Welcome!
Final Burn Neo => FBN Development => Topic started by: JacKc on October 02, 2011, 09:52:09 AM
-
FB Alpha 0.2.97.43 Bugs Reports
This thread was created to organize bugs reports related to FB Alpha v0.2.97.43
Please only post Bug reports in this thread and/or in the FB Alpha Bug Reporting Module (http://www.fbalpha.com/bugs/), for other topics please go to the main release thread here:
http://neo-source.com/index.php?topic=3227.0
Don't think twice about reporting a bug, feel free to post every detail, but please make sure the report is clear and specific, if the bug is a game crash, a savestate would help to recreate it and see whats wrong. Details of your system are also important, you can save the System Information that FBA generates as a text file and attach it with your report. If you have any images / screenshot of the bugs they are very useful as well.
Remember to follow the Forum rules always when posting anything here. :)
Thanks!
If possible, also please attach a save state and screenshots!
-
When i deactivate the autoframe skip i got the "guru meditation"
-
I received a bug with some kaillera clients earlier, both official and unofficial.
And FBA crashes with guru meditation when i do change the connection settings and try to refresh the server list! That happens only when the firewall temporary allows the connection.
Tested different kailleraclient.dll files with temporary allowed firewall policy.
The dlls marked as WORKS still crashes randomly if the connection is not allowed in the firewall exceptions.
Only okai November 02 2007 release works even if you temporary allow the connection to the server and don't crashes after refreshing the list.
official kaillera sdk 0.9 works
okai August 03 2008 / CRC 238165AD / 184 320 kb / crashes
okai November 02 2007 / CRC B59070C3 / 122 880kb / works
FBEvo_3.04 kailleraclinet / CRC 495EC4CD / 32 256 kb / works
(difference from original, only black buttons are fixed)
okai n02 (fix) v0.5r0 (0.9 compatible) - Aug 15 2011 / CRC 2E97B33A / 112 640 kb / works
okai n02 + Ownaclient + fix (0.9 compatible) - Aug 15 2011 / CRC 308C688B / 108 544 kb / crashes
Ownaclient Release 18 / CRC 1D85E3F8 / 108 032 kb / crashes
anti3d kailleraclient 1.1 / CRC 26F62A23 / 32 256 kb / crashes
cpka v14 / CRC 2C72EF9B / 32 256 kb / crashes
I've continue testing different kailleraclient.dll files and crashes stops when kailleraclient.dll is added as exception in the firewall policy.
I'm using comodo firewall.
Adding log from:
okai n02 + Ownaclient + fix (0.9 compatible)
Every client should WORK, ONLY IF you permanently allow the connection to kaillera server, otherwise it crashes when you change the connection settings and try to refresh the list.
Well that never happened in the earlier versions of FB Alpha, that is why I'm posting this, I'm sure that the problem is not in FBA itself still I think that the info can be useful now.
-
hello,
great to see a new release.
however there is a problem with savestates:
- guru meditation crashlog when loading savestate: http://pastebin.com/ujZ2ZrUT
then i started a fresh FBA install with just the new build files and some roms to test it out:
- when the save is actually loading there are visual anomalies like this: http://i.imgur.com/TzQZ1.jpg and the emulation runs at like 30% of the normal speed
-
hello guys, I use google translation
there's a big problem with Pulstar on neogeo
1) final burn alpha do not recognize the rom .
2) We can still play , but there is no sound .
Please fix the problem, I'm a fan of final burn alpha (I hate mame because it is too heavy now)
thanks guys
-
Hi WASABI,
There is no sound problem with pulstar...Just be sure to got the latest Pulstar ROM because in MAME 0.143u4, the V1 has been updated.
{ "089-p1.p1", 0x100000, 0x5e5847a2, 1 | BRF_ESS | BRF_PRG }, // 0 68K code
{ "089-p2.sp2", 0x200000, 0x028b774c, 1 | BRF_ESS | BRF_PRG }, // 1
{ "089-s1.s1", 0x020000, 0xc79fc2c8, 2 | BRF_GRA }, // 2 Text layer tiles
{ "089-c1.c1", 0x400000, 0xf4e97332, 3 | BRF_GRA }, // 3 Sprite data
{ "089-c2.c2", 0x400000, 0x836d14da, 3 | BRF_GRA }, // 4
{ "089-c3.c3", 0x400000, 0x913611c4, 3 | BRF_GRA }, // 5
{ "089-c4.c4", 0x400000, 0x44cef0e3, 3 | BRF_GRA }, // 6
{ "089-c5.c5", 0x400000, 0x89baa1d7, 3 | BRF_GRA }, // 7
{ "089-c6.c6", 0x400000, 0xb2594d56, 3 | BRF_GRA }, // 8
{ "089-c7.c7", 0x200000, 0x6a5618ca, 3 | BRF_GRA }, // 9
{ "089-c8.c8", 0x200000, 0xa223572d, 3 | BRF_GRA }, // 10
{ "089-m1.m1", 0x020000, 0xff3df7c7, 4 | BRF_ESS | BRF_PRG }, // 11 Z80 code
{ "089-v1.v1", 0x400000, 0x6f726ecb, 5 | BRF_SND }, // 12 Sound data
{ "089-v2.v2", 0x400000, 0x9d2db551, 5 | BRF_SND }, // 13
-
ok, I made the right rom but there is always a problem .
at the end of the first boss , in the kinematics , the sound suddenly cut . (not in mame)
thanks .
-
When i deactivate the autoframe skip i got the "guru meditation"
I believe I have fixed this. Can you confirm that the error only happens with the DirectSound3 audio plugin and not with the XAudio2 plugin?
-
About CPS3 loading roms, I have all roms here matching with last mame romsets but when loading any set it give me like I don´t have any and I guess if the FBA isn´t recognizing those because of roms name.
Ex:
redeartha.zip was not found.
redearth.zip was found (G:\emul\cps3\roms\redearth.zip).
• essential program ROM 10 was not found.
• graphics ROM 30 was not found.
• graphics ROM 31 was not found.
• graphics ROM 40 was not found.
• graphics ROM 41 was not found.
• graphics ROM 50 was not found.
Seeing inside of zip file redearth.zip no roms labeled like FBA insists to search are on it, same for other cps3 sets.
-
MAME loads from CHDs for CPS-3. FBA uses the roms from the CHDs, so having just the MAME sets isn't enough for CPS-3.
-
Small bug:
On the hellfire title screen the background is dark red where it should be black.
-
Small bug:
On the hellfire title screen the background is dark red where it should be black.
Nope.. I have all roms and chds here on the same folder and doens´t works....
-
Ultimate Mortal Kombat 3 game is sizzling sounds
-
Small bug:
On the hellfire title screen the background is dark red where it should be black.
Thanks - fixed that one.
-
Regarding the pacman and mspacman drivers; I notice that these games are classed by the emulator as horizontally orientated. They display in their correct 3:4 ratio but to be accurate should they not be outputting a 4:3 rotated screen like other vertical games?
-
Notice some Graphical bugs for Guardians ◦ 電神魔塊 II [grdianszip] on screens before starting stages in FBA (not in MAME) [see attached screenshots]
-
Regarding the pacman and mspacman drivers; I notice that these games are classed by the emulator as horizontally orientated. They display in their correct 3:4 ratio but to be accurate should they not be outputting a 4:3 rotated screen like other vertical games?
Yeah, that was a mistake on my part. Pac-man was one of my first drivers. :) I'll have to see what I can do about modifying the drawing routines and adding in BDF_ORIENTATION_VERTICAL
-
A few other small bugs I've noticed that have been around since before this latest round of updates:
- Shadows in tmnt are not transparent as they should be.
- Some sfx are broken in 1942 (such the loop-the-loop sound).
- Music plays slowly in several games such as raiden and toki.
- Missing music in aliens.
-
http://www.youtube.com/v/Yg9Xn8YbK_E?version=3
-
FBA v0.2.97.12 and v0.2.97.13 will crash after click game->load game...
-
FBA v0.2.97.12 and v0.2.97.13 will crash after click game->load game...
Try deleting config/fba.ini.
-
Try deleting config/fba.ini.
I have deleted config/fba.ini before I reported bug, but same crash.
FBA v0.2.97.11 is run well in my OS.
Is it caused by character? My OS is chinese system.
-
Do you know which characters cause the issue? I would assume they have come from the Megadrive driver.
-
Do you know which characters cause the issue? I would assume they have come from the Megadrive driver.
They have come from the Megadrive driver, but I dont know which characters cause the issue. I can open Game Selection dialog after I modify "nLoadMenuShowX 266240" in fba.ini.
-
The characters which can cause the issue could be the ones from the Megadrive driver from the following sets (Parents and Clones) :
- Alien³
- Astérix and the Great Rescue
- Férias Frustradas do Pica-Pau (Bra)
- Pelé! (Euro, USA)
- Pelé's World Tournament Soccer (Euro, USA)
- Show do Milhão (Bra)
- Show do Milhão Volume 2 (Bra)
-
Do you know which characters cause the issue? I would assume they have come from the Megadrive driver.
Megadrive driver BUG
[hide]
// Ast閞ix and the Great Rescue (Euro)
static struct BurnRomInfo md_asterixRomDesc[] = {
{ "mpr-15961+mpr-15962.bin", 0x200000, 0x4735fee6, BRF_PRG | SEGA_MD_ROM_LOAD16_WORD_SWAP | SEGA_MD_ROM_OFFS_000000 },
};
STD_ROM_PICK(md_asterix)
STD_ROM_FN(md_asterix)
struct BurnDriver BurnDrvmd_asterix = {
"md_asterix", NULL, NULL, NULL, "1993",
"Ast閞ix and the Great Rescue (Euro)\0", NULL, "Sega", "Sega Megadrive",
NULL, NULL, NULL, NULL,
BDF_GAME_WORKING | BDF_16BIT_ONLY, 2, HARDWARE_SEGA_MEGADRIVE, GBF_MISC, 0,
NULL, MegadriveGetZipName, md_asterixRomInfo, md_asterixRomName, NULL, NULL, MegadriveInputInfo, MegadriveDIPInfo,
MegadriveInit, MegadriveExit, MegadriveFrame, NULL, MegadriveScan,
&bMegadriveRecalcPalette, 0x100, 320, 224, 4, 3
};
// Ast閞ix and the Great Rescue (USA)
static struct BurnRomInfo md_asterixuRomDesc[] = {
{ "asterix and the great rescue (usa).bin", 0x200000, 0x7f112cd8, BRF_PRG | SEGA_MD_ROM_LOAD16_WORD_SWAP | SEGA_MD_ROM_OFFS_000000 },
};
STD_ROM_PICK(md_asterixu)
STD_ROM_FN(md_asterixu)
struct BurnDriver BurnDrvmd_asterixu = {
"md_asterixu", "md_asterix", NULL, NULL, "1994",
"Ast閞ix and the Great Rescue (USA)\0", NULL, "Sega", "Sega Megadrive",
NULL, NULL, NULL, NULL,
BDF_GAME_WORKING | BDF_16BIT_ONLY | BDF_CLONE, 2, HARDWARE_SEGA_MEGADRIVE, GBF_MISC, 0,
NULL, MegadriveGetZipName, md_asterixuRomInfo, md_asterixuRomName, NULL, NULL, MegadriveInputInfo, MegadriveDIPInfo,
MegadriveInit, MegadriveExit, MegadriveFrame, NULL, MegadriveScan,
&bMegadriveRecalcPalette, 0x100, 320, 224, 4, 3
};
// Ast閞ix and the Power of the Gods (Euro)
static struct BurnRomInfo md_asterpgRomDesc[] = {
{ "mpr-17719.bin", 0x200000, 0x4ff1d83f, BRF_PRG | SEGA_MD_ROM_LOAD16_WORD_SWAP | SEGA_MD_ROM_OFFS_000000 },
};
STD_ROM_PICK(md_asterpg)
STD_ROM_FN(md_asterpg)
struct BurnDriver BurnDrvmd_asterpg = {
"md_asterpg", NULL, NULL, NULL, "1995",
"Ast閞ix and the Power of the Gods (Euro)\0", NULL, "Sega", "Sega Megadrive",
NULL, NULL, NULL, NULL,
BDF_GAME_WORKING | BDF_16BIT_ONLY, 2, HARDWARE_SEGA_MEGADRIVE, GBF_MISC, 0,
NULL, MegadriveGetZipName, md_asterpgRomInfo, md_asterpgRomName, NULL, NULL, MegadriveInputInfo, MegadriveDIPInfo,
MegadriveInit, MegadriveExit, MegadriveFrame, NULL, MegadriveScan,
&bMegadriveRecalcPalette, 0x100, 320, 224, 4, 3
};
// Ast閞ix and the Power of the Gods (Euro, Prototype)
static struct BurnRomInfo md_asterpgpRomDesc[] = {
{ "asterix and the power of the gods (euro) (beta).bin", 0x200000, 0x45c8b5b7, BRF_PRG | SEGA_MD_ROM_LOAD16_WORD_SWAP | SEGA_MD_ROM_OFFS_000000 },
};
STD_ROM_PICK(md_asterpgp)
STD_ROM_FN(md_asterpgp)
struct BurnDriver BurnDrvmd_asterpgp = {
"md_asterpgp", "md_asterpg", NULL, NULL, "1995",
"Ast閞ix and the Power of the Gods (Euro, Prototype)\0", NULL, "Sega", "Sega Megadrive",
NULL, NULL, NULL, NULL,
BDF_GAME_WORKING | BDF_16BIT_ONLY | BDF_CLONE, 2, HARDWARE_SEGA_MEGADRIVE, GBF_MISC, 0,
NULL, MegadriveGetZipName, md_asterpgpRomInfo, md_asterpgpRomName, NULL, NULL, MegadriveInputInfo, MegadriveDIPInfo,
MegadriveInit, MegadriveExit, MegadriveFrame, NULL, MegadriveScan,
&bMegadriveRecalcPalette, 0x100, 320, 224, 4, 3
};
// F閞ias Frustradas do Pica-Pau (Bra)
static struct BurnRomInfo md_feriasRomDesc[] = {
{ "ferias frustradas do pica-pau (bra).bin", 0x100000, 0x7b2e416d, BRF_PRG | SEGA_MD_ROM_LOAD16_WORD_SWAP | SEGA_MD_ROM_OFFS_000000 },
};
STD_ROM_PICK(md_ferias)
STD_ROM_FN(md_ferias)
struct BurnDriver BurnDrvmd_ferias = {
"md_ferias", NULL, NULL, NULL, "1995",
"F閞ias Frustradas do Pica-Pau (Bra)\0", NULL, "Tec Toy", "Sega Megadrive",
NULL, NULL, NULL, NULL,
BDF_GAME_WORKING | BDF_16BIT_ONLY, 2, HARDWARE_SEGA_MEGADRIVE, GBF_MISC, 0,
NULL, MegadriveGetZipName, md_feriasRomInfo, md_feriasRomName, NULL, NULL, MegadriveInputInfo, MegadriveDIPInfo,
MegadriveInit, MegadriveExit, MegadriveFrame, NULL, MegadriveScan,
&bMegadriveRecalcPalette, 0x100, 320, 224, 4, 3
};
[/hide]
閞---->er
-
I posted a test version on my site.
http://www.barryharris.me.uk/article_view.php?id=164 (http://www.barryharris.me.uk/article_view.php?id=164)
Please can somebody with a Chinese system let me know if this resolves the issue?
Thanks.
-
I posted a test version on my site.
http://www.barryharris.me.uk/article_view.php?id=164 (http://www.barryharris.me.uk/article_view.php?id=164)
Please can somebody with a Chinese system let me know if this resolves the issue?
Thanks.
Cool! It resolves issue which open game selection dialog. Fba is run well now.
Thanks.
-
Thanks for confirming. The ASCII descriptions now lose those characters, but I replaced them in the UniCode descriptions.
-
Strange characters
Astérix and the Great Rescue (Euro) ◦ 獁整楲⁸湡桴片慥⁴敒捳敵⠠畅潲)
md_asterix
Astérix and the Power of the Gods (Euro, Prototype) ◦ 獁整楲⁸湡桴潐敷景琠敨䜠摯䔨牵Ɐ倠潲潴祴数)
md_asterpgp
Férias Frustradas do Pica-Pau (Bra) ◦ 敆楲獡䘠畲瑳慲慤潤倠捩ⵡ慐⁵䈨慲)
md_ferias
Pelé! (Euro, USA) ◦ 敐敬‡䔨牵Ɐ唠䅓) ◦ 摭灟汥e
md_pele
Show do Milhão Volume 2 (Bra) ◦ 桓睯搠楍桬潡嘠汯浵′䈨慲) ◦ 摭獟浤汩慨2Show do Milhão (Bra)
md_sdmilha2
-
In a similar way to how we (Westerners), need to install fonts to show Chinese characters, I imagine you would need to install a font to show some of the more obscure Western characters.
If you set the text to ASCII in the dialog it should display ok, albeit not strictly correctly - but the best you can get with ASCII.
-
@ KOF2112 : the é is a French character and the ã a Spanish one...So you need to install Occidental Western Fonts.
-
Thanks JacK.
-
I think kof2112 refer to this one, garbled
-
Ah, they were a mistake on my part - I didn't escape the \0 in my generator - resulting a NULL character ending up in there - I've fixed that already. :)
-
some other bugs:
d_Megadrive drive:
Turma da M鬾ica na Terra dos Monstros (Bra) -> Turma da Monica na Terra dos Monstros (Bra)
kof98ae:
Select player screen and part player move screen is broken.
fba will crash after enter the following command line:
fba.exe neogeo -w
-
The correct on that name is: Mônica, so its "Turma da Mônica na Terra dos Monstros (Bra)"
Att;
some other bugs:
d_Megadrive drive:
Turma da M鬾ica na Terra dos Monstros (Bra) -> Turma da Monica na Terra dos Monstros (Bra)
kof98ae:
Select player screen and part player move screen is broken.
fba will crash after enter the following command line:
fba.exe neogeo -w
-
@ KOF2112 : the é is a French character and the ã a Spanish one...So you need to install Occidental Western Fonts.
Just to be clear:
· é is used in Spanish also (á, é, í, ó, ú)
· ã is Portuguese, not Spanish ;)
-
Neogeo command line bug - fixed. :)
kof98ae - not sure what the issue is yet.
Megadrive game description - it works fine here - are you missing the font(s) Jack referred to?
-
some graphical glitches now for kof98ae not in previous FBA version...Could it be related to latest changes for neogeo protos, not supported if more than 8 C'roms in the driver or simply because of new MULTISLOT support?
old driver :
STDROMPICKEXT(kof98ae, kof98ae, neogeo)
STD_ROM_FN(kof98ae)
struct BurnDriver BurnDrvkof98ae = {
"kof98ae", "kof98", "neogeo", NULL, "2007",
"The King of Fighters '98 (Anniversary Edition, EGHT hack)\0", NULL, "hack", "Neo Geo",
NULL, NULL, NULL, NULL,
BDF_GAME_WORKING | BDF_CLONE | BDF_HACK, 2, HARDWARE_SNK_NEOGEO | HARDWARE_SNK_MVSCARTRIDGE, GBF_VSFIGHT, FBF_KOF,
NULL, kof98aeRomInfo, kof98aeRomName, NULL, NULL, neogeoInputInfo, neogeoDIPInfo,
NeoInit, NeoExit, NeoFrame, NeoRender, NeoScan, JBF_GAME_WORKING, NULL, NULL, NULL, &NeoRecalcPalette, 0x1000,
304, 224, 4, 3
};
New :
STDROMPICKEXT(kof98ae, kof98ae, neogeo)
STD_ROM_FN(kof98ae)
struct BurnDriver BurnDrvkof98ae = {
"kof98ae", "kof98", "neogeo", NULL, "2007",
"The King of Fighters '98 (Anniversary Edition, EGHT hack)\0", NULL, "hack", "Neo Geo MVS",
NULL, NULL, NULL, NULL,
BDF_GAME_WORKING | BDF_CLONE | BDF_HACK, 2, HARDWARE_PREFIX_CARTRIDGE | HARDWARE_SNK_NEOGEO, GBF_VSFIGHT, FBF_KOF,
NULL, NULL, kof98aeRomInfo, kof98aeRomName, NULL, NULL, neogeoInputInfo, neogeoDIPInfo,
NeoInit, NeoExit, NeoFrame, NeoRender, NeoScan, &NeoRecalcPalette,
0x1000, 304, 224, 4, 3
};
-
I wanted to tell you I found something that could be a bug (or not, depending of your knowledge about this) on Samurai Shodown 4 (Neo-Geo CD).
I tested with a CUE image, the whole folder is about 109MB.
How to reproduce it:
1. Load CUE image
2. Go to "Game Mode" > "Practice"
3. Select characters and prepare to play
4. Since Samurai Shodown IV uses [START] as a command button (to make sucide/throw weapon), PAUSE is made with [SELECT] button. So, press [SELECT]
5. Game is paused and the word "Paused!" appears on the center of the screen.
6. Press [SELECT] to continue training.
What I see:
1. Characters keep freezed, but Stage animations work as usual.
2. You can't PAUSE game again, no mention that you can't escape to anywhere (i.e. to go to Game's Main menu).
Notes:
1. I wanted to add that this "freezing" problem doesn't happen if I choose "Track 01" instead of the CUE image.
2. There's also a graphical error I saw while training... there's a commandlist that falls from the top/left of the window as characters (arrows, ABCD buttons, etc).
The problem is sometimes there's a long vertical line (with the width of a character/button) that falls from the Top/Center of the window and gets fixed on the center of the screen.
I haven't a way to reproduce it (it appears at random). If I see that again, I'll take a capture.
Thanks for all!
-
kof98ae is now working correctly.
-
Picture errors
Thundercade / Twin Formation
Tokusyu Butai U.A.G. (Japan)
-
kof98ae is now working correctly.
Thanks a lot Treble Winner for the fix !!!
-
Thanks Jack. There was a sprite size cap of 64MB, once removed, problem solved. Your post pointed me there. :)
-
Ultimate Mortal Kombat 3 game is still sizzling sounds :rolleyes:
-
It's me or IPS support isn't working for FBA's 0.2.97.13 release?
I've tried to follow these steps:
1. File > Load Game
2. Select (1 click) "Samurai Showdown IV" SET 1 (of course I've IPS patches for this game and they are located in the same place as usual: support/IPS).
- the following 2 can be called as 3rd. or 4th. step in any order -
A. Click "IPS Manager" > Check IPS to use and press OK to close (this confirms IPS patches are located in the correct folder).
B. Check "Apply Patches".
5. Click "Play"
6. Game is loaded, but IPS patches weren't applied.
I hope this isn't a bug... or if it is, then I hope that you could fix it in a easy way. :)
Thanks!
-
It's me or IPS support isn't working for FBA's 0.2.97.13 release?
I've tried to follow these steps:
1. File > Load Game
2. Select (1 click) "Samurai Showdown IV" SET 1 (of course I've IPS patches for this game and they are located in the same place as usual: support/IPS).
- the following 2 can be called as 3rd. or 4th. step in any order -
A. Click "IPS Manager" > Check IPS to use and press OK to close (this confirms IPS patches are located in the correct folder).
B. Check "Apply Patches".
5. Click "Play"
6. Game is loaded, but IPS patches weren't applied.
I hope this isn't a bug... or if it is, then I hope that you could fix it in a easy way. :)
Thanks!
The patches I've tried are working ok. Which specific ones are giving you issues? Could be down to all the recent rom changes in the Neo Geo driver?
-
I suppose that's because of those game changes, as you said.
I tried to load 0 DAY-S & Eddids "Boss Fixes Ver 1.??" made for Samurai Shodown IV SET 1.
The thing is that SET loads OK (no problems), it just doesn't load patches (but it did before .13)
Is there something I could do in order to make them work again?
Thank you!
-
Looking at the patch it has the current filenames and crcs in.
Is there any easy way I can tell if I have the patch applied or not?
I usually test with the kog patch for kof97 - I use that one because it's easy to tell if it worked or not. :)
-
[EDIT]:
Well... I've updated my IPS pack to 4.3 and that problem was solved.
Seems to be that I've to update all my Neo-Geo games on P2 rom.
Thank you so much for your help! ;)
-
Hi.
I *KNOW* this is a bug report for WIP drivers, but I haven't found this issue reported anywhere else- thought it might be of help/interest.
Fba 0.2.97.13:
Cave PGM sound seems to be functioning incorrectly for the following games:
KETSUI
DDP DOJ
DDP DOJBLK
Ketsui sound (music and effects) starts skipping at around the 20sec mark on level 1-1 and doesn't re-sync afterwards.
In DDP DOJ (both variants) the voice channel in the level 1-1 music is missing, and the music skips, doesn't play properly or starts late on boss fights and inbetween-levels score screens.
ESPGALUDA sounds fine- no such issues.
If this is any help, on starting KETSUI I hear a short crackling sound just before the disclaimer screen appears, which I also hear with ESPGALUDA.
These issues don't appear to be present in current MAMEUI32 or SHMUPMAME (3.0b) builds.
Appologies if this isn't news to anyone, but as I said I haven't read about it anywhere else.
I've tried different pci slots for my sound card, and have updated the sound card drivers.
System info:
FB Alpha v0.2.97.13 system information (Thu Nov 24 11:30:52 2011
).
----------------------------------------------------------------------
System information:
OS: Microsoft Windows XP Personal Service Pack 3 (build 2600)
CPU: AuthenticAMD, Athlon 64 / Athlon 64 Mobile
2202 MHz, MMX, SSE, SSE2 (1 system processor)
Physical RAM: 2095856 KB (2046 MB) total, 1427740 KB (1394 MB) avail
Total RAM: 3511456 KB (3429 MB) total, 3008380 KB (2937 MB) avail
FB Alpha: 11620 KB in use (12040 KB peak, 13800 KB virtual)
Installed displays and display adapters:
LG LG 782LE on NVIDIA GeForce3 Ti 500 (primary)
Display size ~34x27cm, Gamma 2.50
Preferred mode 1280x1024, 60.020 Hz (338x270mm, 108.000 MHz)
Max. bandwidth 140 MHz, H sync 30-70 KHz, V sync 56-85 Hz
----------------------------------------------------------------------
FB Alpha information:
Built on Oct 28 2011, 20:30:31, using Visual C++ 10.0.
Optimised for i686 CPUs.
Using Unicode for all text.
Debug functionality absent.
MMX optimisations enabled.
A68K emulation core enabled for MC68000 emulation.
Musashi emulation core enabled for MC68010/MC68EC020 emulation.
Not emulating any game.
Video settings:
Selected module: DirectDraw7 video output
Interface settings: Running in windowed mode, $ix923, 652bpp
Source image 304x224, 32bpp
Module settings: Using Blt() to transfer the image
Using video memory for the final blit
Audio settings:
Selected module: DirectSound3 audio output
Interface settings: Playback at 22050Hz, 100% volume
Module settings: Audio is delayed by approx. 83ms
Input settings:
Selected module: DirectInput8 input
Interface settings: keyboard 0 System keyboard: Keyboard
mouse 0 System mouse: Mouse
Profiling settings:
Interface settings: Profiling module not initialised
----------------------------------------------------------------------
Detailed process information about fba.exe:
fba.exe (base address 0x00400000, size 18204 KB)
Modules loaded by fba.exe:
ntdll.dll (base address 0x7C900000, size 700 KB)
kernel32.dll (base address 0x7C800000, size 984 KB)
USER32.dll (base address 0x7E410000, size 580 KB)
GDI32.dll (base address 0x77F10000, size 292 KB)
COMDLG32.dll (base address 0x763B0000, size 292 KB)
ADVAPI32.dll (base address 0x77DD0000, size 620 KB)
RPCRT4.dll (base address 0x77E70000, size 584 KB)
Secur32.dll (base address 0x77FE0000, size 68 KB)
COMCTL32.dll (base address 0x773D0000, size 1036 KB)
msvcrt.dll (base address 0x77C10000, size 352 KB)
SHLWAPI.dll (base address 0x77F60000, size 472 KB)
SHELL32.dll (base address 0x7C9C0000, size 8284 KB)
WINMM.dll (base address 0x76B40000, size 180 KB)
ole32.dll (base address 0x774E0000, size 1268 KB)
d3d9.dll (base address 0x4FDD0000, size 1688 KB)
d3d8thk.dll (base address 0x6D990000, size 24 KB)
VERSION.dll (base address 0x77C00000, size 32 KB)
d3dx9_43.dll (base address 0x10000000, size 2044 KB)
DDRAW.dll (base address 0x73760000, size 300 KB)
DCIMAN32.dll (base address 0x73BC0000, size 24 KB)
DSOUND.dll (base address 0x73F10000, size 368 KB)
DINPUT8.dll (base address 0x6CE10000, size 224 KB)
IMM32.DLL (base address 0x76390000, size 116 KB)
uxtheme.dll (base address 0x5AD70000, size 224 KB)
MSCTF.dll (base address 0x74720000, size 304 KB)
apphelp.dll (base address 0x77B40000, size 136 KB)
msctfime.ime (base address 0x755C0000, size 184 KB)
HID.DLL (base address 0x688F0000, size 36 KB)
SETUPAPI.DLL (base address 0x77920000, size 972 KB)
WINTRUST.dll (base address 0x76C30000, size 184 KB)
CRYPT32.dll (base address 0x77A80000, size 596 KB)
MSASN1.dll (base address 0x77B20000, size 72 KB)
IMAGEHLP.dll (base address 0x76C90000, size 160 KB)
OLEAUT32.DLL (base address 0x77120000, size 556 KB)
wdmaud.drv (base address 0x72D20000, size 36 KB)
msacm32.drv (base address 0x72D10000, size 32 KB)
MSACM32.dll (base address 0x77BE0000, size 84 KB)
midimap.dll (base address 0x77BD0000, size 28 KB)
KsUser.dll (base address 0x5EF80000, size 16 KB)
CLBCATQ.DLL (base address 0x76FD0000, size 508 KB)
COMRes.dll (base address 0x77050000, size 788 KB)
ntshrui.dll (base address 0x76990000, size 148 KB)
ATL.DLL (base address 0x76B20000, size 68 KB)
NETAPI32.dll (base address 0x5B860000, size 340 KB)
USERENV.dll (base address 0x769C0000, size 720 KB)
----------------------------------------------------------------------
-
Hi.
I *KNOW* this is a bug report for WIP drivers, but I haven't found this issue reported anywhere else- thought it might be of help/interest.
Fba 0.2.97.13:
Cave PGM sound seems to be functioning incorrectly for the following games:
KETSUI
DDP DOJ
DDP DOJBLK
Ketsui sound (music and effects) starts skipping at around the 20sec mark on level 1-1 and doesn't re-sync afterwards.
In DDP DOJ (both variants) the voice channel in the level 1-1 music is missing, and the music skips, doesn't play properly or starts late on boss fights and inbetween-levels score screens.
ESPGALUDA sounds fine- no such issues.
If this is any help, on starting KETSUI I hear a short crackling sound just before the disclaimer screen appears, which I also hear with ESPGALUDA.
These issues don't appear to be present in current MAMEUI32 or SHMUPMAME (3.0b) builds.
Appologies if this isn't news to anyone, but as I said I haven't read about it anywhere else.
I've tried different pci slots for my sound card, and have updated the sound card drivers.
System info:
FB Alpha v0.2.97.13 system information (Thu Nov 24 11:30:52 2011
).
----------------------------------------------------------------------
System information:
OS: Microsoft Windows XP Personal Service Pack 3 (build 2600)
CPU: AuthenticAMD, Athlon 64 / Athlon 64 Mobile
2202 MHz, MMX, SSE, SSE2 (1 system processor)
Physical RAM: 2095856 KB (2046 MB) total, 1427740 KB (1394 MB) avail
Total RAM: 3511456 KB (3429 MB) total, 3008380 KB (2937 MB) avail
FB Alpha: 11620 KB in use (12040 KB peak, 13800 KB virtual)
Installed displays and display adapters:
LG LG 782LE on NVIDIA GeForce3 Ti 500 (primary)
Display size ~34x27cm, Gamma 2.50
Preferred mode 1280x1024, 60.020 Hz (338x270mm, 108.000 MHz)
Max. bandwidth 140 MHz, H sync 30-70 KHz, V sync 56-85 Hz
----------------------------------------------------------------------
FB Alpha information:
Built on Oct 28 2011, 20:30:31, using Visual C++ 10.0.
Optimised for i686 CPUs.
Using Unicode for all text.
Debug functionality absent.
MMX optimisations enabled.
A68K emulation core enabled for MC68000 emulation.
Musashi emulation core enabled for MC68010/MC68EC020 emulation.
Not emulating any game.
Video settings:
Selected module: DirectDraw7 video output
Interface settings: Running in windowed mode, $ix923, 652bpp
Source image 304x224, 32bpp
Module settings: Using Blt() to transfer the image
Using video memory for the final blit
Audio settings:
Selected module: DirectSound3 audio output
Interface settings: Playback at 22050Hz, 100% volume
Module settings: Audio is delayed by approx. 83ms
Input settings:
Selected module: DirectInput8 input
Interface settings: keyboard 0 System keyboard: Keyboard
mouse 0 System mouse: Mouse
Profiling settings:
Interface settings: Profiling module not initialised
----------------------------------------------------------------------
Detailed process information about fba.exe:
fba.exe (base address 0x00400000, size 18204 KB)
Modules loaded by fba.exe:
ntdll.dll (base address 0x7C900000, size 700 KB)
kernel32.dll (base address 0x7C800000, size 984 KB)
USER32.dll (base address 0x7E410000, size 580 KB)
GDI32.dll (base address 0x77F10000, size 292 KB)
COMDLG32.dll (base address 0x763B0000, size 292 KB)
ADVAPI32.dll (base address 0x77DD0000, size 620 KB)
RPCRT4.dll (base address 0x77E70000, size 584 KB)
Secur32.dll (base address 0x77FE0000, size 68 KB)
COMCTL32.dll (base address 0x773D0000, size 1036 KB)
msvcrt.dll (base address 0x77C10000, size 352 KB)
SHLWAPI.dll (base address 0x77F60000, size 472 KB)
SHELL32.dll (base address 0x7C9C0000, size 8284 KB)
WINMM.dll (base address 0x76B40000, size 180 KB)
ole32.dll (base address 0x774E0000, size 1268 KB)
d3d9.dll (base address 0x4FDD0000, size 1688 KB)
d3d8thk.dll (base address 0x6D990000, size 24 KB)
VERSION.dll (base address 0x77C00000, size 32 KB)
d3dx9_43.dll (base address 0x10000000, size 2044 KB)
DDRAW.dll (base address 0x73760000, size 300 KB)
DCIMAN32.dll (base address 0x73BC0000, size 24 KB)
DSOUND.dll (base address 0x73F10000, size 368 KB)
DINPUT8.dll (base address 0x6CE10000, size 224 KB)
IMM32.DLL (base address 0x76390000, size 116 KB)
uxtheme.dll (base address 0x5AD70000, size 224 KB)
MSCTF.dll (base address 0x74720000, size 304 KB)
apphelp.dll (base address 0x77B40000, size 136 KB)
msctfime.ime (base address 0x755C0000, size 184 KB)
HID.DLL (base address 0x688F0000, size 36 KB)
SETUPAPI.DLL (base address 0x77920000, size 972 KB)
WINTRUST.dll (base address 0x76C30000, size 184 KB)
CRYPT32.dll (base address 0x77A80000, size 596 KB)
MSASN1.dll (base address 0x77B20000, size 72 KB)
IMAGEHLP.dll (base address 0x76C90000, size 160 KB)
OLEAUT32.DLL (base address 0x77120000, size 556 KB)
wdmaud.drv (base address 0x72D20000, size 36 KB)
msacm32.drv (base address 0x72D10000, size 32 KB)
MSACM32.dll (base address 0x77BE0000, size 84 KB)
midimap.dll (base address 0x77BD0000, size 28 KB)
KsUser.dll (base address 0x5EF80000, size 16 KB)
CLBCATQ.DLL (base address 0x76FD0000, size 508 KB)
COMRes.dll (base address 0x77050000, size 788 KB)
ntshrui.dll (base address 0x76990000, size 148 KB)
ATL.DLL (base address 0x76B20000, size 68 KB)
NETAPI32.dll (base address 0x5B860000, size 340 KB)
USERENV.dll (base address 0x769C0000, size 720 KB)
----------------------------------------------------------------------
The sound core we're using for PGM is old and crap. Probably sooner, rather than later, I'll be porting MAME's nice core to FBA.
-
...Oh I wouldn't describe any part of FBA as 'crap'.. But I look forward to future builds :)
Keep up the fantastic work.
-
The kof98ae works bad is flawed of pixels
(http://img836.imageshack.us/img836/2271/kof98ae1201193403.png)
-
This bug has been fixed in the 0.2.97.14 release !
-
0.2.97.14 bug:
XP OS user run Fba, don't display menubar.
Fba will crash after md_kof98 is run.
-
0.2.97.14 bug:
XP OS user run Fba, don't display menubar.
Fba will crash after md_kof98 is run.
What build of FB Alpha are you using, 32-bit or 64-bit?, and what Windows XP version you have, x64 or x86 ?
-
Fba don't display menubar in windows XP (x86 and x64)
http://f.ppxclub.com/forum-redirect-goto-findpost-ptid-141075-pid-1896319.html
-
Fba don't display menubar in windows XP (x86 and x64)
http://f.ppxclub.com/forum-redirect-goto-findpost-ptid-141075-pid-1896319.html
Thanks for the information, I will try to fix the issues as soon as possible.
-
Strange one - MinGW builds work fine on XP SP3, VC builds don't. I'll try and take a look tonight.
-
Hmm, well I thought it might be my compile environment so I took a snapshot of my XP VM, setup a VC compile environment from scratch and compiled a binary. The issue was still there.
Also, using my current VC environment to compile the previous release works fine on XP.
The only changes I can see that might affect affect this are Cap's DX changes. Cap, do you have any relatively easy way of using the latest released code with the old method of linking to the DX libs and see if that sorts it?
-
Well, I did, say might. :)
Didn't take me long to chop and change the source to build with libs linked again - didn't resolve the issue! Giving up for now. :(
-
Sure Barry :), I just finished installing "Windows XP Mode" with VMware Player and was able to reproduce the problem.
I will try reverting the modified code back and I'll post the results asap.
SeeYaa!
:biggrin:
-
I get guru meditation when change to Enhaced blitter
FB Alpha v0.2.97.14 fatal exception report (Fri Dec 02 16:53:37 2011
)
----------------------------------------------------------------------
Exception 0xC0000005 (EXCEPTION_ACCESS_VIOLATION) thrown.
EIP: 0x0114
(attempting to write address 0x00000000)
EAX: 0x00FFFFFF, EBX: 0x00FFFFFF, ECX: 0x001DCF4C, EDX: 0x00000000
ESI: 0x00000000, EDI: 0x00000002, ESP: 0x001DCF20, EBP: 0x001DCFD0
----------------------------------------------------------------------
System information:
OS: Microsoft Windows XP Professional Service Pack 3 (build 2600)
CPU: AuthenticAMD, Athlon 64 / Athlon 64 Mobile
2434 MHz, MMX, SSE, SSE2 (1 system processor)
Physical RAM: 1048575 KB (1023 MB) total, 639460 KB ( 624 MB) avail
Total RAM: 2097151 KB (2047 MB) total, 1048575 KB (1023 MB) avail
FB Alpha: 28328 KB in use (28328 KB peak, 59156 KB virtual)
Installed displays and display adapters:
@monitor.inf,%generic%;(Tipo de monitor estándar) @monitor.inf,%no
No EDID data present for this device
----------------------------------------------------------------------
FB Alpha information:
Built on Dec 1 2011, 19:54:54, using Visual C++ 10.0.
Optimised for X86 CPUs.
Using Unicode for all text.
Debug functionality absent.
MMX optimisations enabled.
A68K emulation core enabled for MC68000 emulation.
Musashi emulation core enabled for MC68010/MC68EC020 emulation.
Not emulating any game.
Video settings:
Interface settings: Video plugin not initialised
Audio settings:
Selected module: DirectSound3 audio output
Interface settings: Playback at 22050Hz, 100% volume
Module settings: Audio is delayed by approx. 83ms
Input settings:
Selected module: DirectInput8 input
Interface settings: keyboard 0 System keyboard: Teclado
mouse 0 System mouse: Mouse
joystick 0 Twin USB Vibration Gamepad
joystick 1 Twin USB Vibration Gamepad
Profiling settings:
Interface settings: Profiling module not initialised
----------------------------------------------------------------------
Detailed process information about fba.exe:
fba.exe (base address 0x00CC0000, size 23784 KB)
Exception occurred in module fba.exe:
fba.exe (base address 0x00CC0000, size 23784 KB)
Modules loaded by fba.exe:
ntdll.dll (base address 0x778E0000, size 1264 KB)
kernel32.dll (base address 0x771C0000, size 848 KB)
KERNELBASE.dll (base address 0x75B10000, size 296 KB)
snxhk.dll (base address 0x64D00000, size 208 KB)
USER32.dll (base address 0x76380000, size 804 KB)
GDI32.dll (base address 0x75D30000, size 312 KB)
LPK.dll (base address 0x77A20000, size 40 KB)
USP10.dll (base address 0x75D90000, size 628 KB)
msvcrt.dll (base address 0x770A0000, size 688 KB)
COMDLG32.dll (base address 0x77480000, size 492 KB)
SHLWAPI.dll (base address 0x75FD0000, size 348 KB)
COMCTL32.dll (base address 0x74990000, size 1656 KB)
SHELL32.dll (base address 0x76450000, size 12580 KB)
WINMM.dll (base address 0x70D00000, size 200 KB)
ole32.dll (base address 0x76220000, size 1392 KB)
RPCRT4.dll (base address 0x775D0000, size 644 KB)
ADVAPI32.dll (base address 0x76180000, size 640 KB)
sechost.dll (base address 0x77A30000, size 100 KB)
apphelp.dll (base address 0x75910000, size 300 KB)
AcLayers.dll (base address 0x70C10000, size 560 KB)
SspiCli.dll (base address 0x75960000, size 104 KB)
OLEAUT32.dll (base address 0x75F40000, size 572 KB)
USERENV.dll (base address 0x75090000, size 92 KB)
profapi.dll (base address 0x759D0000, size 44 KB)
WINSPOOL.DRV (base address 0x70AD0000, size 324 KB)
MPR.dll (base address 0x722D0000, size 72 KB)
AcGenral.DLL (base address 0x697B0000, size 2144 KB)
UxTheme.dll (base address 0x74800000, size 256 KB)
samcli.dll (base address 0x741D0000, size 60 KB)
MSACM32.dll (base address 0x70A60000, size 80 KB)
VERSION.dll (base address 0x74F30000, size 36 KB)
sfc.dll (base address 0x6F9C0000, size 12 KB)
sfc_os.DLL (base address 0x6F990000, size 52 KB)
dwmapi.dll (base address 0x74970000, size 76 KB)
SETUPAPI.dll (base address 0x772A0000, size 1652 KB)
CFGMGR32.dll (base address 0x75AB0000, size 156 KB)
DEVOBJ.dll (base address 0x75B60000, size 72 KB)
urlmon.dll (base address 0x76030000, size 1236 KB)
CRYPT32.dll (base address 0x75C10000, size 1136 KB)
MSASN1.dll (base address 0x75AA0000, size 48 KB)
iertutil.dll (base address 0x776E0000, size 2020 KB)
AcXtrnal.DLL (base address 0x69550000, size 2404 KB)
SHUNIMPL.DLL (base address 0x73AE0000, size 24 KB)
SortServer2003Compat.dll (base address 0x72040000, size 52 KB)
IMM32.DLL (base address 0x771A0000, size 124 KB)
MSCTF.dll (base address 0x77500000, size 816 KB)
dsound.dll (base address 0x74350000, size 456 KB)
POWRPROF.dll (base address 0x74220000, size 148 KB)
dinput8.dll (base address 0x72010000, size 192 KB)
ddraw.dll (base address 0x69460000, size 924 KB)
DCIMAN32.dll (base address 0x72000000, size 24 KB)
d3d9.dll (base address 0x6A650000, size 1804 KB)
d3d8thk.dll (base address 0x71370000, size 24 KB)
D3DX9_43.dll (base address 0x670A0000, size 2044 KB)
HID.DLL (base address 0x74DF0000, size 36 KB)
CRYPTBASE.dll (base address 0x759B0000, size 48 KB)
WINTRUST.dll (base address 0x75AE0000, size 180 KB)
CLBCatQ.DLL (base address 0x77A50000, size 524 KB)
MMDevApi.dll (base address 0x74500000, size 228 KB)
PROPSYS.dll (base address 0x74840000, size 980 KB)
AUDIOSES.DLL (base address 0x70A90000, size 216 KB)
EZFRD32.dll (base address 0x10000000, size 92 KB)
DINPUT.dll (base address 0x6C260000, size 148 KB)
nvd3dum.dll (base address 0x06090000, size 3236 KB)
D3DIM700.DLL (base address 0x69C20000, size 816 KB)
----------------------------------------------------------------------
-
I will re-implement everything related to my DirectX changes later from scratch using the burner interface from 029713, but without the 64-bit changes, just to make sure, then start testing from that point. The problem is very weird indeed.
PS: The 64-bit changes I'm referring are the ones at the burner interface only, just in case.
SeeYaa!
:biggrin:
-
Ok, I just confirmed that the menu problem is located at src/burner/win32/ in FBA source v029714. To confirm it, I replaced the whole directory with the win32 directory from v029713 source, then started porting everything from scratch, so far I've implemented almost everything and the Menu is working fine on Windows XP.
I haven't touched src/burner/win32/menu.cpp yet, and there are a few things left to port in src/burner/win32/scrn.cpp.
I will keep reporting as I progress :)
SeeYaa!
:biggrin:
-
i compiled x64 build (neogeo only) with gcc461.. compiling failed without this lib: -ldxguid (this is for IID_Unknown)
-
i compiled x64 build (neogeo only) with gcc461.. compiling failed without this lib: -ldxguid (this is for IID_Unknown)
Are you using the latest official source or an old one?, I searched the whole source for "IID_Unknown" and it isn't used. Could you please post on what object file you got the failure?
I have compiled FBA v0.2.97.14 source with gcc 3.4.5, gcc 4.5.2, gcc 4.6.1 and I get no problems.
SeeYaa!
:biggrin:
-
heh.. sorry cap.. false alarm.. i compiled again from scratch & all turned out fine.. :)
-
treblewinner
I have post my source
http://f.ppxclub.com/forum-redirect-goto-findpost-ptid-141076-pid-1897032.html
-
I found the Windows XP Menu problem last night, after researching the code, the fix resulted to be very simple :p
In src/burner/win32/burner_win32.h from FBA src v0.2.97.14 change this ...
#define _WIN32_WINDOWS 0x0410
//#define _WIN32_WINNT 0x0400
#define _WIN32_IE 0x0500
#if not defined _WIN32_WINNT
#define _WIN32_WINNT 0x0501
#endif
#if not defined WINVER
#define WINVER 0x0501
#endif
#define STRICT
to this...
#define _WIN32_WINDOWS 0x0410
//#define _WIN32_WINNT 0x0400
#define _WIN32_IE 0x0500
#define _WIN32_WINNT 0x0501
#define WINVER 0x0501
#define STRICT
I submitted that fix along with 2 more small changes related to the Neo Geo CDZ interface.
SeeYaa!
:biggrin:
-
Some screenshots...
(http://captaincps.net/neosource/img/vc_winxp_menu_01.png)
(http://captaincps.net/neosource/img/vc_winxp_menu_02.png)
SeeYaa!
:biggrin:
-
Nice work Cap.
And doomking, thanks for releasing the fix as well.
-
Question for iq_132 regarding the revised pacman driver:
I'm just in the process of updating the Xbox build with the latest FBA code. Everything is working well but a couple of the games on this driver go through their hardware test then just display a screen showing DIP switch info with looping audio in the background. There seems to be no way past this screen.
The problem affects puckman, mspacman and clones; although mschamp and all non-pacman-style games are working fine. Just wondered if there was anything singular about these games in the driver that might help pin down the cause of this problem?
-
Question for iq_132 regarding the revised pacman driver:
I'm just in the process of updating the Xbox build with the latest FBA code. Everything is working well but a couple of the games on this driver go through their hardware test then just display a screen showing DIP switch info with looping audio in the background. There seems to be no way past this screen.
The problem affects puckman, mspacman and clones; although mschamp and all non-pacman-style games are working fine. Just wondered if there was anything singular about these games in the driver that might help pin down the cause of this problem?
There were a few mspacman clones that did that with the version iq submitted. I fixed those before release though, along with some dip switch fixes. The problem with the mspacman clones was that they were using the puckman init, but they use a different memory map.
-
There were a few mspacman clones that did that with the version iq submitted. I fixed those before release though, along with some dip switch fixes. The problem with the mspacman clones was that they were using the puckman init, but they use a different memory map.
Out of interest, what were the DIP switch fixes? The Xbox port's UI has an unusual way of handling DIP switch settings and I've had to manually alter several drivers to get them working correctly, so I suspect this may be where the problem lies. I could be wrong but I don't see any other reason why this problem would exist in the Xbox build and not the original build.
-
I can't see the new Neo-Geo CD submenu on .15 update:
(http://img168.imagevenue.com/loc702/th_314955964_Imagen4_122_702lo.jpg) (http://img168.imagevenue.com/img.php?image=314955964_Imagen4_122_702lo.jpg)
Does anyone know why?
Thanks!
-
From memory, mostly changes in non-Pacman memory map drivers, changing DrvDips[0] and DrvDips[1] to DrvDips[2] and DrvDips[3].
-
Notice a minor GFX Bug for Varth: Operation Thunderstorm (World 920714) and clones (see attached files):
- The background is brown instead of black
Also impossible to take a snap (with BASIC Blitter) for these sets and when exit game with shortcut key CTRL+F4, it makes FBA crashes.
I'm using Windows 7 x64.
-
JacKc - not a bug - the game clears the screen using a palette entry - check MAME.
It crashes here on exit too - but I think I know what caused that. :)
-
I fixed the screenshot issue and the crash on exit issue. :)
For more on the CPS-1 screen clearing see this;
www.barryharris.me.uk/article_view.php?id=100
-
Thanks a lot Treble Winner !!!
-
I'm working on Hungarian translation now and I've noticed that the translation of "Adjust CPU speed..." string doesn't work. I've found that string in two places in the template.flt and I've translated both of them but in the emulator it dislayed in English.
-
my build is resources based on some old code from Jan_Klaassen
-
Doomking - please post the fix.
-
ok
-
ok
Thanks doomking. Issue with string translation is now resolved.
It seems there is an issue with GCC builds and strings when building the templates (which lead me down the wrong path!). It looks like a very long-standing issue and since release builds are done with VC, I'm not overly concerned.
I'll now start identifying items that should be loaded as string resources rather than hard-coded.
-
It seems there is an issue with GCC builds and strings when building the templates (which lead me down the wrong path!).
I have updated resources code from my build fba code cleanup for GCC builds (http://neo-source.com/index.php?topic=1446.0)
-
I've noticed that Parodius no longer works. The ROM check shows the C5 ROM as BAD.
-
I have updated resources code from my build fba code cleanup for GCC builds (http://neo-source.com/index.php?topic=1446.0)
Nice catch. That fixes GCC builds exporting a template.
I've changed all the hard-coded strings I can find to translatable strings, so the next release should be the best yet for localisation.
-
I've just finished my Hungarian translation. I've found some strings displayed in English still:
Path #1 - Path #8
Showing 0 of 5032 sets [0 Unavailable sets]
Error
Error finding 'kailleraclient.dll'.
OK
Filters
Board Type
Family
Genre
Neo Geo CDZ system
Title
Slot 1 - Slot 6
Empty
Select CD image
FB Alpha Error
Couldn't initialise CD emulation.
Other Gamma value...
Other...
Previews
Titles
Cheats
High Scores
Samples
IPS
Neo Geo CDZ Games
Neo Geo CDZ Covers
etc.
Will they be corrected in the next release of FBA? A totally translatable build would be nice. Is it possible to inculde translations with the emulator and starting the first time the emulator runs as same as the OS language?
-
I've just finished my Hungarian translation. I've found some strings displayed in English still:
Path #1 - Path #8
Showing 0 of 5032 sets [0 Unavailable sets]
Error
Error finding 'kailleraclient.dll'.
OK
Filters
Board Type
Family
Genre
Neo Geo CDZ system
Title
Slot 1 - Slot 6
Empty
Select CD image
FB Alpha Error
Couldn't initialise CD emulation.
Other Gamma value...
Other...
Previews
Titles
Cheats
High Scores
Samples
IPS
Neo Geo CDZ Games
Neo Geo CDZ Covers
etc.
Will they be corrected in the next release of FBA? A totally translatable build would be nice. Is it possible to inculde translations with the emulator and starting the first time the emulator runs as same as the OS language?
They should all be corrected. If you find any strings that don't translate in the new FB Alpha (0.2.97.16), please let me know and I will resolve any issues.
-
bug:
nbbatman can't be run.
-
PoINT32 => Point
Turma da M\u00f4nica na Terra dos Monstros (Bra)
-
Thanks doomking.
The Megadrive game name is correct, and comes direct from MESS. This is how it looks with the correct language support installed;
-
Unicode Fix..
Set wofj, wofjh, wofch in src\burn\drivers\capcom\d_cps1.cpp
Change
\u8EE2\u5730\u3092\u55B0\u3089\u3046II
to
\u5929\u5730\u3092\u55B0\u3089\u3046II
-
little error for wofch v0.144u3:
New (v0.144u3):
tk2(ch)_07.9a", 0x200004, 0x80000, CRC(d706568e)
Old (v0.144u2 and fba .16):
tk2(ch)_07.9a", 0x200004, 0x80000, CRC(cc9006c9)
-
Unicode Fix..
Set wofj, wofjh, wofch in src\burn\drivers\capcom\d_cps1.cpp
Change
\u8EE2\u5730\u3092\u55B0\u3089\u3046II
to
\u5929\u5730\u3092\u55B0\u3089\u3046II
Thanks - change made.
-
little error for wofch v0.144u3:
New (v0.144u3):
tk2(ch)_07.9a", 0x200004, 0x80000, CRC(d706568e)
Old (v0.144u2 and fba .16):
tk2(ch)_07.9a", 0x200004, 0x80000, CRC(cc9006c9)
Thanks - changed it over.
-
Unicode Fix..
Set wofj, wofjh, wofch in src\burn\drivers\capcom\d_cps1.cpp
Change
\u8EE2\u5730\u3092\u55B0\u3089\u3046II
to
\u5929\u5730\u3092\u55B0\u3089\u3046II
Set dynwarj, dynwarjr
\u8EE2
to
\u5929
-
Hi all.
I´m still having problems to load all CPS3 games here.
All of my zip files matches with last mame release and I cant load it.
Bad... a pop up insist on stay at my screen upon the load rom failure.
Check screen shot please...
-
FB Alpha sets don't match MAME for CPS-3, we don't support CHDs.
Follow the error dialog and get the needed files - use the datfile to help.
-
update first post to FBA 0.2.97.17 release ;p
Please now report bugs concerning this new release version.
-
FB Alpha sets don't match MAME for CPS-3, we don't support CHDs.
Follow the error dialog and get the needed files - use the datfile to help.
OK and what about the pop up rom load screen?
-
OK and what about the pop up rom load screen?
The screen is only appearing to tell you about your missing files. Get the correct files, and the screen won't appear....
-
FB Alpha sets don't match MAME for CPS-3, we don't support CHDs.
Follow the error dialog and get the needed files - use the datfile to help.
The MAME sets also load the flash dumps (in the format they would be on the chip, rather than the format on the CD) to avoid having to go through the countdown each time. You could easily load the same sets as MAME and fake / patch out the CD check ;-)
-
The MAME sets also load the flash dumps (in the format they would be on the chip, rather than the format on the CD) to avoid having to go through the countdown each time. You could easily load the same sets as MAME and fake / patch out the CD check ;-)
Hmm, I might take a look sometime, once I've finished tidying all the current stuff. We've got sound going in plenty of old drivers that didn't work before.
Our biggest stumbling block now is with games that use the H6280 CPU. iq did the initial port, and I have ported since, but we always seem to hit problems. For example, in robocop, it starts reading from 0x1fe000 - 0x1fffff, which doesn't appear to be mapped in MAME at all. Passing the games ROM, seems to make it work for a bit longer, but eventually, it stops doing anything useful, and doesn't get ingame.
Off the top of your head, would you have any experience with the chip that would point is in the right direction? I'm not asking to take a look at it, just wondering if you have had any experience with this or a similar CPU you could share that might point us at our error?
-
The chip is a bit weird in MAME too (half of me thinks the endianess of it is somehow busted too as you have to reverse half the reads/writes vs. what you'd expect when interfacing with other chips, seems like a 16-bit cpu kludged up as an 8-bit one, but maybe that's just the design.)
Of the top of my head it has some kind of built in paging / memory banking / segment type system which might be confusing you (IIRC basic addressing is only 16-bits but then it can select more with a register) Maybe you have something uninitialized, or are just handling that incorrectly.
Of course if your word read/writes on it are messed up vs. MAME (which due to the weirdness i experienced I wouldn't rule out) then it's possible some of your memory handlers are wrong, causing the CPU core to read the wrong data, and crash.. I'd check that avenue too.
-
The chip is a bit weird in MAME too (half of me thinks the endianess of it is somehow busted too as you have to reverse half the reads/writes vs. what you'd expect when interfacing with other chips, seems like a 16-bit cpu kludged up as an 8-bit one, but maybe that's just the design.)
Of the top of my head it has some kind of built in paging / memory banking / segment type system which might be confusing you (IIRC basic addressing is only 16-bits but then it can select more with a register) Maybe you have something uninitialized, or are just handling that incorrectly.
Of course if your word read/writes on it are messed up vs. MAME (which due to the weirdness i experienced I wouldn't rule out) then it's possible some of your memory handlers are wrong, causing the CPU core to read the wrong data, and crash.. I'd check that avenue too.
Thanks Haze - hopefully I should get some time over Christmas to take a look at it again.
-
I've noticed that Parodius no longer works. The ROM check shows the C5 ROM as BAD.
That's the sound rom. Not sure how it is broken atm. :S
-
That's the sound rom. Not sure how it is broken atm. :S
Was a bug in K053260 caused by me changing "unsigned long" to INT32 instead of UINT32. Works now anyway. :)
-
Hi all, i compile this release, after of compiler, open the fba.exe and have a guru meditation at 0x0000a662
could you please help me. Many thanks, and happy new year.
LOG:
FB Alpha v0.2.97.17 fatal exception report (Fri
----------------------------------------------------------------------
Exception 0xC000001D (EXCEPTION_ILLEGAL_INSTRUCTION) thrown.
EIP: 0x0
EAX: 0x00007020, EBX: 0x00000000, ECX: 0x0022D9D0, EDX: 0x00000001
ESI: 0xFFFFFFFF, EDI: 0xFFFFFFFE, ESP: 0x0022D9C0, EBP: 0x00252356
----------------------------------------------------------------------
System information:
OS: Microsoft Windows XP Professional Service Pack 3 (build 2600)
CPU: AuthenticAMD, Athlon XP "Thorton" / Athon XP "Barton"
2191 MHz, MMX, SSE (1 system processor)
Physical RAM: 2096628 KB (2047 MB) total, 1360144 KB (1328 MB) avail
Total RAM: 4038128 KB (3943 MB) total, 3333428 KB (3255 MB) avail
FB Alpha N3O: 6364 KB in use (6364 KB peak, 16736 KB virtual)
Installed displays and display adapters:
Philips Consumer Electronics Co. Philips 105E (105E1) on NVIDIA Ge
Display size ~28x21cm, Gamma 2,87
Max. bandwidth 70 MHz, H sync 30-54 KHz, V sync 50-120 Hz
----------------------------------------------------------------------
FB Alpha information:
Built on Dec 30 2011, 13:51:54, using GCC 4.6.1.
Optimised for X86 CPUs.
Using Unicode for all text.
Debug functionality absent.
MMX optimisations enabled.
A68K emulation core enabled for MC68000 emulation.
Musashi emulation core enabled for MC68010/MC68EC020 emulation.
Not emulating any game.
Video settings:
Interface settings: Video plugin not initialised
Audio settings:
Interface settings: Audio plugin not initialised
Input settings:
Interface settings: Input plugin not initialised
Profiling settings:
Interface settings: Profiling module not initialised
----------------------------------------------------------------------
Detailed process information about fba.exe:
fba.exe (base address 0x00400000, size 23944 KB)
Exception occurred in module fba.exe:
fba.exe (base address 0x00400000, size 23944 KB)
Modules loaded by fba.exe:
ntdll.dll (base address 0x7C910000, size 724 KB)
kernel32.dll (base address 0x7C800000, size 1036 KB)
ADVAPI32.dll (base address 0x77DA0000, size 688 KB)
RPCRT4.dll (base address 0x77E50000, size 584 KB)
Secur32.dll (base address 0x77FC0000, size 68 KB)
COMCTL32.DLL (base address 0x773A0000, size 1036 KB)
msvcrt.dll (base address 0x77BE0000, size 352 KB)
GDI32.dll (base address 0x77EF0000, size 292 KB)
USER32.dll (base address 0x7E390000, size 580 KB)
SHLWAPI.dll (base address 0x77F40000, size 472 KB)
COMDLG32.DLL (base address 0x76360000, size 296 KB)
SHELL32.dll (base address 0x7E6A0000, size 8324 KB)
ole32.dll (base address 0x774B0000, size 1268 KB)
setupapi.dll (base address 0x778F0000, size 988 KB)
WINMM.DLL (base address 0x76B00000, size 184 KB)
IMM32.DLL (base address 0x76340000, size 116 KB)
dsound.dll (base address 0x73E90000, size 368 KB)
VERSION.dll (base address 0x77BD0000, size 32 KB)
dinput8.dll (base address 0x6D320000, size 232 KB)
ddraw.dll (base address 0x736E0000, size 300 KB)
DCIMAN32.dll (base address 0x73B40000, size 24 KB)
d3d9.dll (base address 0x4FD10000, size 1688 KB)
d3d8thk.dll (base address 0x6DEA0000, size 24 KB)
D3DX9_43.dll (base address 0x10000000, size 2044 KB)
uxtheme.dll (base address 0x5B150000, size 224 KB)
MSCTF.dll (base address 0x746B0000, size 304 KB)
msctfime.ime (base address 0x75160000, size 184 KB)
----------------------------------------------------------------------
-
bug:
Fba will crash after run Gradius II.
-
bug:
Fba will crash after run Gradius II.
Thanks - fixed it. :)
Should have affected all games in that driver (d_twin16).
-
Hi all, i compile this release, after of compiler, open the fba.exe and have a guru meditation at 0x0000a662
could you please help me. Many thanks, and happy new year.
LOG:
FB Alpha v0.2.97.17 fatal exception report (Fri
----------------------------------------------------------------------
Exception 0xC000001D (EXCEPTION_ILLEGAL_INSTRUCTION) thrown.
EIP: 0x0
EAX: 0x00007020, EBX: 0x00000000, ECX: 0x0022D9D0, EDX: 0x00000001
ESI: 0xFFFFFFFF, EDI: 0xFFFFFFFE, ESP: 0x0022D9C0, EBP: 0x00252356
----------------------------------------------------------------------
System information:
OS: Microsoft Windows XP Professional Service Pack 3 (build 2600)
CPU: AuthenticAMD, Athlon XP "Thorton" / Athon XP "Barton"
2191 MHz, MMX, SSE (1 system processor)
Physical RAM: 2096628 KB (2047 MB) total, 1360144 KB (1328 MB) avail
Total RAM: 4038128 KB (3943 MB) total, 3333428 KB (3255 MB) avail
FB Alpha N3O: 6364 KB in use (6364 KB peak, 16736 KB virtual)
Installed displays and display adapters:
Philips Consumer Electronics Co. Philips 105E (105E1) on NVIDIA Ge
Display size ~28x21cm, Gamma 2,87
Max. bandwidth 70 MHz, H sync 30-54 KHz, V sync 50-120 Hz
----------------------------------------------------------------------
FB Alpha information:
Built on Dec 30 2011, 13:51:54, using GCC 4.6.1.
Optimised for X86 CPUs.
Using Unicode for all text.
Debug functionality absent.
MMX optimisations enabled.
A68K emulation core enabled for MC68000 emulation.
Musashi emulation core enabled for MC68010/MC68EC020 emulation.
Not emulating any game.
Video settings:
Interface settings: Video plugin not initialised
Audio settings:
Interface settings: Audio plugin not initialised
Input settings:
Interface settings: Input plugin not initialised
Profiling settings:
Interface settings: Profiling module not initialised
----------------------------------------------------------------------
Detailed process information about fba.exe:
fba.exe (base address 0x00400000, size 23944 KB)
Exception occurred in module fba.exe:
fba.exe (base address 0x00400000, size 23944 KB)
Modules loaded by fba.exe:
ntdll.dll (base address 0x7C910000, size 724 KB)
kernel32.dll (base address 0x7C800000, size 1036 KB)
ADVAPI32.dll (base address 0x77DA0000, size 688 KB)
RPCRT4.dll (base address 0x77E50000, size 584 KB)
Secur32.dll (base address 0x77FC0000, size 68 KB)
COMCTL32.DLL (base address 0x773A0000, size 1036 KB)
msvcrt.dll (base address 0x77BE0000, size 352 KB)
GDI32.dll (base address 0x77EF0000, size 292 KB)
USER32.dll (base address 0x7E390000, size 580 KB)
SHLWAPI.dll (base address 0x77F40000, size 472 KB)
COMDLG32.DLL (base address 0x76360000, size 296 KB)
SHELL32.dll (base address 0x7E6A0000, size 8324 KB)
ole32.dll (base address 0x774B0000, size 1268 KB)
setupapi.dll (base address 0x778F0000, size 988 KB)
WINMM.DLL (base address 0x76B00000, size 184 KB)
IMM32.DLL (base address 0x76340000, size 116 KB)
dsound.dll (base address 0x73E90000, size 368 KB)
VERSION.dll (base address 0x77BD0000, size 32 KB)
dinput8.dll (base address 0x6D320000, size 232 KB)
ddraw.dll (base address 0x736E0000, size 300 KB)
DCIMAN32.dll (base address 0x73B40000, size 24 KB)
d3d9.dll (base address 0x4FD10000, size 1688 KB)
d3d8thk.dll (base address 0x6DEA0000, size 24 KB)
D3DX9_43.dll (base address 0x10000000, size 2044 KB)
uxtheme.dll (base address 0x5B150000, size 224 KB)
MSCTF.dll (base address 0x746B0000, size 304 KB)
msctfime.ime (base address 0x75160000, size 184 KB)
----------------------------------------------------------------------
-
Thanks Haze - hopefully I should get some time over Christmas to take a look at it again.
Found the time over the weekend, and found the issue - all the decoic16 based games now have sound.
And Robocop works in the DEC0 driver, I'll add the other games to it soon.
-
Ultimate Mortal Kombat 3 game is still sizzling sounds [(There are still a little) at selection screen, on key down; squeaky sound is heard]
FBA emulator Ultimate Mortal Kombat 3 game to excessive fast play. [ridiculously fast (I'm using FBA rom)]
Edit: Emulator`s demo play to speed normal, but emulator`s game play to excessive fast!
-
Ultimate Mortal Kombat 3 game is still sizzling sounds [(There are still a little) at selection screen, on key down; squeaky sound is heard]
FBA emulator Ultimate Mortal Kombat 3 game to excessive fast play. [ridiculously fast (I'm using FBA rom)]
Edit: Emulator`s demo play to speed normal, but emulator`s game play to excessive fast!
I've purposely been ignoring your rude requests, especially since the Megadrive driver is very much WIP. That said, I have tried the game in question and I notice no issues at all.
-
i was kinda unsure if i should post this as im not using FBA on a PC but on an x**x so it might just be a console issue also this isnt
related to this release as ive had this bug over a few different releases now anyway here's the issue in battleshark the game seems
to lock up if your sub takes a hit it doesnt happen in the same stage in the game every time but is guaranteed to happen any time
between stage 1 and 2 i have tried all clones and the same thing happens
-
I've purposely been ignoring your rude requests, especially since the Megadrive driver is very much WIP. That said, I have tried the game in question and I notice no issues at all.
:confused: :S
I played game at 2 separate computer, and I had the same problems
So the problem is not on my computer!
No scratchy sounds at the original game! :)
Also This is the original speed of the game is actually :rolleyes:
http://www.youtube.com/watch?v=9mT0_PV_-D0
Best Regards
-
Little bug first time running fba.exe.
I'm on Win XP Sp3 with 120 dpi res.
-
actually testing TurboGrafx16 games recently added to FBA and some problem with some games :
- Ghost Manor : Freeze when you go further in the game
- Impossamole : it loads, but locks up immediately and crash FBA
- Parasol Stars - The Story of Bubble Bobble III : it loads, but locks up immediately and crash FBA
-
not sure if this is the correct fix but
in c6280_stream_update() in c6820.cpp
change:
if (!samples) return; // don't update if length is 0
to
if (samples<=0) return; // don't update if length is 0
will make these games run
-
Ah, nice. Thanks lantus. I was probably going to blow way more time than I should have figuring that one out.
-
no problem, some games init with a negative sample value, hence the crash. Like i said, probably not the correct fix for this but it works :)
-
I'll take a look at it later, but having a starting point helps. :)
-
found a bug for you guys
Dragon Breed (both M72 and M81 roms) has graphical "flicker". It worked fine in a earlier release.
-
found a bug for you guys
Dragon Breed (both M72 and M81 roms) has graphical "flicker". It worked fine in a earlier release.
The nec games have very sensitive timing. When I replaced the nec (cpu) core it improved a ton of things, but changed the timings in a lot of games. This broke Ninja Baseball Batman and is probably what's causing the flicker in Dragon Breed.
-
A lot of problems compiling in Windows with MinGW 4.6.1
fba_src_029721.7z
-
A lot of problems compiling in Windows with MinGW 4.6.1
fba_src_029721.7z
In makefile.mingw change this;
ifdef ROM_VERIFY
DEF := $(DEF) -DLSB_FIRST
endif
to;
ifdef LSB_FIRST
DEF := $(DEF) -DLSB_FIRST
endif
-
New final romance:
(http://img594.imageshack.us/img594/4726/fr2ch0128111310.png)
-
No problem here with fba 0.2.97.21 and new Final Romance dump by Razoola...
-
New final romance:
(http://img594.imageshack.us/img594/4726/fr2ch0128111310.png)
I Have the same problem.
This is my ROMset:
# fr2ch.zip
# Files 6
Date Name Size Compressed CRC32
2001-01-01 00:00 098-c2ch.bin 4194304 1055355 226B1263
2001-01-01 00:00 098-c1ch.bin 4194304 1318114 29148BF7
2001-01-01 00:00 098-m1ch.bin 131072 56931 DA4878CF
2001-01-01 00:00 098-p1ch.bin 524288 211464 09675541
2001-01-01 00:00 098-s1ch.bin 131072 26089 0E6A7C73
2001-01-01 00:00 098-v1ch.bin 1048576 671707 6F8CCDDC
# Total 10223616 3339660
-
I Have the same problem.
This is my ROMset:
you have cheats or some kind of high score hack (if FBA does such things like old MAME did) turned on.
incidently, this is why MAME doesn't allow you to boot up with cheats on, despite people requesting it all the time.
-
Ok thanks !
-
you have cheats or some kind of high score hack (if FBA does such things like old MAME did) turned on.
incidently, this is why MAME doesn't allow you to boot up with cheats on, despite people requesting it all the time.
Thanks Haze, you're right, I forgot I had Album feature unlocked (^_^)'
All works OK after pressing F3 (although, cheats were loaded automatically each time FBA started).
Seems to be that Cheat author forgot to type "Default" value to "'0", and he left it as "1"; so this Cheat file was always loaded by default,
This is the correct file:
cheat "Unlock Photo Album"
default 0
0 "Disabled"
1 "Enabled", 0, 0x101ED7, 0x09
However, I didn't see that "cheats" was still enabled, just because at simple view (from menu) it looks disabled (no marks next to text).
So, what I mean is that would be good to add a "checkmark" next to "Enable Cheats" (at the left) to see if they're enabled or not.
Something like this:
-
I have a big sound problem in ninja ryukenden (interference). xaudio or directsound....
All drivers are up to date.
rec input: http://ge.tt/90QJR6D?c
-
found a bug for you guys
Dragon Breed (both M72 and M81 roms) has graphical "flicker". It worked fine in a earlier release.
Fixed this - and fixed the sound in the driver too.
-
Caliber 50 has not Sound
-
Sengoku 3 gives this Error and is not shown as available when its perfect in version 029708.
____________________________________________________________________________
samsho5.zip was found (C:\Users\XXXX\Desktop\ARCADE\Neo-Geo\samsho5.zip).
neogeo.zip was found (C:\Users\XXXXX\Desktop\ARCADE\neogeo.zip).
The ROMset is OK.sengoku3.zip was found (C:\Users\XXXXX\Desktop\ARCADE\sengoku3.zip).
neogeo.zip was found (C:\Users\XXXXX\Desktop\ARCADE\neogeo.zip).
• essential program ROM 261-m1.m1 was not found.
The ROMset exhibits the following problems:
• essential data is missing, the game will not run!
• program data is missing.
__________________________________________
Samsho5 ? :eek:
Also Andro Dundunos gives this error when its fully up to date and all files match,
___________________________________________________________________
androdun.zip was found (C:\Users\XXXXX\Desktop\ARCADE\Neo-Geo\androdun.zip).
neogeo.zip was found (C:\Users\XXXXXX\Desktop\ARCADE\neogeo.zip).
• essential program ROM 049-m1.m1 was not found.
• sound ROM 049-v1.v1 was not found.
The ROMset exhibits the following problems:
• essential data is missing, the game will not run!
• program data is missing.
• sound data is missing.
EDIT: ok there is ALOT of broken Neogeo games , im just going to list them all.
Crossed swords
Fatal Fury 2
Fight Fever
Voltage Fighter Gowcaizer
The Irritating Maze
Legend of Success Joe
Metal Slug 5 (JAMMA PCB)
Pulstar
Sengoku 3
-
Hi SNK_Dude,
All games you've listed work fine under fba 0.2.97.21. Simply it seems your neogeo set is not up to date (especially M1) according to your message. So before reporting broken games (which are not broken), be sure to have the correct neogeo set for latest fba release...
-
I do have the latest set which is the strange part, All files inside the roms are perfectly matched.
and what i mentioned is a bug where it gives the error and shows a completely different rom name like in Sengoku 3s case where it samsho5.zip was found when its a error for Sengoku3.
-
I do have the latest set which is the strange part, All files inside the roms are perfectly matched.
and what i mentioned is a bug where it gives the error and shows a completely different rom name like in Sengoku 3s case where it samsho5.zip was found when its a error for Sengoku3.
Sengoku 3 had a non-terminated unicode title which may have lead to issues with that set and that set only. That said, it still worked fine (but may have displayed title incorrectly, and/or other such things, the issue you saw with looking in and extra zip was probably caused by this).
For the others, like JacKc said, you have the wrong files. The sets used by .08 will not be current for the latest version.
-
Well I have no idea what it could be then since i am using the 0.2.97.21 set.
Edit: I learned why it was giving that error now, I had multiple directories with the same rom name in them.
-
Hi,
I can still confirm this bug:
http://neo-source.com/index.php?topic=1366.msg11338#msg11338
and it's not tied to filters or specific video renderers.
Thanks
-
Hi,
I can still confirm this bug:
http://neo-source.com/index.php?topic=1366.msg11338#msg11338
and it's not tied to filters or specific video renderers.
Thanks
You're right. I thought I had found a fix, but that broke the title in Progear, and also MAME doesn't do anything like it and that has no issues.
I could quite easily fix it by blanking the two specific tiles in a sfa2 specific driver init - which I may do unless any of the team has any better ideas?
-
You're right. I thought I had found a fix, but that broke the title in Progear, and also MAME doesn't do anything like it and that has no issues.
I could quite easily fix it by blanking the two specific tiles in a sfa2 specific driver init - which I may do unless any of the team has any better ideas?
I think it's related to the unemulated palette fade - but for now I'm going to nuke the tiles on init.
-
I think it's related to the unemulated palette fade - but for now I'm going to nuke the tiles on init.
Actually - nuking tiles on init breaks the game over screen - so leaving as is.
Looks like we need emulate palette fade for this one.
-
Made a start on palette fade. Verified palette control register for all CPS-1/2 games (including non-MAME) and added config mapping for it. Will hook up the actual palette calculation tomorrow hopefully.
-
I have found a actual bug this time,
in Crime Fighters the Music and sounds are not properly Emulated , same goes for Metal slug series (crackling in the audio when firing the pistol)
I checked both with mame.
I hope this is what you mean by Bug reporting.
-
Hi,
I've almost lost my mind on this issue..SSF2T and his clones have always been choppy, skipped a lot of frames and seemed to not run correctly (I think that a lot of the complaints in the "Unwanted Frameskip" thread are also because of this).
So I've started to try to analyze the fba sources and I think and hope I've finally found the cause of the problem, so I've recompiled the emulator with my changes and now the game seems to run perfectly.
Below my changes, obviously I'm asking for a confirmation by the devs about it:
file cps_run.cpp, inside Cps2Frame function:
// GetPalette(0, 4); // Get palettes
// CpsObjGet(); // Get objects
for (i = 0; i < 3; i++) {
nNext = ((i + 1) * nDisplayEnd) / 3; // find out next cycle count to run to
while (nNext > nIrqCycles && nInterrupt < MAX_RASTER) {
SekRun(nIrqCycles - SekTotalCycles());
DoIRQ();
}
SekRun(nNext - SekTotalCycles()); // run cpu
}
GetPalette(0, 4); // Get palettes
CpsObjGet(); // Get objects
Pratically I've moved the GetPalette and CpsObjGet calls after the "for" cycle to find out next cycle count to run to.
You can verify the benefits of this change especially while playing on the ryu stage.
I really hope that this doesn't break anything, anyone that knows that game could confirm that actually it's quite unplayable, so please try this fix and let me know.
Thank you
Manlio
-
Hi,
I've almost lost my mind on this issue..SSF2T and his clones have always been choppy, skipped a lot of frames and seemed to not run correctly (I think that a lot of the complaints in the "Unwanted Frameskip" thread are also because of this).
So I've started to try to analyze the fba sources and I think and hope I've finally found the cause of the problem, so I've recompiled the emulator with my changes and now the game seems to run perfectly.
Below my changes, obviously I'm asking for a confirmation by the devs about it:
file cps_run.cpp, inside Cps2Frame function:
// GetPalette(0, 4); // Get palettes
// CpsObjGet(); // Get objects
for (i = 0; i < 3; i++) {
nNext = ((i + 1) * nDisplayEnd) / 3; // find out next cycle count to run to
while (nNext > nIrqCycles && nInterrupt < MAX_RASTER) {
SekRun(nIrqCycles - SekTotalCycles());
DoIRQ();
}
SekRun(nNext - SekTotalCycles()); // run cpu
}
GetPalette(0, 4); // Get palettes
CpsObjGet(); // Get objects
Pratically I've moved the GetPalette and CpsObjGet calls after the "for" cycle to find out next cycle count to run to.
You can verify the benefits of this change especially while playing on the ryu stage.
I really hope that this doesn't break anything, anyone that knows that game could confirm that actually it's quite unplayable, so please try this fix and let me know.
Thank you
Manlio
CPS has had some palette related improvements in the current build, and GetPalette() is no longer called in the frame function, instead it's called when the palette register is written to. This is more accurate than before.
I'll look at the timing of the CpsObjGet() call in the current build though.
-
Thank You,
You and IQ are really great developers that listen users' feedback.
I really hope you get this issue sorted out so that CPS rendering feels smoother.
Thank you very very much for all your priceless effort ^_^
-
One last thing..It seems that the default volume for CPS2 games is a bit louder when compared with other drivers at the minimum master volume.
Thanks
-
libsnes fails to build from source after stuff was moved around, this fixes it:
https://gist.github.com/2004124
Also, including png.h is unneeded for libsnes, and to avoid build breaking for libsnes/Win32:
https://gist.github.com/2004169 (You might want some HAVE_LIBPNG define or something like that instead).
-
Apparently some parts of the last patch were not applied on SVN. Also fixed build issues on OSX. The fopen64() stuff is kinda weird though. It doesn't exist in standard C at least, I'd suppose it's a particular quirk with some systems (large file extensions). It doesn't exist on OSX at least where fopen() is always 64-bit. Probably better to just use fopen()/ftell()/fseek().
https://gist.github.com/2007780
-
About cheat:
Fba limits cheat options number and string length, it results that a part of cheats aren't used.
see here:
http://f.ppxclub.com/141688-1-1
http://f.ppxclub.com/144005-1-1
-
About cheat:
Fba limits cheat options number and string length, it results that a part of cheats aren't used.
see here:
http://f.ppxclub.com/141688-1-1
http://f.ppxclub.com/144005-1-1
I'd consider upping the limit of the options, but 1024 for a line is surely plenty.
You can only fit so much in the dialog, and there is also a limit of 64 addresses, assuming max 20 chars for address, value, cheat type combination, that would be 1280 chars. I might consider doubling to 2048, but I can't see any reason for over 6000 chars in a line at all.
-
Not a big deal but playing 88' Games and clones I noticed in the Long Jump event the shadow the Long Jumper makes in the sand is pink compared to the shadow in Mame which is dark.
-
I'd consider upping the limit of the options, but 1024 for a line is surely plenty.
You can only fit so much in the dialog, and there is also a limit of 64 addresses, assuming max 20 chars for address, value, cheat type combination, that would be 1280 chars. I might consider doubling to 2048, but I can't see any reason for over 6000 chars in a line at all.
You can view wayder cheat, truxton cheat exceed 100 chars for address, fba cheat conver from wayder and pagsy cheat.
http://www.geocities.jp/ued_way/
-
fba cheat - truxton.ini
cheat "Infinite Lives"
default 0
0 "Disabled"
1 "Enabled", 0, 0x81D17, 0x0A
cheat "Infinite Skulls"
default 0
0 "Disabled"
1 "Enabled", 0, 0x81847, 0x0B
cheat "Invincibility"
default 0
0 "Disabled"
1 "Enabled", 0, 0x81838, 0x00
2 "Disabled", 0, 0x81838, 0x80
cheat "Get Max Speed Now!"
default 0
0 "Disabled"
1 "Enabled", 0, 0x81855, 0x22
cheat "Always Max Speed"
default 0
0 "Disabled"
1 "Enabled", 0, 0x81855, 0x22
cheat "Get 4 P Now!"
default 0
0 "Disabled"
1 "Enabled", 0, 0x81D2D, 0x04
cheat "Select Weapon"
default 0
0 "Disabled"
1 "Normal", 0, 0x81845, 0x00
2 "Better", 0, 0x81845, 0x01
3 "Best", 0, 0x81845, 0x02
cheat "Select Fire"
default 0
0 "Disabled"
1 "Red", 0, 0x81841, 0x00, 0, 0x81843, 0x00
2 "Green", 0, 0x81841, 0x08, 0, 0x81843, 0x08
3 "Blue", 0, 0x81841, 0x10, 0, 0x81843, 0x10
cheat " "
cheat ">>> Cheats are added under <<<"
cheat " "
cheat "Rapid Fire"
default 0
0 "Disabled"
1 "Enabled", 0, 0x81849, 0x18
cheat "Select Another Weapon"
default 0
0 "Disabled"
1 "Multi", 0, 0x80D60, 0x01, 0, 0x80D61, 0xF9, 0, 0x80D80, 0x01, 0, 0x80D81, 0xF9, 0, 0x80DA0, 0x01, 0, 0x80DA1, 0xF9, 0, 0x80DC0, 0x01, 0, 0x80DC1, 0xF9, 0, 0x80DE0, 0x01, 0, 0x80DE1, 0xF9, 0, 0x80E00, 0x01, 0, 0x80E01, 0xF9, 0, 0x80E20, 0x01, 0, 0x80E21, 0xF9, 0, 0x80E40, 0x01, 0, 0x80E41, 0xF9, 0, 0x80E60, 0x01, 0, 0x80E61, 0xF9, 0, 0x80E80, 0x01, 0, 0x80E81, 0xF9, 0, 0x80EA0, 0x01, 0, 0x80EA1, 0xF9, 0, 0x80EC0, 0x01, 0, 0x80EC1, 0xF9, 0, 0x80EE0, 0x01, 0, 0x80EE1, 0xF9, 0, 0x80F00, 0x01, 0, 0x80F01, 0xF9, 0, 0x80F20, 0x01, 0, 0x80F21, 0xF9, 0, 0x80F40, 0x01, 0, 0x80F41, 0xF9, 0, 0x80F60, 0x01, 0, 0x80F61, 0xF9, 0, 0x80F80, 0x01, 0, 0x80F81, 0xF9, 0, 0x80FA0, 0x01, 0, 0x80FA1, 0xF9, 0, 0x80FC0, 0x01, 0, 0x80FC1, 0xF9, 0, 0x80FE0, 0x01, 0, 0x80FE1, 0xF9, 0, 0x81000, 0x01, 0, 0x81001, 0xF9, 0, 0x81020, 0x01, 0, 0x81021, 0xF9, 0, 0x81040, 0x01, 0, 0x81041, 0xF9, 0, 0x81060, 0x01, 0, 0x81061, 0xF9, 0, 0x81080, 0x01, 0, 0x81081, 0xF9, 0, 0x810A0, 0x01, 0, 0x810A1, 0xF9, 0, 0x810C0, 0x01, 0, 0x810C1, 0xF9, 0, 0x810E0, 0x01, 0, 0x810E1, 0xF9, 0, 0x81100, 0x01, 0, 0x81101, 0xF9, 0, 0x81120, 0x01, 0, 0x81121, 0xF9, 0, 0x81140, 0x01, 0, 0x81141, 0xF9, 0, 0x81160, 0x01, 0, 0x81161, 0xF9, 0, 0x81180, 0x01, 0, 0x81181, 0xF9, 0, 0x811A0, 0x01, 0, 0x811A1, 0xF9, 0, 0x811C0, 0x01, 0, 0x811C1, 0xF9, 0, 0x811E0, 0x01, 0, 0x811E1, 0xF9, 0, 0x81200, 0x01, 0, 0x81201, 0xF9, 0, 0x81220, 0x01, 0, 0x81221, 0xF9, 0, 0x81240, 0x01, 0, 0x81241, 0xF9, 0, 0x81260, 0x01, 0, 0x81261, 0xF9, 0, 0x81280, 0x01, 0, 0x81281, 0xF9, 0, 0x812A0, 0x01, 0, 0x812A1, 0xF9, 0, 0x812C0, 0x01, 0, 0x812C1, 0xF9, 0, 0x812E0, 0x01, 0, 0x812E1, 0xF9, 0, 0x81300, 0x01, 0, 0x81301, 0xF9, 0, 0x81320, 0x01, 0, 0x81321, 0xF9, 0, 0x81340, 0x01, 0, 0x81341, 0xF9, 0, 0x81360, 0x01, 0, 0x81361, 0xF9, 0, 0x81380, 0x01, 0, 0x81381, 0xF9, 0, 0x813A0, 0x01, 0, 0x813A1, 0xF9, 0, 0x813C0, 0x01, 0, 0x813C1, 0xF9, 0, 0x813E0, 0x01, 0, 0x813E1, 0xF9, 0, 0x81400, 0x01, 0, 0x81401, 0xF9, 0, 0x81420, 0x01, 0, 0x81421, 0xF9, 0, 0x81440, 0x01, 0, 0x81441, 0xF9, 0, 0x81460, 0x01, 0, 0x81461, 0xF9, 0, 0x81480, 0x01, 0, 0x81481, 0xF9, 0, 0x814A0, 0x01, 0, 0x814A1, 0xF9, 0, 0x814C0, 0x01, 0, 0x814C1, 0xF9, 0, 0x81840, 0x00, 0, 0x81841, 0x00, 0, 0x81842, 0x00, 0, 0x81843, 0x00, 0, 0x80D71, 0x80, 0, 0x80D72, 0xF5, 0, 0x80D73, 0x20, 0, 0x80D74, 0xFC, 0, 0x80D91, 0x80, 0, 0x80D92, 0xF5, 0, 0x80D93, 0x20, 0, 0x80D94, 0xFC, 0, 0x80DB1, 0x80, 0, 0x80DB2, 0xF5, 0, 0x80DB3, 0x20, 0, 0x80DB4, 0xFC, 0, 0x80DD1, 0x80, 0, 0x80DD2, 0xF5, 0, 0x80DD3, 0x20, 0, 0x80DD4, 0xFC, 0, 0x80DF1, 0x80, 0, 0x80DF2, 0xF8, 0, 0x80DF3, 0xFF, 0, 0x80DF4, 0xF8, 0, 0x80E11, 0x80, 0, 0x80E12, 0xF8, 0, 0x80E13, 0xFF, 0, 0x80E14, 0xF8, 0, 0x80E31, 0x80, 0, 0x80E32, 0xF8, 0, 0x80E33, 0xFF, 0, 0x80E34, 0xF8, 0, 0x80E51, 0x80, 0, 0x80E52, 0xF8, 0, 0x80E53, 0xFF, 0, 0x80E54, 0xF8, 0, 0x80E71, 0x80, 0, 0x80E72, 0xF8, 0, 0x80E73, 0x01, 0, 0x80E74, 0x00, 0, 0x80E91, 0x80, 0, 0x80E92, 0xF8, 0, 0x80E93, 0x01, 0, 0x80E94, 0x00, 0, 0x80EB1, 0x80, 0, 0x80EB2, 0xF8, 0, 0x80EB3, 0x01, 0, 0x80EB4, 0x00, 0, 0x80ED1, 0x80, 0, 0x80ED2, 0xF8, 0, 0x80ED3, 0x01, 0, 0x80ED4, 0x00, 0, 0x80EF1, 0x80, 0, 0x80EF2, 0xF8, 0, 0x80EF3, 0x38, 0, 0x80EF4, 0xFC, 0, 0x80F11, 0x80, 0, 0x80F12, 0xF8, 0, 0x80F13, 0x38, 0, 0x80F14, 0xFC, 0, 0x80F31, 0x80, 0, 0x80F32, 0xF8, 0, 0x80F33, 0x38, 0, 0x80F34, 0xFC, 0, 0x80F51, 0x80, 0, 0x80F52, 0xF8, 0, 0x80F53, 0x38, 0, 0x80F54, 0xFC, 0, 0x80F71, 0x80, 0, 0x80F72, 0xF8, 0, 0x80F73, 0x30, 0, 0x80F74, 0xFC, 0, 0x80F91, 0x80, 0, 0x80F92, 0xF8, 0, 0x80F93, 0x30, 0, 0x80F94, 0xFC, 0, 0x80FB1, 0x80, 0, 0x80FB2, 0xF8, 0, 0x80FB3, 0x30, 0, 0x80FB4, 0xFC, 0, 0x80FD1, 0x80, 0, 0x80FD2, 0xF8, 0, 0x80FD3, 0x30, 0, 0x80FD4, 0xFC, 0, 0x80FF1, 0x80, 0, 0x80FF2, 0xF8, 0, 0x80FF3, 0x25, 0, 0x80FF4, 0xFC, 0, 0x81011, 0x80, 0, 0x81012, 0xF8, 0, 0x81013, 0x25, 0, 0x81014, 0xFC, 0, 0x81031, 0x80, 0, 0x81032, 0xF8, 0, 0x81033, 0x25, 0, 0x81034, 0xFC, 0, 0x81051, 0x80, 0, 0x81052, 0xF8, 0, 0x81053, 0x25, 0, 0x81054, 0xFC, 0, 0x81071, 0x80, 0, 0x81072, 0xF8, 0, 0x81073, 0x08, 0, 0x81074, 0xFC, 0, 0x81091, 0x80, 0, 0x81092, 0xF8, 0, 0x81093, 0x08, 0, 0x81094, 0xFC, 0, 0x810B1, 0x80, 0, 0x810B2, 0xF8, 0, 0x810B3, 0x08, 0, 0x810B4, 0xFC, 0, 0x810D1, 0x80, 0, 0x810D2, 0xF8, 0, 0x810D3, 0x08, 0, 0x810D4, 0xFC, 0, 0x810F1, 0x80, 0, 0x810F2, 0xF8, 0, 0x810F3, 0x10, 0, 0x810F4, 0xFC, 0, 0x81111, 0x80, 0, 0x81112, 0xF8, 0, 0x81113, 0x10, 0, 0x81114, 0xFC, 0, 0x81131, 0x80, 0, 0x81132, 0xF8, 0, 0x81133, 0x10, 0, 0x81134, 0xFC, 0, 0x81151, 0x80, 0, 0x81152, 0xF8, 0, 0x81153, 0x10, 0, 0x81154, 0xFC, 0, 0x81171, 0x80, 0, 0x81172, 0xF8, 0, 0x81173, 0x1B, 0, 0x81174, 0xFC, 0, 0x81191, 0x80, 0, 0x81192, 0xF8, 0, 0x81193, 0x1B, 0, 0x81194, 0xFC, 0, 0x811B1, 0x80, 0, 0x811B2, 0xF8, 0, 0x811B3, 0x1B, 0, 0x811B4, 0xFC, 0, 0x811D1, 0x80, 0, 0x811D2, 0xF8, 0, 0x811D3, 0x1B, 0, 0x811D4, 0xFC, 0, 0x811F1, 0x80, 0, 0x811F2, 0x00, 0, 0x811F3, 0x3B, 0, 0x811F4, 0x04, 0, 0x81211, 0x80, 0, 0x81212, 0x00, 0, 0x81213, 0x3B, 0, 0x81214, 0x04, 0, 0x81231, 0x80, 0, 0x81232, 0x00, 0, 0x81233, 0x3B, 0, 0x81234, 0x04, 0, 0x81251, 0x80, 0, 0x81252, 0x00, 0, 0x81253, 0x3B, 0, 0x81254, 0x04, 0, 0x81271, 0x80, 0, 0x81272, 0x00, 0, 0x81273, 0x34, 0, 0x81274, 0x04, 0, 0x81291, 0x80, 0, 0x81292, 0x00, 0, 0x81293, 0x34, 0, 0x81294, 0x04, 0, 0x812B1, 0x80, 0, 0x812B2, 0x00, 0, 0x812B3, 0x34, 0, 0x812B4, 0x04, 0, 0x812D1, 0x80, 0, 0x812D2, 0x00, 0, 0x812D3, 0x34, 0, 0x812D4, 0x04, 0, 0x812F1, 0x80, 0, 0x812F2, 0x00, 0, 0x812F3, 0x2B, 0, 0x812F4, 0x04, 0, 0x81311, 0x80, 0, 0x81312, 0x00, 0, 0x81313, 0x2B, 0, 0x81314, 0x04, 0, 0x81331, 0x80, 0, 0x81332, 0x00, 0, 0x81333, 0x2B, 0, 0x81334, 0x04, 0, 0x81351, 0x80, 0, 0x81352, 0x00, 0, 0x81353, 0x2B, 0, 0x81354, 0x04, 0, 0x81371, 0x80, 0, 0x81372, 0x00, 0, 0x81373, 0x05, 0, 0x81374, 0xF4, 0, 0x81391, 0x80, 0, 0x81392, 0x00, 0, 0x81393, 0x05, 0, 0x81394, 0xF4, 0, 0x813B1, 0x80, 0, 0x813B2, 0x00, 0, 0x813B3, 0x05, 0, 0x813B4, 0xF4, 0, 0x813D1, 0x80, 0, 0x813D2, 0x00, 0, 0x813D3, 0x05, 0, 0x813D4, 0xF4, 0, 0x813F1, 0x80, 0, 0x813F2, 0x00, 0, 0x813F3, 0x0C, 0, 0x813F4, 0xF4, 0, 0x81411, 0x80, 0, 0x81412, 0x00, 0, 0x81413, 0x0C, 0, 0x81414, 0xF4, 0, 0x81431, 0x80, 0, 0x81432, 0x00, 0, 0x81433, 0x0C, 0, 0x81434, 0xF4, 0, 0x81451, 0x80, 0, 0x81452, 0x00, 0, 0x81453, 0x0C, 0, 0x81454, 0xF4, 0, 0x81471, 0x80, 0, 0x81472, 0x00, 0, 0x81473, 0x15, 0, 0x81474, 0xF4, 0, 0x81491, 0x80, 0, 0x81492, 0x00, 0, 0x81493, 0x15, 0, 0x81494, 0xF4, 0, 0x814B1, 0x80, 0, 0x814B2, 0x00, 0, 0x814B3, 0x15, 0, 0x814B4, 0xF4, 0, 0x814D1, 0x80, 0, 0x814D2, 0x00, 0, 0x814D3, 0x15, 0, 0x814D4, 0xF4
2 "Wide", 0, 0x80D60, 0x04, 0, 0x80D61, 0x00, 0, 0x80D80, 0x04, 0, 0x80D81, 0x00, 0, 0x80DA0, 0x04, 0, 0x80DA1, 0x00, 0, 0x80DC0, 0x04, 0, 0x80DC1, 0x00, 0, 0x80DE0, 0x04, 0, 0x80DE1, 0x00, 0, 0x80E00, 0x04, 0, 0x80E01, 0x00, 0, 0x80E20, 0x04, 0, 0x80E21, 0x00, 0, 0x80E40, 0x04, 0, 0x80E41, 0x00, 0, 0x80E60, 0x04, 0, 0x80E61, 0x00, 0, 0x80E80, 0x04, 0, 0x80E81, 0x00, 0, 0x80EA0, 0x04, 0, 0x80EA1, 0x00, 0, 0x80EC0, 0x04, 0, 0x80EC1, 0x00, 0, 0x80EE0, 0x04, 0, 0x80EE1, 0x00, 0, 0x80F00, 0x04, 0, 0x80F01, 0x00, 0, 0x80F20, 0x04, 0, 0x80F21, 0x00, 0, 0x80F40, 0x04, 0, 0x80F41, 0x00, 0, 0x80F60, 0x04, 0, 0x80F61, 0x00, 0, 0x80F80, 0x04, 0, 0x80F81, 0x00, 0, 0x80FA0, 0x04, 0, 0x80FA1, 0x00, 0, 0x80FC0, 0x04, 0, 0x80FC1, 0x00, 0, 0x80FE0, 0x04, 0, 0x80FE1, 0x00, 0, 0x81000, 0x04, 0, 0x81001, 0x00, 0, 0x81020, 0x04, 0, 0x81021, 0x00, 0, 0x81040, 0x04, 0, 0x81041, 0x00, 0, 0x81060, 0x04, 0, 0x81061, 0x00, 0, 0x81080, 0x04, 0, 0x81081, 0x00, 0, 0x810A0, 0x04, 0, 0x810A1, 0x00, 0, 0x810C0, 0x04, 0, 0x810C1, 0x00, 0, 0x810E0, 0x04, 0, 0x810E1, 0x00, 0, 0x81100, 0x04, 0, 0x81101, 0x00, 0, 0x81120, 0x04, 0, 0x81121, 0x00, 0, 0x81140, 0x04, 0, 0x81141, 0x00, 0, 0x81160, 0x04, 0, 0x81161, 0x00, 0, 0x81180, 0x04, 0, 0x81181, 0x00, 0, 0x811A0, 0x04, 0, 0x811A1, 0x00, 0, 0x811C0, 0x04, 0, 0x811C1, 0x00, 0, 0x811E0, 0x04, 0, 0x811E1, 0x00, 0, 0x81200, 0x04, 0, 0x81201, 0x00, 0, 0x81220, 0x04, 0, 0x81221, 0x00, 0, 0x81240, 0x04, 0, 0x81241, 0x00, 0, 0x81260, 0x04, 0, 0x81261, 0x00, 0, 0x81280, 0x04, 0, 0x81281, 0x00, 0, 0x812A0, 0x04, 0, 0x812A1, 0x00, 0, 0x812C0, 0x04, 0, 0x812C1, 0x00, 0, 0x812E0, 0x04, 0, 0x812E1, 0x00, 0, 0x81300, 0x04, 0, 0x81301, 0x00, 0, 0x81320, 0x04, 0, 0x81321, 0x00, 0, 0x81340, 0x04, 0, 0x81341, 0x00, 0, 0x81360, 0x04, 0, 0x81361, 0x00, 0, 0x81380, 0x04, 0, 0x81381, 0x00, 0, 0x813A0, 0x04, 0, 0x813A1, 0x00, 0, 0x813C0, 0x04, 0, 0x813C1, 0x00, 0, 0x813E0, 0x04, 0, 0x813E1, 0x00, 0, 0x81400, 0x04, 0, 0x81401, 0x00, 0, 0x81420, 0x04, 0, 0x81421, 0x00, 0, 0x81440, 0x04, 0, 0x81441, 0x00, 0, 0x81460, 0x04, 0, 0x81461, 0x00, 0, 0x81480, 0x04, 0, 0x81481, 0x00, 0, 0x814A0, 0x04, 0, 0x814A1, 0x00, 0, 0x814C0, 0x04, 0, 0x814C1, 0x00, 0, 0x81840, 0x00, 0, 0x81841, 0x00, 0, 0x81842, 0x00, 0, 0x81843, 0x00, 0, 0x80D71, 0x60, 0, 0x80D72, 0xFC, 0, 0x80D73, 0x00, 0, 0x80D74, 0xFC, 0, 0x80D91, 0x60, 0, 0x80D92, 0xFC, 0, 0x80D93, 0x00, 0, 0x80D94, 0xFC, 0, 0x80DB1, 0x60, 0, 0x80DB2, 0xFC, 0, 0x80DB3, 0x00, 0, 0x80DB4, 0xFC, 0, 0x80DD1, 0x60, 0, 0x80DD2, 0xFC, 0, 0x80DD3, 0x00, 0, 0x80DD4, 0xFC, 0, 0x80DF1, 0x60, 0, 0x80DF2, 0xFC, 0, 0x80DF3, 0x3F, 0, 0x80DF4, 0xFC, 0, 0x80E11, 0x60, 0, 0x80E12, 0xFC, 0, 0x80E13, 0x3F, 0, 0x80E14, 0xFC, 0, 0x80E31, 0x60, 0, 0x80E32, 0xFC, 0, 0x80E33, 0x3F, 0, 0x80E34, 0xFC, 0, 0x80E51, 0x60, 0, 0x80E52, 0xFC, 0, 0x80E53, 0x3F, 0, 0x80E54, 0xFC, 0, 0x80E71, 0x60, 0, 0x80E72, 0xFC, 0, 0x80E73, 0x01, 0, 0x80E74, 0xFC, 0, 0x80E91, 0x60, 0, 0x80E92, 0xFC, 0, 0x80E93, 0x01, 0, 0x80E94, 0xFC, 0, 0x80EB1, 0x60, 0, 0x80EB2, 0xFC, 0, 0x80EB3, 0x01, 0, 0x80EB4, 0xFC, 0, 0x80ED1, 0x60, 0, 0x80ED2, 0xFC, 0, 0x80ED3, 0x01, 0, 0x80ED4, 0xFC, 0, 0x80EF1, 0x60, 0, 0x80EF2, 0xFF, 0, 0x80EF3, 0x3E, 0, 0x80EF4, 0x02, 0, 0x80F11, 0x60, 0, 0x80F12, 0xFF, 0, 0x80F13, 0x3E, 0, 0x80F14, 0x02, 0, 0x80F31, 0x60, 0, 0x80F32, 0xFF, 0, 0x80F33, 0x3E, 0, 0x80F34, 0x02, 0, 0x80F51, 0x60, 0, 0x80F52, 0xFF, 0, 0x80F53, 0x3E, 0, 0x80F54, 0x02, 0, 0x80F71, 0x60, 0, 0x80F72, 0xFF, 0, 0x80F73, 0x3D, 0, 0x80F74, 0x02, 0, 0x80F91, 0x60, 0, 0x80F92, 0xFF, 0, 0x80F93, 0x3D, 0, 0x80F94, 0x02, 0, 0x80FB1, 0x60, 0, 0x80FB2, 0xFF, 0, 0x80FB3, 0x3D, 0, 0x80FB4, 0x02, 0, 0x80FD1, 0x60, 0, 0x80FD2, 0xFF, 0, 0x80FD3, 0x3D, 0, 0x80FD4, 0x02, 0, 0x80FF1, 0x60, 0, 0x80FF2, 0xFF, 0, 0x80FF3, 0x3C, 0, 0x80FF4, 0x02, 0, 0x81011, 0x60, 0, 0x81012, 0xFF, 0, 0x81013, 0x3C, 0, 0x81014, 0x02, 0, 0x81031, 0x60, 0, 0x81032, 0xFF, 0, 0x81033, 0x3C, 0, 0x81034, 0x02, 0, 0x81051, 0x60, 0, 0x81052, 0xFF, 0, 0x81053, 0x3C, 0, 0x81054, 0x02, 0, 0x81071, 0x60, 0, 0x81072, 0xFF, 0, 0x81073, 0x02, 0, 0x81074, 0xF6, 0, 0x81091, 0x60, 0, 0x81092, 0xFF, 0, 0x81093, 0x02, 0, 0x81094, 0xF6, 0, 0x810B1, 0x60, 0, 0x810B2, 0xFF, 0, 0x810B3, 0x02, 0, 0x810B4, 0xF6, 0, 0x810D1, 0x60, 0, 0x810D2, 0xFF, 0, 0x810D3, 0x02, 0, 0x810D4, 0xF6, 0, 0x810F1, 0x60, 0, 0x810F2, 0xFF, 0, 0x810F3, 0x03, 0, 0x810F4, 0xF6, 0, 0x81111, 0x60, 0, 0x81112, 0xFF, 0, 0x81113, 0x03, 0, 0x81114, 0xF6, 0, 0x81131, 0x60, 0, 0x81132, 0xFF, 0, 0x81133, 0x03, 0, 0x81134, 0xF6, 0, 0x81151, 0x60, 0, 0x81152, 0xFF, 0, 0x81153, 0x03, 0, 0x81154, 0xF6, 0, 0x81171, 0x60, 0, 0x81172, 0xFF, 0, 0x81173, 0x04, 0, 0x81174, 0xF6, 0, 0x81191, 0x60, 0, 0x81192, 0xFF, 0, 0x81193, 0x04, 0, 0x81194, 0xF6, 0, 0x811B1, 0x60, 0, 0x811B2, 0xFF, 0, 0x811B3, 0x04, 0, 0x811B4, 0xF6, 0, 0x811D1, 0x60, 0, 0x811D2, 0xFF, 0, 0x811D3, 0x04, 0, 0x811D4, 0xF6, 0, 0x811F1, 0x60, 0, 0x811F2, 0x08, 0, 0x811F3, 0x3B, 0, 0x811F4, 0x08, 0, 0x81211, 0x60, 0, 0x81212, 0x08, 0, 0x81213, 0x3B, 0, 0x81214, 0x08, 0, 0x81231, 0x60, 0, 0x81232, 0x08, 0, 0x81233, 0x3B, 0, 0x81234, 0x08, 0, 0x81251, 0x60, 0, 0x81252, 0x08, 0, 0x81253, 0x3B, 0, 0x81254, 0x08, 0, 0x81271, 0x60, 0, 0x81272, 0x08, 0, 0x81273, 0x3A, 0, 0x81274, 0x08, 0, 0x81291, 0x60, 0, 0x81292, 0x08, 0, 0x81293, 0x3A, 0, 0x81294, 0x08, 0, 0x812B1, 0x60, 0, 0x812B2, 0x08, 0, 0x812B3, 0x3A, 0, 0x812B4, 0x08, 0, 0x812D1, 0x60, 0, 0x812D2, 0x08, 0, 0x812D3, 0x3A, 0, 0x812D4, 0x08, 0, 0x812F1, 0x60, 0, 0x812F2, 0x08, 0, 0x812F3, 0x39, 0, 0x812F4, 0x08, 0, 0x81311, 0x60, 0, 0x81312, 0x08, 0, 0x81313, 0x39, 0, 0x81314, 0x08, 0, 0x81331, 0x60, 0, 0x81332, 0x08, 0, 0x81333, 0x39, 0, 0x81334, 0x08, 0, 0x81351, 0x60, 0, 0x81352, 0x08, 0, 0x81353, 0x39, 0, 0x81354, 0x08, 0, 0x81371, 0x60, 0, 0x81372, 0x08, 0, 0x81373, 0x05, 0, 0x81374, 0xF0, 0, 0x81391, 0x60, 0, 0x81392, 0x08, 0, 0x81393, 0x05, 0, 0x81394, 0xF0, 0, 0x813B1, 0x60, 0, 0x813B2, 0x08, 0, 0x813B3, 0x05, 0, 0x813B4, 0xF0, 0, 0x813D1, 0x60, 0, 0x813D2, 0x08, 0, 0x813D3, 0x05, 0, 0x813D4, 0xF0, 0, 0x813F1, 0x60, 0, 0x813F2, 0x08, 0, 0x813F3, 0x06, 0, 0x813F4, 0xF0, 0, 0x81411, 0x60, 0, 0x81412, 0x08, 0, 0x81413, 0x06, 0, 0x81414, 0xF0, 0, 0x81431, 0x60, 0, 0x81432, 0x08, 0, 0x81433, 0x06, 0, 0x81434, 0xF0, 0, 0x81451, 0x60, 0, 0x81452, 0x08, 0, 0x81453, 0x06, 0, 0x81454, 0xF0, 0, 0x81471, 0x60, 0, 0x81472, 0x08, 0, 0x81473, 0x07, 0, 0x81474, 0xF0, 0, 0x81491, 0x60, 0, 0x81492, 0x08, 0, 0x81493, 0x07, 0, 0x81494, 0xF0, 0, 0x814B1, 0x60, 0, 0x814B2, 0x08, 0, 0x814B3, 0x07, 0, 0x814B4, 0xF0, 0, 0x814D1, 0x60, 0, 0x814D2, 0x08, 0, 0x814D3, 0x07, 0, 0x814D4, 0xF0
I'm doubling to 8192...
-
fba v0.2.97.23 will crash after load a savestate in cps1, neogeo, pgm games.
how to reproduce the bug:
1. save a state in game. (ex: dino, orlegend)
2. close emulator and launch emulator again.
3. load savastate then crash. (if it dosen't crash, try again plz)
-
fba v0.2.97.23 will crash after load a savestate in cps1, neogeo, pgm games.
how to reproduce the bug:
1. save a state in game. (ex: dino, orlegend)
2. close emulator and launch emulator again.
3. load savastate then crash. (if it dosen't crash, try again plz)
Savestates work fine in the current version.
-
Savestates work fine in the current version.
thx for the quick reply. :smilie:
I test it with fba v0.2.97.23 x64 build on os win7 x64, it does crash after loading the savestate.
but the x86 build works fine. :rolleyes:
-
thx for the quick reply. :smilie:
I test it with fba v0.2.97.23 x64 build on os win7 x64, it does crash after loading the savestate.
but the x86 build works fine. :rolleyes:
Sorry, I for some reason I read that as 0.2.95.23. If you created the savestate with the x86 build then it will need to be loaded in the x86 build, they won't be transferrable. Thinking about it we need to make sure that the savestates are flagged as x64/x86 and only load in the correct build, or find a way to make them transferrable (if feasible).
-
Sorry, I for some reason I read that as 0.2.95.23. If you created the savestate with the x86 build then it will need to be loaded in the x86 build, they won't be transferrable. Thinking about it we need to make sure that the savestates are flagged as x64/x86 and only load in the correct build, or find a way to make them transferrable (if feasible).
I know x64/x86 savedatas are not transferrable. I save and load savedatas with the x64 build.
Can you reproduce the bug ?
And sometimes, after you load a savestate, the game speed will become slow.
-
Force load unavailable set, the "Loading" don't kill.
-
Not really a bug... :biggrin:
The year in Copyright.
And in "License": All material not covered above © 2004-2010 Team FB Alpha.
:cool:
-
Not really a bug... :biggrin:
The year in Copyright.
And in "License": All material not covered above © 2004-2010 Team FB Alpha.
:cool:
Thanks for reporting these Bison, I'll look at the unavailable issue soon.
-
Force load unavailable set, the "Loading" don't kill.
Ok, fixed, thanks again for reporting.
-
savestate crash bug is still exist in fba 0.2.97.24.
how to reproduce the bug:
- run fba64.
- save a state in game. (ex: dino, orlegend, samsho2)
- close emulator and launch emulator again.
- load savastate then crash. (if it dosen't crash, try again plz)
another crash bug since the first release of fba64.
after launch fba64, press F6 to open game list dialog will crash the emulator.
click the menu to open game list is ok.
-
cannot go beyond the calendar on ddp2, I get a message "27 EXPIRED XXXXX CMD ID:01 K:01 R:0" with XXXXX being an incremental counter...
maybe I'm not using the right roms...
-
cannot go beyond the calendar on ddp2, I get a message "27 EXPIRED XXXXX CMD ID:01 K:01 R:0" with XXXXX being an incremental counter...
maybe I'm not using the right roms...
your ddp2 rom needs a new file to work correctly i think it's called ddp2_igs00127a_japan.bin
-
savestate crash bug is still exist in fba 0.2.97.24.
how to reproduce the bug:
- run fba64.
- save a state in game. (ex: dino, orlegend, samsho2)
- close emulator and launch emulator again.
- load savastate then crash. (if it dosen't crash, try again plz)
another crash bug since the first release of fba64.
after launch fba64, press F6 to open game list dialog will crash the emulator.
click the menu to open game list is ok.
Savestates work fine for me (using the procedure you posted). And F6 has always worked in x86 and x64 (it's the way I open the dialog).
Try doing a clean install to a new folder and see if that fixes your issues.
-
I updated the iOS version and I've got issues with latest pgm update:
with bee storm (v100, v101, v102) & demonfront v105 I get crash just at first arm7 call in the drvframe function.
Only demonfront v102 (china) works.
in
void Arm7_program_write_byte_32le(UINT32 addr, UINT8 data)
{
#if defined FBA_DEBUG
if (!DebugCPU_ARM7Initted) bprintf(PRINT_ERROR, _T("Arm7_program_write_byte_32le called without init\n"));
#endif
#ifdef DEBUG_LOG
bprintf (PRINT_NORMAL, _T("%5.5x, %2.2x wb\n"), addr, data);
#endif
if (membase[WRITE][addr >> PAGE_SHIFT] != NULL) {
membase[WRITE][addr >> PAGE_SHIFT][addr & PAGE_BYTE_AND] = data;
return;
}
if (pWriteByteHandler) {
pWriteByteHandler(addr, data);
}
}
I got a crash due to wrong addr (addr>>PAGE_SHIFT is way over PAGE_COUNT)
any clue ?
also: I have the same crash with all GCC 4.2 optim turned off, both on x86 (iphone simulator) and arm7 hardware (iphone/ipad)
-
Having same issue with the regular PC version as well.
-
I updated the iOS version and I've got issues with latest pgm update:
with bee storm (v100, v101, v102) & demonfront v105 I get crash just at first arm7 call in the drvframe function.
Only demonfront v102 (china) works.
in
void Arm7_program_write_byte_32le(UINT32 addr, UINT8 data)
{
#if defined FBA_DEBUG
if (!DebugCPU_ARM7Initted) bprintf(PRINT_ERROR, _T("Arm7_program_write_byte_32le called without init\n"));
#endif
#ifdef DEBUG_LOG
bprintf (PRINT_NORMAL, _T("%5.5x, %2.2x wb\n"), addr, data);
#endif
if (membase[WRITE][addr >> PAGE_SHIFT] != NULL) {
membase[WRITE][addr >> PAGE_SHIFT][addr & PAGE_BYTE_AND] = data;
return;
}
if (pWriteByteHandler) {
pWriteByteHandler(addr, data);
}
}
I got a crash due to wrong addr (addr>>PAGE_SHIFT is way over PAGE_COUNT)
any clue ?
also: I have the same crash with all GCC 4.2 optim turned off, both on x86 (iphone simulator) and arm7 hardware (iphone/ipad)
Check the SVN. r472 should fix this.
-
src\burn\drv\cps3\d_cps3.cpp
struct BurnDriver BurnDrvSfiii3u = {
"sfiii3u", "sfiii3", NULL, NULL, "1999",
"Street Fighter III 3rd Strike: Fight for the Future (USA 990608)\0", NULL, "Capcom", "CPS-3",
NULL, NULL, NULL, NULL,
BDF_GAME_WORKING | BDF_16BIT_ONLY, 2, HARDWARE_CAPCOM_CPS3, GBF_VSFIGHT, FBF_SF,
NULL, sfiii3uRomInfo, sfiii3uRomName, NULL, NULL, cps3InputInfo, usaDIPInfo,
sfiii3Init, cps3Exit, cps3Frame, NULL, cps3Scan, &cps3_palette_change, 0x40000,
384, 224, 4, 3
};
sfiii3u must be added "BDF_GAME_WORKING | BDF_16BIT_ONLY | BDF_CLONE"...
-
Is the sound stuttering on Demon Front normal? During continue screen, and after character selection, the audio stutters.
Looking at Haze's video. I'm not getting BGM during the character select screen nor during tutorial screen.
-
for the arm7 crash issue, there is another function to patch:
void Arm7_program_write_dword_32le(UINT32 addr, UINT32 data)
{
#if defined FBA_DEBUG
if (!DebugCPU_ARM7Initted) bprintf(PRINT_ERROR, _T("Arm7_program_write_dword_32le called without init\n"));
#endif
addr &= MAX_MEMORY_AND;
also all the ddp3 (ddpdoj) roms are flagged as GBF_SCRFIGHT instaed of GBF_VERTSHOOT
-
for the arm7 crash issue, there is another function to patch:
void Arm7_program_write_dword_32le(UINT32 addr, UINT32 data)
{
#if defined FBA_DEBUG
if (!DebugCPU_ARM7Initted) bprintf(PRINT_ERROR, _T("Arm7_program_write_dword_32le called without init\n"));
#endif
addr &= MAX_MEMORY_AND;
also all the ddp3 (ddpdoj) roms are flagged as GBF_SCRFIGHT instaed of GBF_VERTSHOOT
Fixed. Thanks. :)
Is the sound stuttering on Demon Front normal? During continue screen, and after character selection, the audio stutters.
Looking at Haze's video. I'm not getting BGM during the character select screen nor during tutorial screen.
Sound in FBA for pgm games is horrible. I need to motivate myself to port MAME's newer ics2115 core, but can't seem to do it.
-
found an issue in pdrift - during gameplay the overhead map radar flickers and disappears. Its not .24 specific but its new-ish
-
found an issue in pdrift - during gameplay the overhead map radar flickers and disappears. Its not .24 specific but its new-ish
It's a timing issue with an IRQ, it's always been a problem, and thus far an actual fix has eluded me.
-
interesting..i swear it never used to happen but of course i cant be certain :)
-
I had a hack that fired the IRQ much earlier on in the frame, but it still had issues with the track display and othe stuff* disappearing. It must have been less pronounced, so I might look at the timing again.
* It's the System16B sprite layer that has the issue.
-
In Shadow Warriors and all of its clones there is sound distortion when hitting a enemy and being hit by a enemy.
This bug has been around for awhile now.
not sure how else to explain it.
-
I think there might be a minor bug in 0.2.97.24, I have the same GUI issue (running XP Home 32-bit, SP3) as mentioned by Beulu on the main http://www.barryharris.me.uk site.
Here's the original comment:
Nice job again ! - I have a small bug in the rom selection screen (on XP SP3). (See n°1) - Could you put back the old way to see how many roms by category and not all the roms ? (or put an option to swith ?) (See n°2) http://alexledieu1.free.fr/fba_bug.png
Also, another possible minor bug or two. When selecting "Genuine" rom's only, Metal Slug 5 (bootleg) for Neo Geo and Pang! 3 (950511 Euro, Bootleg) for CPS1 still show on the game list.
Metal Slug 5 http://tinypic.com/r/1zbrua0/6
Pang! 3 http://tinypic.com/r/15gc6lt/6
-
zip name:
sgyxz -> wofhb
wof3js wof3jsa wof3sj wof3sja wofsj wofsja P3 coin and F2 invalid.
-
zip name:
sgyxz -> wofhb
wof3js wof3jsa wof3sj wof3sja wofsj wofsja P3 coin and F2 invalid.
The zipname will remain consistent with MAME - if/when it changes in MAME it will change in FB Alpha.
I know about the P3 coin and F2 issues with those sets. Haven't found where/if the inputs get read yet.
-
mistake
the real is:
Dinosaur Hunters, strange noises
probably for the preliminar sound??
-
mistake
the real is:
Dinosaur Hunters, strange noises
probably for the preliminar sound??
It's a bootleg with poor quality samples.
-
My roms folder is a wreck, but still, FBA gets halfway through scanning and crashes.
Problem signature:
Problem Event Name: APPCRASH
Application Name: fba.exe
Application Version: 0.2.97.25
Application Timestamp: 4f9aec8e
Fault Module Name: StackHash_2685
Fault Module Version: 6.1.7601.17725
Fault Module Timestamp: 4ec49b8f
Exception Code: c0000374
Exception Offset: 000ce6c3
OS Version: 6.1.7601.2.1.0.256.1
Locale ID: 1033
Additional Information 1: 2685
Additional Information 2: 2685a4fc63a3c2f9d1dc2d9c7773ce9a
Additional Information 3: 0cc3
Additional Information 4: 0cc321a6721f3866b13ebd5587ba696e
-
Hi guys,
noticed a bug in FBA 0.2.97.25:
PGM - Demonfront: Start emulation, don´t coin up, let the intro play. On some parts the music speeds up.
jb
EDIT: Removed Neo-Geo report
-
O.k. after some testing reposting second noticed bug again:
Neo-Geo: Stumbled over this: http://www.pouet.net/prod.php?which=59253 -> Let the demo run (cart or cd) after some (random) time the music will "hang".
This works in MAME (except a problem? with some samples ).
-
IPS Configuration bug:
When I config the game ips, sometimes Fba create ini files of other game in directory config/ips after select ips or click deselect all.
About Load Game Dialogs:
I have adjusted load game dialog, bug when I deselect filters or change options, It changes small auto after I open load game dialogs again.
-
The hardware tab stays with unchecked box while every submenu box is checked. Kinda weird thing.
A question here! When I ran for first time FBA it want to scan for roms, it takes damn too long, the scanning usually takes too long. I don't know why that happens. Is there anyway to improve this?
So... I'm totally not sure if this suggestions may come in help as far as I don't know how the scan
works actually.
When I got 10 games and add one more to the directory it doesnt need to scan every hardware, it should scan only under checked hardware in the list and skip the rest. I mean I check pgm but uncheck everything else... so the scan will just skip everything and check only pgm system then.
Also another thing.
For example if I don't have neogeo.zip or pgm.zip the emulator can check first for bioses and exclude the hardware with missing bioses, just will skip that neogeo or pgm roms from the scan, theres no point to scan for roms when you got your bioses missing. Or well then theres can be made option to force scan even if bios is missing so to see if the roms are correct.
-
Macro input:
Some neogeo games need to use macro input. eg. King of Fighters, Street Fighter, Samurai Shodown and so on. But macro input is supported by neogeo.
Name isn't usd utf8:
Turma da M鬾ica na Terra dos Monstros (Bra) -> Turma da Mônica na Terra dos Monstros (Bra)
It is wrong that samsho5, samsho5b, samsho5h year display 2003. 2004 is correct.
-
Penguin Bros does not load up and has a Guru Meditation Error #c00000005.01671f4d
-
Macro input:
Some neogeo games need to use macro input. eg. King of Fighters, Street Fighter, Samurai Shodown and so on. But macro input is supported by neogeo.
Name isn't usd utf8:
Turma da M鬾ica na Terra dos Monstros (Bra) -> Turma da Mônica na Terra dos Monstros (Bra)
It is wrong that samsho5, samsho5b, samsho5h year display 2003. 2004 is correct.
Hi money_114,
according to neogeomuseum.snkplaymore.co.jp (http://neogeomuseum.snkplaymore.co.jp/english/catalogue/index.php) :
SAMURAI SHODOWN V Fighting SNK P/Yuki Ent MVS Cartridge:2003/10/10
NEOGEO ROM-cart:2003/12/11
SAMURAI SHODOWN V SPECIAL Fighting SNK P/Yuki Ent MVS Cartridge:2004/04/22
NEOGEO ROM-cart:2004/07/15
-
Hi money_114,
according to neogeomuseum.snkplaymore.co.jp (http://neogeomuseum.snkplaymore.co.jp/english/catalogue/index.php) :
SAMURAI SHODOWN V Fighting SNK P/Yuki Ent MVS Cartridge:2003/10/10
NEOGEO ROM-cart:2003/12/11
SAMURAI SHODOWN V SPECIAL Fighting SNK P/Yuki Ent MVS Cartridge:2004/04/22
NEOGEO ROM-cart:2004/07/15
Sorry, it's my mistake. I mean that year wrong is samsh5sp samsh5sph samsh5spho.
Thanks for your correction.
-
Thanks money_114 :smilie:
Effectively year for samsh5sp and clones is wrong (2003 instead of 2004)...Will be fixed in next release :wink:
thanks for reporting bugs...Looking actually utf8 problem with Turma da Mônica na Terra dos Monstros (Bra).
best regards,
JacKc.
-
When saving input preset for neogeo games, neogeo.ini is not created anymore but preset.ini.
-
kov&kovplus and clone bug:
when play to stage 6, the game will skip back coin screen after character go to next level from snow level.
-
Wrote a driver for Pac-Land. This game is not good. lol
some slight issues with this game certain sound fx and in level game music is missing any chance this could be sorted
for pacland fans like me :redface:
-
cannot go beyond the calendar on ddp2, I get a message "27 EXPIRED XXXXX CMD ID:01 K:01 R:0" with XXXXX being an incremental counter...
It also happens on real hardware, what exactly causes this error?
Thanks in advance for any help.
-
some slight issues with this game certain sound fx and in level game music is missing any chance this could be sorted
for pacland fans like me :redface:
I'll take a look. I'm really bad with sound though. I'm a bit tone deaf and have little attention for detail (this is why beta testers like JackC, for example, are a huge asset to the FBA project).
It also happens on real hardware, what exactly causes this error?
Thanks in advance for any help.
That's related to the protection device on the cart. "27" refers to ASIC27A -- the ARM7 cpu on the cartridge. What it means is that the 68k and arm7 aren't communicating properly.
If this happens on the cart normally that means igs did some shitty bug testing or that something is off with your cart. lol
-
That's related to the protection device on the cart. "27" refers to ASIC27A -- the ARM7 cpu on the cartridge. What it means is that the 68k and arm7 aren't communicating properly.
If this happens on the cart normally that means igs did some shitty bug testing or that something is off with your cart. lol
I figured it out earlier today. The oscillator that drives the onboard 68k is also used for the ARM. From the oscillator the signal is split then goes to a 22ohm resistor then to the ARM clock in. On my board the 22ohm was reading at 330ohm so I replaced it and the error is gone.
-
I'll take a look. I'm really bad with sound though. I'm a bit tone deaf and have little attention for detail (this is why beta testers like JackC, for example, are a huge asset to the FBA project).
well thx for just taking a look the music is kinda iconic in this game
-
well thx for just taking a look the music is kinda iconic in this game
While I was off doing useless things, Barry took a look and fixed this issue. :)
-
well thankyou very much barry harris for taking the time to sort the issues in the pac-land driver
-
well thankyou very much barry harris for taking the time to sort the issues in the pac-land driver
The main issue was that we were only using 3 of the sound channels. Pac Land (and Baraduke and Sky Kid) use 8 channels.
-
kov&kovplus and clone bug:
when play to stage 6, the game will skip back coin screen after character go to next level from snow level.
Thanks money_114 for the report. This bug has been confirmed...We are still tracking this bug and try to find a solution soon.
-
In the lastest versions I can´t do nothing with neogeo memory cards, the emulator don´t let me create or insert.In old versions I have not this problem, all works perfect, also i try lo load memory cards from ol versions but i can´t.Is this a bug???I am doing something bad??.For example, I load "art of fighting" then go to memory card menu and all is not markable.Don´t let me create a new memory card or insert one.
Regards,
-
In the lastest versions I can´t do nothing with neogeo memory cards, the emulator don´t let me create or insert.In old versions I have not this problem, all works perfect, also i try lo load memory cards from ol versions but i can´t.Is this a bug???I am doing something bad??.For example, I load "art of fighting" then go to memory card menu and all is not markable.Don´t let me create a new memory card or insert one.
Regards,
menu.cpp
if ((BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_SNK_NEOGEO
|| (BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_PREFIX_CARTRIDGE
|| (BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_SNK_DEDICATED_PCB)
{
EnableMenuItem(hMenu, MENU_INTERPOLATE_1, MF_GRAYED | MF_BYCOMMAND);
EnableMenuItem(hMenu, MENU_INTERPOLATE_3, MF_GRAYED | MF_BYCOMMAND);
if (!kNetGame) {
EnableMenuItem(hMenu, MENU_MEMCARD_CREATE, MF_ENABLED | MF_BYCOMMAND);
EnableMenuItem(hMenu, MENU_MEMCARD_SELECT, MF_ENABLED | MF_BYCOMMAND);
if (nMemoryCardStatus & 1) {
if (nMemoryCardStatus & 2) {
EnableMenuItem(hMenu, MENU_MEMCARD_EJECT, MF_ENABLED | MF_BYCOMMAND);
} else {
EnableMenuItem(hMenu, MENU_MEMCARD_INSERT, MF_ENABLED | MF_BYCOMMAND);
}
}
}
}
scrn.cpp
case MENU_MEMCARD_CREATE:
if (bDrvOkay && UseDialogs() && !kNetGame && ((BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_SNK_NEOGEO
|| (BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_PREFIX_CARTRIDGE
|| (BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_SNK_DEDICATED_PCB))
{
InputSetCooperativeLevel(false, bAlwaysProcessKeyboardInput);
AudBlankSound();
MemCardEject();
MemCardCreate();
MemCardInsert();
GameInpCheckMouse();
}
break;
case MENU_MEMCARD_SELECT:
if (bDrvOkay && UseDialogs() && !kNetGame && ((BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_SNK_NEOGEO
|| (BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_PREFIX_CARTRIDGE
|| (BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_SNK_DEDICATED_PCB))
{
InputSetCooperativeLevel(false, bAlwaysProcessKeyboardInput);
AudBlankSound();
MemCardEject();
MemCardSelect();
MemCardInsert();
GameInpCheckMouse();
}
break;
case MENU_MEMCARD_INSERT:
if (bDrvOkay && !kNetGame && ((BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_SNK_NEOGEO
|| (BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_PREFIX_CARTRIDGE
|| (BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_SNK_DEDICATED_PCB))
{
MemCardInsert();
}
break;
case MENU_MEMCARD_EJECT:
if (bDrvOkay && !kNetGame && ((BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_SNK_NEOGEO
|| (BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_PREFIX_CARTRIDGE
|| (BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_SNK_DEDICATED_PCB))
{
MemCardEject();
}
break;
case MENU_MEMCARD_TOGGLE:
if (bDrvOkay && !kNetGame && ((BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_SNK_NEOGEO
|| (BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_PREFIX_CARTRIDGE
|| (BurnDrvGetHardwareCode() & HARDWARE_PUBLIC_MASK) == HARDWARE_SNK_DEDICATED_PCB))
{
MemCardToggle();
}
break;
(https://lh4.googleusercontent.com/-iv0eUYlLM6I/UBPKGAWwBNI/AAAAAAAAAHE/3J7LZDWrNXg/s656/fbap_memory_card.png)
(https://lh3.googleusercontent.com/--MyKDmSi80I/UBPKGBVzxgI/AAAAAAAAAHM/b-mBoWFaqrw/s656/fbap_data_save.png)
download: http://www.sendspace.com/file/s4zvw2
-
Custom macros not work for "street fighters" series.
I tested from 0.2.96.99~0.2.97.26
Custom macros never works on any of them.
Custom macros works for neo geo sets and others,but not for those with 3x macros buildin by fba.
Also I can only define only one custom macro.
-
// --- Inputs -----------------------------------------------------------------
input "P1 Coin" switch 0x06
input "P1 Start" switch 0x02
input "P1 Up" switch 0xC8
input "P1 Down" switch 0xD0
input "P1 Left" switch 0xCB
input "P1 Right" switch 0xCD
input "P1 Weak Punch" switch 0x1E
input "P1 Medium Punch" switch 0x1F
input "P1 Strong Punch" switch 0x20
input "P1 Weak Kick" switch 0x2C
input "P1 Medium Kick" switch 0x2D
input "P1 Strong Kick" switch 0x2E
input "P2 Coin" switch 0x07
input "P2 Start" switch 0x03
input "P2 Up" switch 0x4002
input "P2 Down" switch 0x4003
input "P2 Left" switch 0x4000
input "P2 Right" switch 0x4001
input "P2 Weak Punch" switch 0x4080
input "P2 Medium Punch" switch 0x4081
input "P2 Strong Punch" switch 0x4082
input "P2 Weak Kick" switch 0x4083
input "P2 Medium Kick" switch 0x4084
input "P2 Strong Kick" switch 0x4085
input "Reset" switch 0x3D
input "Diagnostic" switch 0x3C
input "Service" switch 0x0A
macro "P1 3x Punch" switch 0x21
macro "P1 3x Kick" switch 0x2F
macro "P2 3x Punch" undefined
macro "P2 3x Kick" undefined
custom "P1 Weak Punch + Weak Kick" switch 0x1E "P1 Weak Punch" 0x01 "P1 Weak Kick" 0x01
custom "P1 Medium Punch + Medium Kick" undefined "P1 Medium Punch" 0x01 "P1 Medium Kick" 0x01
custom "P1 Strong Punch + Strong Kick" undefined "P1 Strong Punch" 0x01 "P1 Strong Kick" 0x01
-
doomking,can you show me your all region settings of windows control panel?
I found it makes different with fba custom macros.
For enus,i can only define one custom macro for street fighter series
For zhcn,Just as I said above.
I need all your region settings,Thanks.
Istill think this is a bug as we may have different region settings.
-
sedusg, 沒有windows的問題
custom macros是官方fba還未修正問題
-
中文交流還是直接方便些啊。
FB Alpha Plus好久没有更新了吧?
doomking你是自行编译的吗?有没有一直以来的旧版本的啊?
Anyway,this has been comfirmed to be a bug of FB Alpha.
-
Red Earth and clones are not playable..
-
No problem here with redearth and clones...Are you sure your ROMset is up to date with FBA 0.2.97.26?
-
Ooops sorry.
my game config was too old :p
it work very well now, thanks Jack
-
sedusg, 我知你是ppxclub的會員
現在官方FB Alpha Plus!沒有更新
-
No problem here with redearth and clones...Are you sure your ROMset is up to date with FBA 0.2.97.26?
What did you use to make your roms compatible with 02.97.26 ? i can't seem to get Neogeo roms running , everything else runs fine CP1-CPS2-CPS3 etc.
-
there's usually a dat file that comes with each FBA release use the dat with romcentre and scan all your roms
romcentre will fix em for ya and make them compatable with FBA 0.2.97.26
-
{ "242-v5ae.bin", 0x400000, 0xafdd9660, 5 | BRF_SND }, // 20
KOF98AE sound is bad.
Maybe v5ae is not supported.
-
on 0.2.97.27 (and I think others) The sound on Gunsmoke is incorrect, and no sound FX :( working great on MAME
-
Real Bout Fatal Fury Special with AES bios doesn't work. It just shows a black screen.
-
- The Irritating Maze only shows black screen.
- X-Men crash after the big white boss (at the beginning of stage 3).
-
Capcom Sport's Club (All Variants - CPS2 System) sound seems abnormally quieter compared to other games.
-
Made a start on palette fade. Verified palette control register for all CPS-1/2 games (including non-MAME) and added config mapping for it. Will hook up the actual palette calculation tomorrow hopefully.
This sounds quite promising. I'm wondering if there will be any way to sort out some nasty graphical bugs in certain CPS2 games. Sadly, to many huge CPS2 fans, the games affected by the unemulated palette fade effect just so happen to be 2D works of art. :redface:
In the link below exists a thread I created which touches on what these graphical issues look like:
http://neo-source.com/index.php?topic=1695.0
Words can't express how amazing FBA is thanks to the extremely talented and generous team behind it. It's truly remarkable.
-
kov&kovplus and clone bug:
when play to stage 6, the game will skip back coin screen after character go to next level from snow level.
Thanks goes to iq_132, finally kov & kovplus and clone bug is now fixed :smilie:
-
Custom macros not work for "street fighters" series.
I tested from 0.2.96.99~0.2.97.26
Custom macros never works on any of them.
Custom macros works for neo geo sets and others,but not for those with 3x macros buildin by fba.
Also I can only define only one custom macro.
I'm still having problems with Macros/Custom keys, even with 0.2.97.27.
I tried to create two custom input keys for Samurai Shodown IV (samsho4.ini)
This is what I added below "input" lines:
custom "P1 ABC" switch 0x10 "P1 Button A" 0x01 "P1 Button B" 0x01 "P1 Button C" 0x01
custom "P1 BCD" switch 0x11 "P1 Button B" 0x01 "P1 Button C" 0x01 "P1 Button D" 0x01
What's expected?1. I should be able to see custom input names.
2. I should be able to see these 2 custom inputs I added.
3. Custom input keys should work
What I see:1. Custom key for "P1 ABC" didn't appear, which means that emulator recognized the second custom key only (for "P1 BCD").
2. Custom input name for "P1 BCD" can't be seen, but letter "W" (which is switch 0x11) appears next to name space. Althought, this custom key is working as expected.
3. After closing emulator, and samsho4.ini is opened, I only see the key I made for "P1 BCD" and without a name.
custom "" switch 0x11 "P1 Button B" 0x01 "P1 Button C" 0x01 "P1 Button D" 0x01
Is there any way to fix this? What Am I doing wrong?
Or, better... could you add, Barry, at least 3-Button keys such as "ABC" and "BCD" for Neo-Geo games, please?
I'm not asking for 2-Button keys like "AB", "CD" since they aren't needed, and could be help for "unfair gaming".
The thing is, in my case... my keyboard, as in the most of the cases, just doesn't allow user to press 3 buttons at the same time.
So that's the only limitation I have at playing this game in particular.
Thanks in advance.
-
error loading replays in kof2001.zip
the strange thing is: replays load fine, but when I restart PC I got guru meditation.... =[
x64 version
Error log file attached
-
The Irritating Maze (Neo-Geo) still hangs on a black screen upon loading. Using version FB Alpha 0.2.97.28 (XP Binary) OS:XP SP3.
-
Strikers 1945 Plus (Neo Geo) lags pretty hard most of the time when it shouldn't be in comparison to the real thing (on the actual MVS/AES, slowdowns are minimal). Setting the CPU setting higher seems to fix it, but is not ideal if you want speed closer to its true form.
Iron Clad (Neo Geo) has issues where playing on European settings causes the screen to black out. Normally on MAME and presumably on real hardware, this would only cause the screen to dim a little.
Ninja Commando (Neo Geo) has an issue where the screen blacks out in the Prehistoric stage when you fight the first mid-boss. Behavior does not match how it should be running on MAME and a real Neo-Geo system.
Truxton 2 has some desync issues in recording playbacks, and varius other Toaplan games seem to have issues with dipswitch changes (for example, you cannot change the region in Batsugun and Dogyuun).
For Batsugun, voices on the character select screen are very high-pitched. Applies to Special as well.
-
Displays the following error when I run full screen.
Anyone know a solution? Previous versions did not have this error, and this ever happens with any filter.
(http://img705.imageshack.us/img705/6306/fba.png)
(http://img4.imageshack.us/img4/4402/fba2.png)
-
I've just encountered a regression in Gunbird 1 on FBA 0.2.97.28.
This happens when you play as Marion and when you press and hold the fire button and then release.
The 'attack' animation will become 'stuck' to the screen and there's a very annoying sound pop/bug that will not go away.
I can confirm that this didn't happen with previous versions (I can be certain about FBA 0.2.97.26 at the very least). I tested with both the Win32 version and my own libretro version.
Here's a screenshot of what the bug looks like -
http://i.imgur.com/ahIFe.png
note how all the 'sprites' stick glued to the screen and never disappear (the attack sprite is cut in half as well). You can't hear the bad sound it makes either.
-
I've just encountered a regression in Gunbird 1 on FBA 0.2.97.28.
This happens when you play as Marion and when you press and hold the fire button and then release.
The 'attack' animation will become 'stuck' to the screen and there's a very annoying sound pop/bug that will not go away.
I can confirm that this didn't happen with previous versions (I can be certain about FBA 0.2.97.26 at the very least). I tested with both the Win32 version and my own libretro version.
Here's a screenshot of what the bug looks like -
http://i.imgur.com/ahIFe.png
note how all the 'sprites' stick glued to the screen and never disappear (the attack sprite is cut in half as well). You can't hear the bad sound it makes either.
It's related to changeset 891. Looks like the opcode hack should only be applied to 68000 and 68008 only? Any thoughts iq?
-
It's related to changeset 891. Looks like the opcode hack should only be applied to 68000 and 68008 only? Any thoughts iq?
Yeah, I didn't think about it affecting the 32-bit variants. I'll have to make some modifications to that -- I wonder if that is affecting my taito_f3 driver as well? Hmm...
-
The bug I've mentioned about the screen completely blacking out in some Neo Geo games seems to affect The Last Blade 2 when pausing with AES mode as well. In this case, the screen is only supposed to dim a little when paused.
Additional note, these bugs were found in the 32-bit version, have not tried the 64-bit version yet.
-
Sangokushi II: San Jian Sheng miss button C.
C button have fire effect.
-
Hi,
if I try to set some BLitter options, for example Soft FX and 2Xsai, I have this bug message:
How can I solve? I have this problem with others blitter setting :(
FB Alpha v0.2.97.12 fatal exception report (Tue Jan 01 23:52:37 2013
)
----------------------------------------------------------------------
Exception 0xC0000005 (EXCEPTION_ACCESS_VIOLATION) thrown.
EIP: 0x0136
(attempting to write address 0x07818000)
EAX: 0x037A9298, EBX: 0x00000260, ECX: 0x00000000, EDX: 0x07818000
ESI: 0x00000A00, EDI: 0x037A94F8, ESP: 0x0040D384, EBP: 0x0040D3AC
----------------------------------------------------------------------
System information:
OS: Microsoft Windows 7 Service Pack 1 (build 7601)
CPU: GenuineIntel, P6 family
2050 MHz, MMX, SSE, SSE2 (8 system processors)
Physical RAM: 4189688 KB (4091 MB) total, 2283452 KB (2229 MB) avail
Total RAM: 4194303 KB (4095 MB) total, 4194303 KB (4095 MB) avail
FB Alpha: 20880 KB in use (23528 KB peak, 46296 KB virtual)
Installed displays and display adapters:
@oem21.inf,%samsung%;Samsung @oem21.inf,%smp2450hd%;SyncMaster P24
Display size ~53x30cm, Gamma 2,20
Preferred mode 1920x1080, -1,#IO Hz (530x300mm, 148,500 MHz)
Max. bandwidth 170 MHz, H sync 30-81 KHz, V sync 56-60 Hz
----------------------------------------------------------------------
FB Alpha information:
Built on Oct 15 2011, 08:49:11, using Visual C++ 10.0.
Optimised for i686 CPUs.
Using Unicode for all text.
Debug functionality absent.
MMX optimisations enabled.
A68K emulation core enabled for MC68000 emulation.
Musashi emulation core enabled for MC68010/MC68EC020 emulation.
Not emulating any game.
Video settings:
Selected module: DirectDraw7 Software Effects video output
Interface settings: Running in windowed mode, $ix1216, 856bpp
Source image 304x224, 16bpp
Module settings: Using Blt() to transfer the image
Prescaling using 2xSaI (2x zoom)
Using video memory for effect buffers
Audio settings:
Selected module: DirectSound3 audio output
Interface settings: Playback at 22050Hz, 100% volume
Module settings: Audio is delayed by approx. 83ms
Input settings:
Selected module: DirectInput8 input
Interface settings: keyboard 0 System keyboard: Tastiera
mouse 0 System mouse: Mouse
joystick 0 Logitech Cordless RumblePad 2 USB
joystick 1 QanBa Joystick Plus
Profiling settings:
Interface settings: Profiling module not initialised
----------------------------------------------------------------------
Detailed process information about fba.exe:
fba.exe (base address 0x00E50000, size 17912 KB)
Exception occurred in module fba.exe:
fba.exe (base address 0x00E50000, size 17912 KB)
Modules loaded by fba.exe:
ntdll.dll (base address 0x77180000, size 1536 KB)
kernel32.dll (base address 0x74970000, size 1088 KB)
KERNELBASE.dll (base address 0x765D0000, size 284 KB)
USER32.dll (base address 0x767B0000, size 1024 KB)
GDI32.dll (base address 0x75CD0000, size 576 KB)
LPK.dll (base address 0x75EE0000, size 40 KB)
USP10.dll (base address 0x74C80000, size 628 KB)
msvcrt.dll (base address 0x76030000, size 688 KB)
ADVAPI32.dll (base address 0x748C0000, size 640 KB)
sechost.dll (base address 0x75AC0000, size 100 KB)
RPCRT4.dll (base address 0x76370000, size 960 KB)
SspiCli.dll (base address 0x74860000, size 384 KB)
CRYPTBASE.dll (base address 0x74850000, size 48 KB)
COMDLG32.dll (base address 0x75FB0000, size 492 KB)
SHLWAPI.dll (base address 0x74A80000, size 348 KB)
COMCTL32.dll (base address 0x6F310000, size 1656 KB)
SHELL32.dll (base address 0x74D20000, size 12584 KB)
WINMM.dll (base address 0x70D70000, size 200 KB)
ole32.dll (base address 0x76620000, size 1392 KB)
d3d9.dll (base address 0x683D0000, size 1804 KB)
VERSION.dll (base address 0x73150000, size 36 KB)
d3d8thk.dll (base address 0x6F500000, size 24 KB)
dwmapi.dll (base address 0x6E480000, size 76 KB)
d3dx9_43.dll (base address 0x67780000, size 2044 KB)
DDRAW.dll (base address 0x69BE0000, size 924 KB)
DCIMAN32.dll (base address 0x70430000, size 24 KB)
SETUPAPI.dll (base address 0x74AE0000, size 1652 KB)
CFGMGR32.dll (base address 0x75A00000, size 156 KB)
OLEAUT32.dll (base address 0x75A30000, size 572 KB)
DEVOBJ.dll (base address 0x76190000, size 72 KB)
DSOUND.dll (base address 0x6A490000, size 456 KB)
POWRPROF.dll (base address 0x6C0F0000, size 148 KB)
DINPUT8.dll (base address 0x69FA0000, size 192 KB)
IMM32.DLL (base address 0x760E0000, size 384 KB)
MSCTF.dll (base address 0x75AE0000, size 816 KB)
uxtheme.dll (base address 0x6EC80000, size 512 KB)
HID.DLL (base address 0x738A0000, size 36 KB)
WINTRUST.dll (base address 0x75F80000, size 180 KB)
CRYPT32.dll (base address 0x764B0000, size 1144 KB)
MSASN1.dll (base address 0x76140000, size 48 KB)
CLBCatQ.DLL (base address 0x75970000, size 524 KB)
MMDevApi.dll (base address 0x69A80000, size 228 KB)
PROPSYS.dll (base address 0x6F200000, size 980 KB)
AUDIOSES.DLL (base address 0x67230000, size 216 KB)
atiu9pag.dll (base address 0x6A470000, size 100 KB)
idmmkb.dll (base address 0x00E40000, size 32 KB)
WmJoyF32.dll (base address 0x10000000, size 260 KB)
DINPUT.dll (base address 0x6A510000, size 148 KB)
aticfx32.dll (base address 0x64B10000, size 932 KB)
atiumdag.dll (base address 0x67D80000, size 5452 KB)
atiumdva.dll (base address 0x685A0000, size 2672 KB)
----------------------------------------------------------------------
-
Sangokushi II: San Jian Sheng miss button C.
C button have fire effect.
Thanks - I added it the current tree.
-
varios chinese bootlegs from Sangokushi II has is function too
-
Hello and congratulations for FB alpha.
i'd like to report a very boring bug.
i currently use FBA x64 0.2.97.28. i generated the DATS for arcade / md / pce / tb16 / sgx
i completed the romsets thx to CMPRO. all is complete. BUT when i'm scanning again the paths with FBA, there are some missing roms again, in MD sets & PCE sets.
i followed the councils : to put MD in path 10, PCE in path 11 & so on. i put Arcade in the 20th path.
i noticed FBA doesn't ask anymore MD missing roms (?) but always PCE.
i can take example for tvbasket (usa).pce. it isn't in the generated DAT but FBA asks for it again & again when i finished to Rescan all folders.
i put some screenshots in order you understand. http://ge.tt/4c4fwdU (http://ge.tt/4c4fwdU) (picture 1 to 4)
thank you a lot. i hope it will be corrected in the next release.
To be simple : FBA doesn't generated correcty the DATs & all the roms it needs. (especially MD & PCE)
keep going !^^
-
Hi greatexerox, I sedn you a PM :)
-
Hello and congratulations for FB alpha.
i'd like to report a very boring bug.
i currently use FBA x64 0.2.97.28. i generated the DATS for arcade / md / pce / tb16 / sgx
i completed the romsets thx to CMPRO. all is complete. BUT when i'm scanning again the paths with FBA, there are some missing roms again, in MD sets & PCE sets.
i followed the councils : to put MD in path 10, PCE in path 11 & so on. i put Arcade in the 20th path.
i noticed FBA doesn't ask anymore MD missing roms (?) but always PCE.
i can take example for tvbasket (usa).pce. it isn't in the generated DAT but FBA asks for it again & again when i finished to Rescan all folders.
i put some screenshots in order you understand. http://ge.tt/4c4fwdU (http://ge.tt/4c4fwdU) (picture 1 to 4)
thank you a lot. i hope it will be corrected in the next release.
To be simple : FBA doesn't generated correcty the DATs & all the roms it needs. (especially MD & PCE)
keep going !^^
The supplementary and console sets should be in paths 11-20 as per the documentation. You should move your Megadrive path from path 10.
As for tvbasket (usa).pce, there is no rom with that name in FB Alpha anywhere.
-
first path:
arcade
i moved like you said
MD >> path 11
& others in path 12 13 14
and the 224 "missing" roms disapeared. now all seems completed 5740/5740
i think the problem was i put MD in path 10 & others 11 12 13
that's a really strange bug than FBA asks for some pce games for others regions. (1943kai (japan).pce, tv basket (usa).pce ...) if we don't put the others systems in the correct path. that's crazy.
anyway now it's ok. maybe these bugs will be corrected in the future & we can choose every paths we want.
i thank you a lot, Treble winner !
-
Again, to reiterate, FBA will never ask you for roms with those names, because they are not in any of the driver rom definitions.
-
i don't understand why i got this messages. anyway thank you again !
-
Still finding an issue with battle shark and all clones
(http://bisonsas.googlepages.com/bshark-04-10-052412.png)
Seems to freeze at this point unless you skip it and start level 2 but then it will randomly lockup on any level after that
i know you guys are busy and i hate to be a pest :) but any chance someone could take a wee look at this see if you can spot
the bug thats causing this
-
hi guys
I do not speak very good English , So excuse me
I found a bug in pulstar , I put the video on youtube
there is a problem of music 3.40 minutes
https://www.youtube.com/watch?v=9SER5vPz93k
thanks
-
Hi!
language does not change.
I do: Misc >> UI language >> Select UI language >> Select Francais and nothing... Continue in English .
Thnx
-
Finding an issue:
If hardware is checked in game selection dialog, Fba can't removed check.
-
A strange, old bug:
Megadrive driver - Streets of Rage II
The Clones seems fine but the parent has this strange behaviour: when you start beating the baddies the game starts freezing for fractions of second continously.
Thank you
-
Operation Wolf/Bear (opwolf, opwolfb, opwolfj, opwolfu, opwolfa) - Problem with sound - bad crashing sound
-
Dynasty Wars (US Set 1) (dynwar) - Problem with Load State - game running too slow from load state.
-
Turbo Force (turbofrc) - Graphic Glitch on Stage 3 - alien form should be on under not above the background.
-
Super Chase- Criminal Termination (superchs superchsj, superchsu) - No sound, problem with graphic- flashing road.
-
Bug in konami driver for xmen and clones :
- Guru meditation on tilte screen
fba.error.log
FB Alpha v0.2.97.29 fatal exception report (Wed Jul 10 20:59:56 2013
)
----------------------------------------------------------------------
Exception 0xC0000005 (EXCEPTION_ACCESS_VIOLATION) thrown.
EIP: 0x00DD
(attempting to read address 0x0B4CFFF1)
EAX: 0x00000000, EBX: 0x020656ED, ECX: 0x0001FFFD, EDX: 0x0021FFD1
ESI: 0xFFFFFF00, EDI: 0x00008000, ESP: 0x0027CF68, EBP: 0x0B2B0020
----------------------------------------------------------------------
System information:
OS: Microsoft Windows 7 Service Pack 1 (build 7601)
CPU: GenuineIntel, Celeron "Covington" / Pentium II "Deschutes"
2926 MHz, MMX, SSE, SSE2 (4 system processors)
Physical RAM: 4054112 KB (3959 MB) total, 2967220 KB (2897 MB) avail
Total RAM: 4194303 KB (4095 MB) total, 4194303 KB (4095 MB) avail
FB Alpha: 51260 KB in use (80676 KB peak, 86120 KB virtual)
Installed displays and display adapters:
@oem18.inf,%samsung%;Samsung @oem18.inf,%smp2270ha%;SyncMaster P22
Display size ~48x27cm, Gamma 2,20
Preferred mode 1920x1080, 60,000 Hz (510x287mm, 148,500 MHz)
Max. bandwidth 170 MHz, H sync 30-75 KHz, V sync 56-61 Hz
----------------------------------------------------------------------
FB Alpha information:
Built on Jun 29 2013, 09:41:35, using Visual C++ 2012.
Optimised for X86 CPUs.
Using Unicode for all text.
Debug functionality absent.
MMX optimisations enabled.
A68K emulation core enabled for MC68000 emulation.
Musashi emulation core enabled for MC68010/MC68EC020 emulation.
Emulating xmen2pa (X-Men (2 Players ver AAA))
Vertical refresh is 60,00 Hz.
CPU running at 100% of normal frequency.
Video settings:
Selected module: DirectDraw7 video output
Interface settings: Running in windowed mode, $ix1240, 874bpp
Source image 288x224, 32bpp
Module settings: Using Blt() to transfer the image
Using video memory for the final blit
Audio settings:
Selected module: DirectSound3 audio output
Interface settings: Playback at 22050Hz, 100% volume
Module settings: Audio is delayed by approx. 83ms
Input settings:
Selected module: DirectInput8 input
Interface settings: keyboard 0 System keyboard: Clavier
mouse 0 System mouse: Souris
joystick 0 Controller (Gamepad F310)
Profiling settings:
Interface settings: Profiling module not initialised
----------------------------------------------------------------------
Detailed process information about fba.exe:
fba.exe (base address 0x00A10000, size 26696 KB)
Exception occurred in module fba.exe:
fba.exe (base address 0x00A10000, size 26696 KB)
Modules loaded by fba.exe:
ntdll.dll (base address 0x77520000, size 1536 KB)
kernel32.dll (base address 0x767B0000, size 1088 KB)
KERNELBASE.dll (base address 0x75000000, size 284 KB)
USER32.dll (base address 0x76F40000, size 1024 KB)
GDI32.dll (base address 0x76CB0000, size 576 KB)
LPK.dll (base address 0x774F0000, size 40 KB)
USP10.dll (base address 0x76EA0000, size 628 KB)
msvcrt.dll (base address 0x76240000, size 688 KB)
ADVAPI32.dll (base address 0x77040000, size 640 KB)
sechost.dll (base address 0x76E80000, size 100 KB)
RPCRT4.dll (base address 0x76390000, size 960 KB)
SspiCli.dll (base address 0x74F70000, size 384 KB)
CRYPTBASE.dll (base address 0x74F60000, size 48 KB)
COMDLG32.dll (base address 0x76AD0000, size 492 KB)
SHLWAPI.dll (base address 0x76580000, size 348 KB)
COMCTL32.dll (base address 0x710A0000, size 1656 KB)
SHELL32.dll (base address 0x751A0000, size 12584 KB)
WINMM.dll (base address 0x70B70000, size 200 KB)
ole32.dll (base address 0x760D0000, size 1392 KB)
IMM32.DLL (base address 0x76070000, size 384 KB)
MSCTF.dll (base address 0x76B50000, size 816 KB)
dsound.dll (base address 0x73240000, size 456 KB)
POWRPROF.dll (base address 0x6CA50000, size 148 KB)
SETUPAPI.dll (base address 0x768C0000, size 1652 KB)
CFGMGR32.dll (base address 0x74FD0000, size 156 KB)
OLEAUT32.dll (base address 0x762F0000, size 572 KB)
DEVOBJ.dll (base address 0x76000000, size 72 KB)
dinput8.dll (base address 0x71E80000, size 192 KB)
ddraw.dll (base address 0x71A20000, size 924 KB)
DCIMAN32.dll (base address 0x72260000, size 24 KB)
dwmapi.dll (base address 0x70AE0000, size 76 KB)
d3d9.dll (base address 0x71850000, size 1804 KB)
VERSION.dll (base address 0x74E90000, size 36 KB)
d3d8thk.dll (base address 0x72250000, size 24 KB)
D3DX9_43.dll (base address 0x69A50000, size 2044 KB)
uxtheme.dll (base address 0x70BB0000, size 512 KB)
HID.DLL (base address 0x72240000, size 36 KB)
WINTRUST.dll (base address 0x76490000, size 180 KB)
CRYPT32.dll (base address 0x75060000, size 1144 KB)
MSASN1.dll (base address 0x75FF0000, size 48 KB)
CLBCatQ.DLL (base address 0x76C20000, size 524 KB)
MMDevApi.dll (base address 0x691F0000, size 228 KB)
PROPSYS.dll (base address 0x708D0000, size 980 KB)
AUDIOSES.DLL (base address 0x691B0000, size 216 KB)
RICHED20.DLL (base address 0x717D0000, size 472 KB)
nvd3dum.dll (base address 0x63110000, size 12152 KB)
PSAPI.DLL (base address 0x75180000, size 20 KB)
avrt.dll (base address 0x71FB0000, size 28 KB)
----------------------------------------------------------------------
-
Robocop 2 (robocop2, robocop2j. robocop2u) - Graphic Glitch - sprites and background on left and right border are cut off. Some sprites have incorrect position.
Boogie Wings/ The Great Ragtime Show (boogwing, boogwinga, ragtime, ragtimea) - graphic glitch on last stage - lamps light should be behind the whell mechanics
-
Ganbare! Marine Kun (Marine 2K0411 JPN) (ganbare) - HOPPER JAM ERROR -message and don't run.
-
Thunder Dragon 2/ Big Bang (tdragon2, tdragon2a, bigbang) - minor graphic glitch on Stage 2 - Tank is above not the under background. (look in the middle center)
Grahic glitches on In the Hunt (World, US) (inthunt, inthuntu) and Kaitei Daisensou (Japan) (kaiteids)
-
Air Duel (Japan) (airduel) Irem 1990 (M72 Hardware) - Problem with sound - sound is not full emulated - lack of background music.
-
Continental Circus (Japan) (contcircj) - Problem with screen display- shaking screen
Three Wonders (wonder 3 910520 etc) (3wonders) - Incorrect colors of background graphic (should be both black)
Wing Shooting Championship V1.00 (wschampb) - ROMS "U06' AND 'U07' are bad
Vendetta (World 2 Players ver. W) (vendetta2p ) -graphic glitches
-
Vendetta (World 2 Players ver. W)(vendetta2p)- graphic glitches
-
Line of Fire / Bakudan Yarou (World, FD1094 317-0136)(loffire) - no sound, bad emulation, unplayable.
-
三国志II: 三國英雄傳 ◦ Sangokushi II: Sanguo YingXiongZhuan (Chinese bootleg set 3, 921005 Asia) [sgyxz]
Game freezes after this screen but music is still playing (see attached files)
-
Graphical bug with KOF2003
-
Only with x64 version.
-
Pausing the game (in AES) results in a black screen (with BG music),
(http://i.imgur.com/hmXGOHZ.png)
instead of the "PAUSE" screen (also with BG music).
(http://i.imgur.com/7gWTzGh.jpg)
(lastbld2)
-
Hello Guys !
Concerning the CrouchingPoneyHiddenDragon 's neogeo title
a ROM is available on the Net, since fiew days ... :
www.lecortex.com/CPHD/CPHDdemo.zip Rom , about 51 MB
>leCortex: To celebrate this new year, we are releasing a playable demo of CPHD featuring
"The Mercant"....
Bye,
Fred/FRONT !
-
Wow! That's a pretty sweet demo. ^^ I'm sure JackC will have this added pretty quick. :)
-
Hello Guys !
iq_132 , you're welcome ... it's still a pleasure for me, to give us somes good news concerning Neo Geo development :)
->for CrouchingPoneyHiddenDragon, more information, interviews pictures & videos here :
https://www.facebook.com/CrouchingPoneyHiddenDragon
http://www.lecortex.com/CPHD/ LECORTEX's homepage
PS:
Furrtek, also developed for this CPHD's project, differents Hardware (Open source hardware, or not) : memory card, 'Hayaku programmer' ,etc
Actually, he present to us His recent creation: 'Fusion NeoGeo MVS to AES converter' ("Open source hardware MVS converter") By Sean Gonsalves aka Furrtek & Kyuusaku
More informations here :
http://www.indiegogo.com/projects/fusion-neogeo-mvs-to-aes-converter
http://wiki.neogeodev.org/index.php?title=Fusion_converter&t=20140105190558
https://www.facebook.com/pages/Fusion-Neogeo-Converter/181000805428036
PS: jeff KURTZ ( Knight's Chance 's project) presents to us 'FUSION'... like this :
>JEFF:
'If you're looking for a good and inexpensive MVS->AES converter to play your MVS copy of Knight's Chance on, why not check out this converter from furrtek, one of the Neo tech gurus from http://wiki.neogeodev.org/ '.
For the end, still Concerning Neo Geo development , this is few neo-geo Homebrews ( Work In Progress) :
http://ngpace.blogspot.com.au/ 'Neo-Kong' :NGPACE - A Neo Geo System Board Replacement Project by Tcdev/Mark McDougall
https://www.facebook.com/pages/Digital-Systems-Design/204244112933542 'Neo Geo Manic Panic ' by Digital-Systems-Design
http://www.iocero.com/EventDetail.aspx?idEvent=27312 'Galaga Emu or Phoenix, or..etc on NEOGEOCD Project' By BEY/iocero.com
Bye,
Frederic from FRANCE !
PS: sorry for the publicity for my friends ....
-
Crouching Poney Hidden dragon (DEMO)...
-
Very nice! A new HBMAME has been released for it. No idea what the object of the game is, though.
I have no idea what the object of HBMAME is either, but i do detect some jealousy in your tone :)
-
Hello Guys !
Big News
few days ago, The ASM 68k source code to 'Art of Fighting' was found on an old PC9821 (Neo Geo dev machine) .... :)
https://app.box.com/s/5on2e593tqt2c8ketc68
http://www.mameworld.info/ubbthreads/showthreaded.php?Number=320783
...
Its the full source code for Art of Fighting which can be assembled on PC with a little messing and runs as expected.
Thought it might be of some interest to 68k coders :)
After, The official 'Neogeo Hardware Specification', it' s an other Great News for people interested by the development on this system !!!
https://archive.org/details/NeoGeoHardwareSpecification
Bye
Fred / FRONT !
-
Contra (US, Set 1) (contra)- background missing, slowdowns, minor graphic glitches.
-
Task Force Harrier (tharrier)- spites graphic glitches
-
Donkey Kong (US set 1)(dkong) - some sounds are missing- insert coin and mario jump.
-
You'll need the samples for those. Fba doesn't do discrete sounds Atm. http://samples.mameworld.info/
-
Hi guys, I absolutely love FB Alpha! Thanks for the great interface and everything :)
I signed up to report a couple quirks that I came across in certain games.
19XX - Sometimes the water or background textures are a weird color after savestate load.
Aquajack - The water changes color/palette, similar to 19XX but more easily reproducable. (after savestate load) Now I'm having trouble reproducing this one.. Maybe it was a fluke?
Samuari Aces - corrupted background tiles when returning from savestates March 30, 2014 - FIXED
Battle Garegga - Crashes on savestate load with guru meditation error. Might have to save one or 2 savestates with this game before the problem shows. March 28, 2014 - FIXED - see my thread below this one :)
Fire Shark / Same! Same! Same! - when using samples (support\samples\fireshrk.zip) and reloading a save state, the music continues playing uninterrupted instead of starting at the savestate point. Also, there is a weird distortion after short sfx, like the insert coin noise. FIXED :)
best regards,
- dink
-
A small issue/buglet in the classic Konami game Scramble[EDIT: And Super Cobra], when the bomb explodes, it makes a loud/distorted crackling noise.
EDIT: Fixed - see my thread below
best regards,
- dink
-
NEOGEO CD BUGS....
Pulstar:
It is impossible to select 1P game start at the main menu. Only 2P option works.
Neo Driftout:
Current stage music does not stop after the level finishes. It carries on playing over every menu screen or even the Game Over screen.
Choutetsu Brikin'ger (Ironclad):
2 player ships appear on screen simultaniously, one on top of the other, even if you only select 1 player game.
-
Hello,
I found a problem with the M62 emulation (Kid Niki, Youjyuden, etc). After loading a savestate the audio dies and makes a loud irritating noise.
EDIT: well that was an easy fix, an updated d_m62.cpp driver will be available in my thread below.
EDIT2: I spoke too soon, the sound still hangs after x amount (random, usually between 1 and 5) state reloads. It kinda sounds like the noise an old console makes when you remove the cartridge while the game is plugged in - it might even be a bank switching issue.. hrm, any ideas guys?
best regards,
- dink
-
Sonic The Hedgehog 3 (Sega Megadrive) bug.
Windows XP Binary / 32-Bit.
This is difficult to explain but here goes...
No water / invisible water in Angel Island, Act 2:
Start Angel Island: Act 2, contine right as normal and head down to the bottom of the stage. You will hear a splashing sound, and see a water splash effect as you jump, but the water itself is totally invisible.
If you are playing as Tails, it looks funny as he is swimming through thin air! :D
*No cheats, codes or saves were used whilst playing. Game was started from a new 1 player game slot, playing as Sonic & Tails together.
Sorry i can't even post screenshots but i am away from my PC.
-
Hello,
Just documenting 2 UI abnormalities I've come across:
In the Select Game list / Filters: Clicking the hardware check doesn't clear it and all checks below it. FIXED
@Bootup:
Startup-Logo window causes the main window to show missing scanlines right below the about window.
Here's a picture to explain what I mean: http://s16.postimg.org/jlzoxgeet/fbwindowafterstartup.png
And a developing oddity:
Somehow the debug console facility in FB Alpha conflicts with gnu debugger, making it impossible to load & run (and get a backtrace on crash). gdb --args fbads.exe and 'r+enter' yeilds:
Welcome to the FB Alpha [DEBUG] debug console.
[hangs here]
EDIT, a couple more things - driver related :)
Raiden, the tempo of the in-game music is slow, like around 80bpm, it should be a lot faster. Try it in mame and then fba, and see what I mean.
Gyruss - no in game music both fixed
best regards,
- dink
-
Just wondering have any issues ever been reported for Asura Buster on PC FBA?? when using FBL on level 3 the game will
lock up but when using MAME it doesn't obviously it could be an issue confined to the xbox versions of FBL
-
Just wondering have any issues ever been reported for Asura Buster on PC FBA?? when using FBL on level 3 the game will
lock up but when using MAME it doesn't obviously it could be an issue confined to the xbox versions of FBL
Yes, the one game does lock up. It's an FBA issue. No idea what it does that. I am 90% sure it is related to the 32-bit 68k handling in fba. not sure why..
-
Test Sangokushi II:
wofhfh sgyxz wofh wofha wof3sj wof3sja wofch miss button C.
-
Test Sangokushi II:
wofhfh sgyxz wofh wofha wof3sj wof3sja wofch miss button C.
Thanks, its fixed :)
-
Hey dink, wanted to let you know that you have Super Mod powers with the status upgrade. You can go through and cross off stuff you've already done in this thread if you like. :)
-
Task Force Harrier (tharrier)- spites graphic glitches
Fixed!
-
iq_132: excellent :)
btw, tharrier was playing sounds from the wrong bank - it sounds awesome now [see svn] :biggrin:
EDIT:
Well, we almost got it. It seems there exists 1 last problem..
Its a weird shaking effect at the end of level 1, and 3, after you beat the bosses - your player sprite will shake/vibrate "violently" as the computer takes over and drives on to the next level.
I've attached 2 savestates that show this problem (just wait a second for the boss to blow up completely)
EDIT (again): The score is missing as well, it should appear above and slightly to the right of 1UP at the top of the screen.
best regards,
- dink
-
Nice work with Task Force Harrier there was another game in the nmk16 driver that had graphical problems although much worse
as i recall it was Rapid Hero. Not sure if there are any more games in the driver with graphical bugs?? only tried the two already
mentioned
-
gamez fan, with the latest svn version, Rapid Hero seems perfect graphically, its just missing the sound because we don't emulate the sound cpu yet (tmp90841).
best regards,
- dink
-
gamez fan, with the latest svn version, Rapid Hero seems perfect graphically, its just missing the sound because we don't emulate the sound cpu yet (tmp90841).
best regards,
- dink
Oh Ok then there's always a danger for me when reporting bugs as i use the xbox variant of FBA called FBL which has some cores
which are older and because of that some of the issues with certain games might be fixed in the latest SRC Anyway good to hear
it plays fine now in FBA
-
iq_132: excellent :)
btw, tharrier was playing sounds from the wrong bank - it sounds awesome now [see svn] :biggrin:
EDIT:
Well, we almost got it. It seems there exists 1 last problem..
Its a weird shaking effect at the end of level 1, and 3, after you beat the bosses - your player sprite will shake/vibrate "violently" as the computer takes over and drives on to the next level.
I've attached 2 savestates that show this problem (just wait a second for the boss to blow up completely)
EDIT (again): The score is missing as well, it should appear above and slightly to the right of 1UP at the top of the screen.
best regards,
- dink
Fixed the score... as for the sprite issue... that's weird. I have no idea. :\
-
@ gamez fan, well don't let that stop you from reporting any issues you might find :)
@ iq_132, well, I guess we can consider the shakey ship part a special effect ;p
best regards,
- dink
-
Gauntlet - game crash after first stage.
Alien Storm - some minor glitches on cutscenes
Excelsor - minor glitches
Night Striker - game freeze
-
Gauntlet - game crash after first stage.
Alien Storm - some minor glitches on cutscenes
Excelsor - minor glitches
Night Striker - game freeze
WILOWAR, thanks for the bug reports. So far, the good news:
Night Striker plays fine now, I just need to tighten up the inputs a little bit like I had done for Afterburner.
I'll edit this post as I progress to the next games in the list :)
best regards,
- dink
-
Ok, thx dink. I have found another bug on Dodonpachi and Gauntlet 2:
Player 1 inputs don't response , only second player.
Stage 2 are not playable, graphic mess
-
WILOWAR, thanks again for the bug infos! :)
EDIT: I played a few levels of Alien Storm, Excelsior (excelsr.zip, right?), DoDonpachi and they all seem to be in good working order with the latest developers version of FBA, Gauntlet 1 & 2 seem like the only ones left suffering from problems, let's see what can be done..
best regards,
- dink
-
Hi. I have found that glitch on Alien Storm
-
Ugg, I can't seem to come up with a solution the corrupted level2 in Gauntlet 1 & 2, I put many hours into it yet the fix eludes me.. Hopefully iq_132 or Treble Winner have better luck with it than I did!
best regards,
- dink
-
The SVN FBA from about a month ago had an issue where the default bullet sprites in the game Varth looked like plus (+). With the current svn, the bullet sprites are back to normal. I'm just curious as to which change was made to fix this? (I can't find anything)
thanks,
- dink
-
Most likely the 68K issue....
-
Most likely the 68K issue....
Ugg, that always seems to get me.. Last time it gave me trouble I had it surgically removed, and replaced with your compiled .obj's :)
-
Ugg, that always seems to get me.. Last time it gave me trouble I had it surgically removed, and replaced with your compiled .obj's :)
There is still the argument of removing a68k. I can't see it has much use these days.... does it?
-
Kev, I'd say keep it, at least until the issue with the .c core gets fixed when compiled with a newer gcc.
An example of what I mean: Varth has the wrong sprites for bullets, Altered Beast[the non-fd1074 version] leaves behind sprites, Enduro Racer crashes and a bunch of other weird issues with some games happen when FBA (or more specifically, the Musashi c core) is compiled with gcc 3.6.1 to the latest version. Switching to the ASM core at least solves that issue. :)
best regards,
- dink
-
A lot of drivers disable A68K anyway, so having it around doesn't really provide much.
Official binaries are MSVC compiled anyway, and have no issues.
I'd be interested to see if any of our changes have impacted on M68K at some point or if we've always had the issue with newer GCCs?
I wouldn't shed any tears if we nuked A68K though. :)
-
I'm not sure if this was addressed but I just noticed in older neogeo games that have high pitches make a crackling noise , has this been fixed?
-
SNK_Dude, can you be more specific? Which game(s), at which point in the game, etc.
best regards,
- dink
-
SNK_Dude, can you be more specific? Which game(s), at which point in the game, etc.
best regards,
- dink
well lets see,
Art of Fighting 2
Football Frenzy
are 2 games I noticed it in , it seems to happen at random in the emulator.
In AOF 2 it was when YOU WON or YOU LOST appeared on the screen and with the high pitched TING noise, the noise was crackly.
In Football Frenzy it happened when you were picking out a play.
This happened a few times before with other neogeo games with high pitched noises , cant seem to make it happen at will .
Sorry dink, I'm very bad at describing things well :(
Edit: I THINK I found the problem.........4 point 3rd order FM interpolation was on.........
-
EDIT: nevermind what I wrote earlier, I found the problem that sometimes makes a nasty static noise, thanks for reporting it, you did a good job explaining it :) The 4point 3rd order interpolation is supposed to sound better, not worse.. hehe..
best regards,
- dink
-
I found a sprite priority problem in Konami's Ajax/Typhoon, the tanks and boats go ontop of the bridges, while travling on land or water. Attached are pics and savestates to get to the exact location.
-
Here's another savestate further into the game, where the tanks drive ontop of the houses/buildings :)
-
Does anyone have time to take a look at the Gauntlet 1/2 driver? If it can't be fixed, can I suggest we disable it or make it debug-only because in its current state its not playable.
The problem: tiles are messed up on the next level in both games. Start with Gauntlet 2, Just find the first exit, and it'll take you to level 2. In Gauntlet 1, the second level doesn't look too corrupted, but it really is - try the second level using Mame and you'll see what I mean - the level is laid out proper... or go a little to the right and enter the "exit to level 4" box in FBA, and the game will get stuck after it draws the next screen.
best regards,
- dink
-
Reverting the changes to the driver in SVC commit 783 return it to it's best state (ie, the only issue being the missing POKEY sounds).
-
Treble Winner, I sorta see what happened, in #783 the SlapStick protection emulation was separated from the Gauntlet driver into a separate file so it could be used by other drivers - and perhaps a type-o or slight oversight caused this bug. Now if I could only slip out of this code-burnout, I would try to fix it :) [I probably just need a little break]
Btw, we never had POKEY sounds due to a lack of a POKEY driver.
best regards,
- dink
-
I know what happened - I wrote the driver. :)
Was just pointing out exactly what broke it.
I never got around to porting a POKEY core, hence that known issue.
-
I know what happened - I wrote the driver. :)
Was just pointing out exactly what broke it.
I never got around to porting a POKEY core, hence that known issue.
And I broke it. 😁
-
Treble Winner, Oops, I didn't mean to sound like a d*ck, sometimes I say the obvious when I'm trying to get a better understanding of things.. and some times I'm waaay off, I thought you meant that backing out change 783 removed the pokey sounds.. my bad! :redface:
-
Image fight (Japan, Revision A) (d_m72.cpp) - all the graphics are completely corrupt.
If you look closely, you can sorta see the titlescreen under the corrupt garbage on the screen, while its running.
The game works fine in the last public release, from June 2013 - 0.2.97.29, but not in svn.
-
Image fight (Japan, Revision A) (d_m72.cpp) - all the graphics are completely corrupt.
If you look closely, you can sorta see the titlescreen under the corrupt garbage on the screen, while its running.
The game works fine in the last public release, from June 2013 - 0.2.97.29, but not in svn.
Can't verify ^^
-
Can you think of anything that might cause this? I tried the basic blitter, new fba ini, deleted config/games/imgf*, re-compiled d_m72.cpp. The other Irem m72 games work fine.. I'm out of ideas :redface:
The highscore table seems to be the only thing that isn't totally corrupted:
EDIT: I also backlevel'd d_m72.cpp from a year ago and recompiled, no dice.
-
No corrupted graphics too for Image fight (Japan, Revision A) (d_m72.cpp)...
-
Thanks for checking, guys :)
I'm not crazy after all, the corruption I was seeing was actually caused by an overlapping src/dest address in memcpy();
Technical info:
Here's the deal, when using memcpy(), the destination address cannot overlap the source address, if it does, "unexpected results" may happen. I didn't come up with this rule, I guess some sort of [lib]C standard did..perhaps? :)
in imgfightRomLoadCallback();
...
memcpy (DrvV30ROM + 0x40000, DrvV30ROM + 0x20000, 0x40000);
I quick change to memmove() solves the problem! It allows for overlapping addresses. I guess this is specific to certain versions of gcc.. mine is 4.6.1. Curious what version iq_132 and JacKc is using?
-
I don't know exactly which version i am using as it was iq_132 who sent me it...But it seems to be an old MinGW version.
-
JacKc, could you run: gcc --version :)
-
Need to be careful with memmove though. Some hardware expects the data in both locations.
A better fix would be to address why the drivers are trying to copy more data than they have available (and just copy the correct amount of data) and continue with memcpy.
-
EDIT: re: image fight, the memmove function is just relocating the rom into the way imgfight expects it, see m72.c from mame..
(fba loads it linearily, but imgfight wants 0x20000 bytes at the beginning of the rom then the rest of the rom loaded at 0x40000)
excerpt from man on bsd,
void *
memmove(void *dst, const void *src, size_t len);
DESCRIPTION
The memmove() function copies len bytes from src to dst. The two strings may overlap;
the copy is always done in a non-destructive manner. <-
-
JacKc, could you run: gcc --version :)
Let's do it !
C:\MinGW>gcc --version
gcc (4.3.3-tdm-1 mingw32) 4.3.3
Copyright (C) 2008 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
-
JacKc, Thanks :)
-
I dont like reporting here as im using a build called FBL rather than the latest FBA but..... :biggrin:
Risky Challenge and the clone Gusson Oyoyo seem to have problems i was wondering any chance of the fix for
this from MAME being inserted into the NEC CPU core and the irem decryption tables...???
If it's already been done then accept my apologies ;p
-
I dont like reporting here as im using a build called FBL rather than the latest FBA but..... :biggrin:
Risky Challenge and the clone Gusson Oyoyo seem to have problems i was wondering any chance of the fix for
this from MAME being inserted into the NEC CPU core and the irem decryption tables...???
If it's already been done then accept my apologies ;p
Fbal uses an older NEC core. Doesn't support those games
-
Fbal uses an older NEC core. Doesn't support those games
Oh no worries then just thought i'd ask as it's a nice wee quirky game i may try to roll the fixes into my src but like the
game it's a Risky Challenge when i start messing around with CPU Cores :biggrin:
-
In the game Tokio / Scramble Formation (tokiob.zip w/tokio.zip, drv/taito/d_bublbobl.cpp), about every screenfull of background traveled, there is a little gap between where the backgrounds should link together. Here's a pic to show what I mean, you'll see this line happen about every screenfull. (When you get to the clouds section of the level, its even more apparent)
-
Here's a savestate that'll take you to the clouds section, where the gap in the background (where it should be seamless) is more apparent. I tried to repair it myself with that I learned from Treble Winner on the Contra level 2 fix, but it got the best of me.. :(
-
IIRC the hardware doesn't use tilemaps so it would be different from Contra.
I'll try and take a look at it.
-
I figured it would be fun to try to solve this one myself, just for something to do, it turns out that the vsync irq was happening too late (total interleaves 100, irq at interleave 98), and the cpu didn't have enough time between when the irq was triggered to when the screen was updated to update the sprite/video ram. I spent half a day debugging in DrvVideoUpdate() to come to this conclusion hehe. :eek: :p
-
Hi guys,
Would you please take a minute to confirm this behavior:
Load a game from the list (f6), but make sure the game is at least on the second page (or farther down) the list.
After the game loads, go back to the list with f6, and do you see the game you've previously selected in the list highlighted with the blue bar? If its not there, click the "down" arrow to the rightside of the list twice, now is it visible?
thanks & best regards,
- dink
-
Found a bug in Tigeroad (us) that doesn't happen in MAME.
When going though the doorway to the third (I think) Boss, all the background tiles are mixed up. After a second or so, it will go to normal.
Attached you'll find a savestate, just load it then go through the doorway to see what I mean.
Pics: 1) at the doorway, 2) going though the doorway, and the new room scrolls into place, 3) in the room, now things are normal again.
(note: the problematic tiles are drawn with draw_background(0); in DrvDraw() in pre90s/d_tigeroad.cpp)
-
Sweet. I'll take a look tonight or tomorrow.
-
I have this problem in Sol Divide. :eek:
http://postimg.org/image/e0s26m33n/
-
Freddy,
Eek! To help me track down the problem, let me send you a debug version of FBA via pm, please run it and attach a screencap and the file zzBurnDebug.html (you'll have to zip it before attaching it to a message here)
thanks,
- dink
-
It seems in Metal Slug 5 the A B C or D buttons do nothing during gameplay, the directions work just not the face buttons.
-
Freddy,
Eek! To help me track down the problem, let me send you a debug version of FBA via pm, please run it and attach a screencap and the file zzBurnDebug.html (you'll have to zip it before attaching it to a message here)
thanks,
- dink
problem solved ;p
-
It seems in Metal Slug 5 the A B C or D buttons do nothing during gameplay, the directions work just not the face buttons.
Just tested and nothing wrong for me with the A B C or D buttons...
(http://i.imgur.com/vz0CDTz.png) (http://i.imgur.com/6IatSv7.png)
Which pad, gamepad are you using? Which Os and which FBA version?
-
Just tested and nothing wrong for me with the A B C or D buttons...
(http://i.imgur.com/vz0CDTz.png) (http://i.imgur.com/6IatSv7.png)
Which pad, gamepad are you using? Which Os and which FBA version?
Mayflash PS to PC controller converter, Windows 7, FBA 029730 32 bit since 64 bit is kinda buggy.
OK found the problem, its the universe bios 3.1 thats causing that strange glitch where the face buttons dont work during gameplay.
Edit: its ALL versions of the universe bios doing it, guess I better send a bug report to Raz.
-
Thanks for the infos SNK_Dude :smilie:
Could you tell me which setting are you using with unibios 3.1 (Arcade or Console, Region, etc...) to see if i could reproduce this bug?
best regards,
JacKc.
-
Thanks for the infos SNK_Dude :smilie:
Could you tell me which setting are you using with unibios 3.1 (Arcade or Console, Region, etc...) to see if i could reproduce this bug?
best regards,
JacKc.
http://unibios.free.fr/knownissues.html
Start metal slug 5 in arcade mode then switch to console mode in the Uni-bios, in order for the controls to work you have to get a gameover which is strange I know :P
Just set your lives to 1 in the F2 menu and hurry and die and then they will work .
Edit: seems the sound problem still persists from previous versions in metal slug like when you shoot your pistol and the sound is crackly , can you please check on that as well?
-
SNK_Dude, try Audio -> Disable FM Interpolation, it usually fixes any crackles/distortion w/Neogeo games.
best regards,
- dink
-
SNK_Dude, try Audio -> Disable FM Interpolation, it usually fixes any crackles/distortion w/Neogeo games.
best regards,
- dink
Tried that already, its still crackly, give it a try and you will see.....well hear what I mean :P
-
Tried that already, its still crackly, give it a try and you will see.....well hear what I mean :P
The gun sound sounds alright to me, I'll admit that in mame the gun sound is a bit muted sounding, but I don't think it sounds bad in fba at all. What do you guys think?
-
The only potential for any issue I see is that we don't clip the result for each stage in burn_ym2610.cpp. Only the final result is clipped.
In reality this may make very little difference.
-
The only potential for any issue I see is that we don't clip the result for each stage in burn_ym2610.cpp. Only the final result is clipped.
In reality this may make very little difference.
I tried checking for clipping at all stages, then I realized it was mixing with a 32bit int, so that wouldn't have made a difference - but I ran the new code anyways, and sure enough, it made no difference. I've seen this before, I think its coming from the psg/ssg part of the ay8910 - in some games (like the games from the new(to FBA)-UPL driver iq_132 had recently wrote), the ssg sound channel was about 5x louder than the fm parts (Atomic Robo Kid, Mutant Night), somehow this had been fixed in MAME when they re-wrote the entire 8910 code - at this point it would be impossible to graft in a fix, so I took the ez route and simply lowered the psg(ssg) channels to match the fm ones. I can't see doing it for Metal Slug 5 though, the crackle of the gun actually sounds pretty good/non-irritating to me, even with headphones. Thanks for reporting it, though :)
best regards,
- dink
-
not sure if it applies to Metal Slug 5 because I'm not hearing it, but some of the later NeoGeo games were clearly developed using an inaccurate emulator (probably NeoRageX) as part of the development.
As a result Nightmare in the Dark has a really annoying crackle when you jump, it didn't happen in very old versions of MAME, and it didn't happen in NeoRageX, but it did happen once we started emulating actual limits of the chip properly, and it does happen on real hardware (so we emulate it) Not respecting the chip limits actually breaks some other games in worse ways!
-
I tried checking for clipping at all stages, then I realized it was mixing with a 32bit int, so that wouldn't have made a difference - but I ran the new code anyways, and sure enough, it made no difference. I've seen this before, I think its coming from the psg/ssg part of the ay8910 - in some games (like the games from the new(to FBA)-UPL driver iq_132 had recently wrote), the ssg sound channel was about 5x louder than the fm parts (Atomic Robo Kid, Mutant Night), somehow this had been fixed in MAME when they re-wrote the entire 8910 code - at this point it would be impossible to graft in a fix, so I took the ez route and simply lowered the psg(ssg) channels to match the fm ones. I can't see doing it for Metal Slug 5 though, the crackle of the gun actually sounds pretty good/non-irritating to me, even with headphones. Thanks for reporting it, though :)
best regards,
- dink
Since I added support for the independent sound routes you can reduce the volume of each channel individually.
I can't test at the minute but I always assumed the gun sound was ADPCM, so not related to the AY part anyway?
-
Since I added support for the independent sound routes you can reduce the volume of each channel individually.
I can't test at the minute but I always assumed the gun sound was ADPCM, so not related to the AY part anyway?
It could just as well be :)
-
not sure if it applies to Metal Slug 5 because I'm not hearing it, but some of the later NeoGeo games were clearly developed using an inaccurate emulator (probably NeoRageX) as part of the development.
As a result Nightmare in the Dark has a really annoying crackle when you jump, it didn't happen in very old versions of MAME, and it didn't happen in NeoRageX, but it did happen once we started emulating actual limits of the chip properly, and it does happen on real hardware (so we emulate it) Not respecting the chip limits actually breaks some other games in worse ways!
The dark-side of emulation. I guess it was easier to use a PC emulator for a quick test than a devkit?
-
64 bits version is stamped 0.2.97.32 ^^
-
64 bits version is stamped 0.2.97.32 ^^
Jees - you downloaded it quick - was up for like two minutes before I replaced it.
Download again to fix to get the correct version. :)
-
^_^
-
Sound, except for explosion, doesn't appear to be working in Galaga (even with the samples in place).
-
The explosion is the only sample. The other sound should be emulated and working.....
-
The explosion is the only sample. The other sound should be emulated and working.....
Thats the problem, there are no other sounds :/
EDIT: nice fix, thanks.
-
In the Megadrive driver, Wonder Boy in Monster World, the top 4 lines seem to display random tiles instead of the info boxes, I can't seem to figure this one out..
best regards, [attached - pic from FBA & pic from Gens/how it should look]
- dink
-
Hi dink. Could fix a Megadrive emulation in FBA. It's a one garbage. :smilie:
-
Oh yeah, in Wonder Boy I knew of this bug... and this two black lines on the side... it's terrible. :eek:
By the way... it's a very fun game!
-
Fixed DrawWindow() 32-cell mode.
-
Treble Winner, you're the king!! :D Thank you, my good man :)
best regards,
- dink
-
Hello everyone, Somehow when I load Gradius III screen just stays black except for the first time I loaded the (World) version which screen was gray and black and nothing happened. I tried loading that same version again and screen was just black for a few mins I would say like 5 mins before I tried loading it again. I tried loading all 3 versions with basically same results. Don't know if it just takes a while to load but that seems pretty long.
-
Almost forgot. I'm using the X64 version and operating system is Windows 8. Every other game I tried works fine so far
-
Never mind everything is working with Gradius III
-
Glad to hear it! :)
best regards & happy gaming,
- dink
-
A dev-related issue:
Regarding gamelist.txt - for the first 44 games in the list, the Hardware column is moved over to the remarks column. Weird, eh?
-
Thanks Dink on Gradius III response.
I did find a issue with Kero Kero Keroppi's Let's Play Together (USA, Version 2.0). This one loads but screen is all scrambled. The Japan version is fine.
-
Fixed. ^^
-
you da man IQ
-
Hey! Someone owes me a sticker.
-
Hey! Someone owes me a sticker.
You're not allowed stickers!
-
You're not allowed stickers!
Aww!!
-
you'll get a sticker when the Midway driver is implemented. Oh no my cat is gonna die. lol
-
I was testing Quiz Tonosama no Yabou (Japan) and it loads but then the screen just stays black.
-
Found a problem with Super Chase Criminal Termination. The screen freezes during the demo and some text appear. Here is a picture
-
There also seems to be a issue with The Irritating Maze / Ultra Denryu Iraira Bou. Game won't boot up. Just a black screen. FBA just totally freezes
-
There also seems to be a issue with The Irritating Maze / Ultra Denryu Iraira Bou. Game won't boot up. Just a black screen. FBA just totally freezes
The good news is, that one has already been fixed - the fix will be in the next release :)
best regards & thanks for all the bugreports!
- dink
-
Anytime Dink. I'll continue reporting as I get to them.
-
Hello guys, The King of Fighters 2003 games don't work correctly. The only ones I got to run with no issues is bootleg 2 and KOF 2004 Ultra Plus bootleg. The other ones seem to have graphic issues. This also applies to KOF 2003 dedicated PCB version.
-
Hi csmart,
could you please give more details and/or snapshots about the kof2003 graphical issues?
-
JacKc, I just tried kof2003 (neogeo mvs), the game is entirely broken it seems. It says insert coin / credit 00 at the bottom of the screen, but there are no graphics in the middle of the screen at the titlescreen, and if/when you get the game to start, everything is out of place/distorted. Do you get a similar result?
-
JacKc, I just tried kof2003 (neogeo mvs), the game is entirely broken it seems. It says insert coin / credit 00 at the bottom of the screen, but there are no graphics in the middle of the screen at the titlescreen, and if/when you get the game to start, everything is out of place/distorted. Do you get a similar result?
Could it be a NeoGeo bio's issue..??? i know Iron-Clad only works with Japan and USA Bios and has close to the same
issues you mentioned above if using say Euro Bio's
-
No major problem here concerning kof2003 except some minor glitches with lifebar and character's name under lifebar during play [tested with FBA debug build 0.2.97.32 but same for FBA unicode build 0.2.97.31 with MVS Asia/Europe ver.6 (1 slot) bios and MVS Japan ver. 6 (?slot)]:
(http://i.imgur.com/xQhRAqe.png) (http://i.imgur.com/HdOnrme.png)
(http://i.imgur.com/YCo64bh.png) (http://i.imgur.com/G7uA0Gu.png)
(http://i.imgur.com/YPicSwL.png) (http://i.imgur.com/fPlMrkp.png)
-
My problem tends to be the same as Dink's. Only a couple of bootlegs worked fine. Would using the Neo Geo universal bios have to do anything with it.
-
EDIT: see the next message instead.
Crud, I can't get it working after messing around a bit with the dipswitch settings (bios,etc). I tried exactly the same settings as JacKc used, "MVS Asia/Europe ver.6 (1 slot) bios and MVS Japan ver. 6 (?slot)"
hmmm.. any more ideas? :)
-
I got it!
Its one of those memcpy() overlap problems. (derek techno-babbles on: if the destination/source address overlap while its copying, it will work sometimes, but not all the time. Or sometimes it will work, then later on that day it won't work depending on if the executable it loaded in a different code segment. ..Or it will work perfect on some systems, but not others. I probably missed a couple more reasons why it would fail. :P memcpy() can be a real pain in the ass :) memmove() is a perfect replacement for memcpy, because it does exactly the same thing except it doesn't fail on overlapping src/dest buffer copies.)
EDIT: fixed checked into svn. Thanks again for reporting the problem, cmart :)
best regards,
- dink
-
Dont like reporting bugs but Robocop 2 has similar graphical issues to caveman ninja before it was fixed same kinda thing
with the sprites Plus the alignment seems off in MAME for example you can only move the character to left or right edge of the screen
but i noticed in FBA i cant move all the way right and can actually disappear of the left hand side of the screen
-
My pleasure Dink
-
Dont like reporting bugs but Robocop 2 has similar graphical issues to caveman ninja before it was fixed same kinda thing
with the sprites Plus the alignment seems off in MAME for example you can only move the character to left or right edge of the screen
but i noticed in FBA i cant move all the way right and can actually disappear of the left hand side of the screen
gamez fan, Thanks for reporting it! Should be fixed now :) Please don't hesitate to report other bugs.
On the other hand..
I noticed that in the driving scene (level 3, I think) the road is always straight, looks like the raster effects weren't implemented into FBA's deco16ic :( The part is still playable thought & its really short, like a minigame. I might try to implement it someday - because it would also fix the driving scene in The Cliffhanger (edrandy).
best regards,
- dink
-
Small Graphical Problem in super Burger time it seems to sometimes affect the smaller burgers the big ones are ok
(http://www.fshare.it/image.jpeg?i=m41409871wh)
Looks like someone sat on the burger bottom left of the screen it doesn't look good at all LOL
-
gamez fan, surprizingly enough, that how its supposed to look (its the same in MAME).
best regards,
- dink
-
VS Block Breaker (Asia) and it's clone Saru-Kani-Hamu-Zou (Japan) just gives me a black screen once it loads and does not go anywhere from there
-
gamez fan, surprizingly enough, that how its supposed to look (its the same in MAME).
best regards,
- dink
Hmm in my MAME set it doesn't look this way the burgers and fillings are displayed correctly here's a video from
youtube
https://www.youtube.com/watch?feature=player_detailpage&v=MHOKqB8Or3I#t=315 (https://www.youtube.com/watch?feature=player_detailpage&v=MHOKqB8Or3I#t=315)
-
from the MAME source
Super Burgertime has a few bugs:
Some sprites clip at the edges of the screen.
Some burgers (from crushing an enemy) appear with wrong colour.
Colour cycle on title screen doesn't work first time around.
These are NOT driver bugs! They all exist in the original game.
-
Well that clears that up then LOL
-
Hmm in my MAME set it doesn't look this way the burgers and fillings are displayed correctly here's a video from
youtube
https://www.youtube.com/watch?feature=player_detailpage&v=MHOKqB8Or3I#t=315 (https://www.youtube.com/watch?feature=player_detailpage&v=MHOKqB8Or3I#t=315)
That weird white colored patty happens when you squash an enemy in the burger, but in the video, he didn't squash an enemy. Run fba and mame side-by-side with super burger time, and when the title screen goes to game-demo mode, you'll see what I mean.
best regards,
- dink
-
Mmmm... burger time. That game makes me so hungry!
-
Mmmm... burger time. That game makes me so hungry!
Ha! Yes, I love it.. The classic version (Colecovision) is even better IMO :D
-
How's Black Tiger play in FBA the game seems to always crash before starting level 3 in FBL ive read all the release threads
i dont see any mention of it ever being broken only thing i noticed was iq_132 did fine tune the driver a while back
i implemented these code changes into the FBL black tiger driver and still the game crashes
-
gamez fan, please try to play the FBAlpha version of Black Tiger to level 3, if it crashes here I can fix it asap. The reason I ask you to play it - I suck terribly at the game.
p.s. at the end of level 2, make a savestate (check your pm first) so getting back to the crash is easy.
best regards,
- dink
-
@dink PM sent
-
Confirmed black Tiger crashes in FBA just after the 2nd boss
FB Alpha [DEBUG] v0.2.97.32 fatal exception report (Sat Sep 06 19:13:1
----------------------------------------------------------------------
Exception 0xC0000005 (EXCEPTION_ACCESS_VIOLATION) thrown.
EIP: 0x006B
(attempting to read address 0x07CB3621)
EAX: 0x00000000, EBX: 0x00001C00, ECX: 0x00007001, EDX: 0x00000040
ESI: 0x00000000, EDI: 0x00007000, ESP: 0x0022C890, EBP: 0x07CAC620
-
Seems this is the cause
</font><font color=#FF3F3F>ZetTotalCycles called when no CPU open
ZetTotalCycles called when no CPU open
I noticed this code was disabled in the driver i hoped by enabling it again it might sort the crash in black tiger
void __fastcall blacktiger_out(UINT16 port, UINT8 data)
{
switch (port & 0xff)
{
case 0x00:
{
// INT64 cycles = ZetTotalCycles();
// ZetClose();
// ZetOpen(1);
// INT32 nCycles = ((INT64)cycles * nCyclesTotal[1] / nCyclesTotal[0]);
// if (nCycles <= ZetTotalCycles()) return;
// BurnTimerUpdate(nCycles);
// ZetClose();
// ZetOpen(0);
after enabling it the game still crashes after level 2 i guess if FBA development was as easy as that we'd all be doing it LOL
-
gamez fan, its crashing while indexing the bg ram. The zettotalcycles when no cpu open is quite common, btw.
Well, I poked around for about an hour, but can't figure it out. Right when level 3 starts, it crashes right here:
in draw_bg():
INT32 attr = DrvBgRAM[(ofst << 1) | 1];
I tried increasing the size of DrvBgRAM, but then the background seems to go missing on the later levels.
Catches:
DrvVidRamBankswitch only indexes 4 (0-3) banks (0x1000 each)
The game tries to switch to bank 5 (4) before level 1 and 3. If I disable the 4 bank limit in the bankswitch function ( & 3), the background goes missing in the first level.
Here's a savestate to take you right to the end of level 2, thanks to gamez fan for providing it.
Looks like we're gonna need help from someone better at this than me, like IQ and Treble Winner :)
best regards,
- dink
-
Hmmm i see this in the MAME video src file for black tiger
#define BGRAM_BANK_SIZE 0x1000
#define BGRAM_BANKS 4
i only count 3 in FBA
INT32 nBank = (bank & 3) * 0x1000;
ZetMapArea(0xc000, 0xcfff, 0, DrvBgRAM + nBank);
ZetMapArea(0xc000, 0xcfff, 1, DrvBgRAM + nBank);
ZetMapArea(0xc000, 0xcfff, 2, DrvBgRAM + nBank);
so i added a forth
INT32 nBank = (bank & 3) * 0x1000;
ZetMapArea(0xc000, 0xcfff, 0, DrvBgRAM + nBank);
ZetMapArea(0xc000, 0xcfff, 1, DrvBgRAM + nBank);
ZetMapArea(0xc000, 0xcfff, 2, DrvBgRAM + nBank);
ZetMapArea(0xc000, 0xcfff, 3, DrvBgRAM + nBank);
then made DrvBgRAM = Next; Next += 0x008000;
i can now get to level 3 in FBL without the game crashing but the background is missing some progress i guess LOL
-
Hmmm i see this in the MAME video src file for black tiger
#define BGRAM_BANK_SIZE 0x1000
#define BGRAM_BANKS 4
i only count 3 in FBA
INT32 nBank = (bank & 3) * 0x1000;
ZetMapArea(0xc000, 0xcfff, 0, DrvBgRAM + nBank);
ZetMapArea(0xc000, 0xcfff, 1, DrvBgRAM + nBank);
ZetMapArea(0xc000, 0xcfff, 2, DrvBgRAM + nBank);
so i added a forth
INT32 nBank = (bank & 3) * 0x1000;
ZetMapArea(0xc000, 0xcfff, 0, DrvBgRAM + nBank);
ZetMapArea(0xc000, 0xcfff, 1, DrvBgRAM + nBank);
ZetMapArea(0xc000, 0xcfff, 2, DrvBgRAM + nBank);
ZetMapArea(0xc000, 0xcfff, 3, DrvBgRAM + nBank);
then made DrvBgRAM = Next; Next += 0x008000;
i can now get to level 3 in FBL without the game crashing but the background is missing some progress i guess LOL
I came to exactly the same conclusion, a missing background in level3 :/
But let me explain ZetMapArea. The 3rd param is for read/write/fetch access to the bank, and the bank pointer is specified in the next parameter: DrvBgRAM + nBank
Which is calculated here:
INT32 nBank = (bank & 3) * 0x1000;
bank & 3 means that it will allow bank to = 0 - 3, thats 4 banks. The & 3 part is like a limiter, anything over 3 will go to 0.
I spent a retarted amount of time on this one with no luck, we really could use a hand from IQ or Treble Winner :)
best regards,
- dink
-
I came to exactly the same conclusion, a missing background in level3 :/
But let me explain ZetMapArea. The 3rd param is for read/write/fetch access to the bank, and the bank pointer is specified in the next parameter: DrvBgRAM + nBank
Which is calculated here:
INT32 nBank = (bank & 3) * 0x1000;
bank & 3 means that it will allow bank to = 0 - 3, thats 4 banks. The & 3 part is like a limiter, anything over 3 will go to 0.
I spent a retarted amount of time on this one with no luck, we really could use a hand from IQ or Treble Winner :)
best regards,
- dink
I should be home tomorrow. I'll take a crack at fixing this then.
-
I did some more poking around (aimlessly), and if you put right under INT32 scrolly = ....:
if (!type) {
scrolly = ((*DrvScrolly)) & (0x7ff >> 2);
}
You'll actually get a background in level3, but its not in sync. with the sprites. It seems like things aren't being calculated properly in 4x8 mode.
Here's some pics, the first is fba w/that above codesnippet, and blacktiger level3 how its supposed to look. (clipped from yt)
best regards,
- dink
-
I did some more poking around (aimlessly), and if you put right under INT32 scrolly = ....:
if (!type) {
scrolly = ((*DrvScrolly)) & (0x7ff >> 2);
}
You'll actually get a background in level3, but its not in sync. with the sprites. It seems like things aren't being calculated properly in 4x8 mode.
Here's some pics, the first is fba w/that above codesnippet, and blacktiger level3 how its supposed to look. (clipped from yt)
best regards,
- dink
Fixed. You're going to be rather unhappy with me when you see the fix. lol
-
Nice one!! was it the priority masks that caused the problems..???
-
Nice one!! was it the priority masks that caused the problems..???
Sure thing man.
Find this:
ofst = (sx & 0x0f) + ((sy & 0x0f) << 4) + ((sx & 0x30) << 4) + ((sy & 0x70) << 7);
make it this:
ofst = (sx & 0x0f) + ((sy & 0x0f) << 4) + ((sx & 0x30) << 4) + ((sy & 0x70) << 6);
-
Fixed. You're going to be rather unhappy with me when you see the fix. lol
Great job :D
best regards,
- dink
-
Sure thing man.
Find this:
ofst = (sx & 0x0f) + ((sy & 0x0f) << 4) + ((sx & 0x30) << 4) + ((sy & 0x70) << 7);
make it this:
ofst = (sx & 0x0f) + ((sy & 0x0f) << 4) + ((sx & 0x30) << 4) + ((sy & 0x70) << 6);
Easy when you know your stuff LOL thanks for the fix!!!
-
Bug in search, sometimes it doesn't find everything.
For example. I type Mahou into the search box, and it brings up
Dimahoo
Great Mahou Daisakusen (clone of Dimahoo)
Sorcer Striker
Mahou Daisakusen (clone of Sorcer Striker)
(So whats the problem?) It didn't bring up:
Shippu Mahou Daisakusen / Kingdom Grandprix
ideas? looking at the code, it looks like everything is alright...
-
Vendetta - theres a lot of sprite priority issues, and clipping in the first boss-fight on the back of the truck.
The sprite priority issues start to happen around level 3, I think. There's people that come out behind doors, but you see them before they come out. Also, on the ship, there are scooba-divers that swim right through the ship. I don't have images or states to back this up at this time, will try to get them in the next few days!
Clipping: see the attached pic (c.o. JacKc) - the left side of the truck is missing sometimes, happens at the first boss encounter:
-
PCEngine, rarely something will show up sort of off by a few pixels. I think it has something to do with windowing, but, its just a guess :)
-
Violent Soldier / Sinistron (same game/different name) on PCEngine has what appears to be a similar problem, just start the game and play for a few seconds, you'll see what I mean. In the below picture, the only thing out-of-place is that brown horizontal line towards the top.
Something weird: if you play for a few more screens-worth of the game, that weird brown line will actually line up with the tiles above it.
best regards,
- dink
-
iq_132, I think the left-side clipping fix for Vendetta caused some right-side sprite clipping in X-Men.. uhoh.. :) Gamez Fan found this one btw, its really easy to spot - load X-men, insert coin, select character, and watch the intro-cutscene on the right-hand side, sprites start to disapear prematurely.
best regards,
- dink
-
iq_132, I think the left-side clipping fix for Vendetta caused some right-side sprite clipping in X-Men.. uhoh.. :) Gamez Fan found this one btw, its really easy to spot - load X-men, insert coin, select character, and watch the intro-cutscene on the right-hand side, sprites start to disapear prematurely.
best regards,
- dink
He already knows i sent him a PM about it im always the bearer of bad news LOL on a seperate note i still cant suss why
Gradius 3 has gone all green for me :S
-
Cyber Tank seems to have sound issues it works sometimes other times it doesn't plus it's quite low in FBL
in FBA however ive yet to hear it boot with sound ive loaded it a few times now
-
Toki needs to be hooked up to seibu sound and have the internal old/buggy seibusound removed. :)
-
Toki needs to be hooked up to seibu sound and have the internal old/buggy seibusound removed. :)
Done. :D
Cyber Tank seems to have sound issues it works sometimes other times it doesn't plus it's quite low in FBL
in FBA however ive yet to hear it boot with sound ive loaded it a few times now
Doesn't seem to be a pc issue. ^^
-
error about state saving and loading
:S
game: kof97
I saved the state to a file called "xxx.fs"in the mobile A ,then copy it to another mobile B,and load the state file "xxx.fs".
fba crashed! but load the xxx.fs in mobile A is ok.
how to solve this problem!
thanks
-
error about state saving and loading
:S
game: kof97
I saved the state to a file called "xxx.fs"in the mobile A ,then copy it to another mobile B,and load the state file "xxx.fs".
fba crashed! but load the xxx.fs in mobile A is ok.
how to solve this problem!
thanks
Are you using FBA in windows .97.31?
-
Some problems with DJ Boy if you leave the attract mode running after a few cycles the colours like when the game 1st boots
start to reappear shortly afterwards the game will freeze confirmed by dink in FBA also i thought id mention airbuster in
FBL does have some graphical problems as well mostly scrolling based maybe thats not the case in FBA though
For the record in the short playthough of level 1 and part of level 2 in DJ Boy i seen no problems of course the game
could mees up later though
-
Doesn't seem to be a pc issue. ^^
I've more than likely made a mistake with the sound defines then in the driver i might look at it again sometime
but the game does have control issues in FBL also which are kinda game breaking so if it had sound wont really make
much of a difference if you know what i mean LOL
-
I hate being the bearer of bad news but that ole bug with Battle Shark is still happening just tried it in FBA
-
Are you using FBA in windows .97.31?
Using FBA .97.31 in Android.
I have saved a state in mobile,and copy it to windows . That's not work too!
-
Hi my friends!
In the Violent Storm game, the transparency effects appears even when is disabled.
By the way, great release!
-
Hi my friends!
In the Violent Storm game, the transparency effects appears even when is disabled.
By the way, great release!
Those aren't accessory blend effects, but original alpha blending used by the Konami hardware. (it's supposed to be there)
-
I dont know if it is bug, Colecovision and SG-1000 games arent added parent/clone relation.
-
Background glitch on Martial Champions (EAB) (mtlchamp):
-
Small graphic glitch on Vendetta (vendetta 2pd)
-
Background glitch on Martial Champions (EAB) (mtlchamp):
Martial champions is very broken. I just forgot to mark it as not working.
-
Bishi Bashi Championship Mini Game Senshuken (ver JAA, 3 Players) graphic issues
-
Bishi Bashi Championship Mini Game Senshuken (ver JAA, 3 Players) graphic issues
One of the graphics roms is a bad dump. Once it's re dumped, the game will work fine. (not a bug)
-
Glitch for Dink. - Small graphic glitch on Turbo Force
-
Glitch for Dink- Bad priority sprites on Thunder Dragon 2
-
Glitch for Dink. - Small graphic glitch on Turbo Force
The Aero Fighter driver needs priorities hooked up. (This will require a bit of re-writing the graphics routines).
-
Glitch for Dink - Some highlights are bad on Violent Storm (EAC)
-
Glitch for Dink - Some shadows are bad on Violent Storm (EAC)
That's a highlight. I need to figure out why those don't show. Can you give me a save state at that point in the game please?
-
Ok, This game Violent Storm has a bad highlights also in the end of Stage 2 - train stage.
Save state for iq 132 :smilie:
-
"Break the Statue" text oddity: (savestate included)
To be fair, the text is broken in all the vids of the game I can find online, so its probably still broken in MAME as well..
-
Can't load Garou: Mark of Wolves.... Guru Meditation!
-
Can't load Garou: Mark of Wolves.... Guru Meditation!
Hi Freddy, thanks for the report, I found & fixed the problem!
best regards,
- dink
-
Two Konami games with Bugs confirmed in FBA....
Escape Kids fails on Boot
SPY Special Project Y Graphical Errors
-
FBA request from a friend: Hook-up sound to Donkey Kong 3.
Donkey Kong 3 has no sound, it uses 2xNES APU (2a03) which we have :)
best regards,
- dink
-
FBA request from a friend: Hook-up sound to Donkey Kong 3.
Donkey Kong 3 has no sound, it uses 2xNES APU (2a03) which we have :)
best regards,
- dink
Hooked up! Still buggy though. :S I think our nes apu core is a bit buggy (big surprise given who ported it)
-
great :) when I get some time I'll try to fix the nes apu for multi chips (if possible).
best regards,
- dink
-
spy.zip / Special Project Y / konami/d_spy.cpp, after changing over to the new loading routines, all the graphics have gone totally weird-like or missing.
-
spy.zip / Special Project Y / konami/d_spy.cpp, after changing over to the new loading routines, all the graphics have gone totally weird-like or missing.
Fixed.
-
三国志II: 三國英雄傳 ◦ Sangokushi II: Sanguo YingXiongZhuan (Chinese bootleg set 3, 921005 Asia) [sgyxz]
game freezes at this screen :
(http://i.imgur.com/E1ghhqJ.png)
-
Today added correct color PROMs set for Super Punch-Out!! (Japan) :
Before :
(http://i.imgur.com/DVS95KF.png)
After :
(http://i.imgur.com/PXl74Oj.png)
code :
// Super Punch-Out!! (Japan)
static struct BurnRomInfo spnchoutjRomDesc[] = {
{ "chs1c8la.bin", 0x2000, 0xdc2a592b, 1 | BRF_PRG | BRF_ESS }, // 0 Z80 Code
{ "chs1c8ka.bin", 0x2000, 0xce687182, 1 | BRF_PRG | BRF_ESS }, // 1
{ "chs1-c.8j", 0x2000, 0x1fa629e8, 1 | BRF_PRG | BRF_ESS }, // 2
{ "chs1-c.8h", 0x2000, 0x15a6c068, 1 | BRF_PRG | BRF_ESS }, // 3
{ "chs1c8fa.bin", 0x4000, 0xf745b5d5, 1 | BRF_PRG | BRF_ESS }, // 4
{ "chp1-c.4k", 0x2000, 0xcb6ef376, 2 | BRF_PRG | BRF_ESS }, // 5 N2A03 Code
{ "b_4c_01a.bin", 0x2000, 0xb017e1e9, 3 | BRF_GRA }, // 6 Top Screen Background Characters
{ "b_4d_01a.bin", 0x2000, 0xe3de9d18, 3 | BRF_GRA }, // 7
{ "chp1-b.4a", 0x2000, 0xc075f831, 4 | BRF_GRA }, // 8 Bottom Screen Background Characters
{ "chp1-b.4b", 0x2000, 0xc4cc2b5a, 4 | BRF_GRA }, // 9
{ "chs1-v.2r", 0x4000, 0xff33405d, 5 | BRF_GRA }, // 10 Big Sprite Characters
{ "chs1-v.2t", 0x4000, 0xf507818b, 5 | BRF_GRA }, // 11
{ "chs1-v.2u", 0x4000, 0x0995fc95, 5 | BRF_GRA }, // 12
{ "chs1-v.2v", 0x2000, 0xf44d9878, 5 | BRF_GRA }, // 13
{ "chs1-v.3r", 0x4000, 0x09570945, 5 | BRF_GRA }, // 14
{ "chs1-v.3t", 0x4000, 0x42c6861c, 5 | BRF_GRA }, // 15
{ "chs1-v.3u", 0x4000, 0xbf5d02dd, 5 | BRF_GRA }, // 16
{ "chs1-v.3v", 0x2000, 0x5673f4fc, 5 | BRF_GRA }, // 17
{ "chs1-v.4r", 0x4000, 0x8e155758, 5 | BRF_GRA }, // 18
{ "chs1-v.4t", 0x4000, 0xb4e43448, 5 | BRF_GRA }, // 19
{ "chs1-v.4u", 0x4000, 0x74e0d956, 5 | BRF_GRA }, // 20
{ "chp1-v.6p", 0x2000, 0x75be7aae, 6 | BRF_GRA }, // 21 Bottom Screen Big Sprite Characters
{ "chp1-v.6n", 0x2000, 0xdaf74de0, 6 | BRF_GRA }, // 22
{ "chp1-v.8p", 0x2000, 0x4cb7ea82, 6 | BRF_GRA }, // 23
{ "chp1-v.8n", 0x2000, 0x1c0d09aa, 6 | BRF_GRA }, // 24
{ "chs1-b-6e_white.6e", 0x0200, 0x8efd867f, 0 | BRF_GRA }, // 25 Color PROMS
{ "chs1-b-6f_white.6f", 0x0200, 0x279d6cbc, 0 | BRF_GRA }, // 26
{ "chs1-b-7f_white.7f", 0x0200, 0xcad6b7ad, 0 | BRF_GRA }, // 27
{ "chs1-b-7e_white.7e", 0x0200, 0x9e170f64, 7 | BRF_GRA }, // 28
{ "chs1-b-8e_white.8e", 0x0200, 0x3a2e333b, 7 | BRF_GRA }, // 29
{ "chs1-b-8f_white.8f", 0x0200, 0x1663eed7, 7 | BRF_GRA }, // 30
{ "chs1-v.2d", 0x0100, 0x71dc0d48, 0 | BRF_OPT }, // 31 Timing PROM
{ "chs1c6pa.bin", 0x4000, 0xd05fb730, 8 | BRF_SND }, // 32 VLM5030 Samples
{ "chs1-b-6e_pink.6e", 0x0200, 0x0ad4d727, 7 | BRF_OPT },
{ "chs1-b-6f_pink.6f", 0x0200, 0x86f5cfdb, 7 | BRF_OPT },
{ "chs1-b-7f_pink.7f", 0x0200, 0x8bd406f8, 7 | BRF_OPT },
{ "chs1-b-7e_pink.7e", 0x0200, 0x4c7e3a67, 0 | BRF_OPT },
{ "chs1-b-8e_pink.8e", 0x0200, 0xec659313, 0 | BRF_OPT },
{ "chs1-b-8f_pink.8f", 0x0200, 0x8b493c09, 0 | BRF_OPT },
};
STD_ROM_PICK(spnchoutj)
STD_ROM_FN(spnchoutj)
Sadly Punch-Out !! have some GFX glitches since it was added in FBA 0.2.97.30 :
(http://i.imgur.com/pIuZ0tg.png)
According to MAME, it seems Punch-Out!! has a different GFX loading routine than Punch-Out!! (Japan) :redface:
-
Punch-Out !! Fixed :smilie:
(http://i.imgur.com/UmTwfGB.png)
(http://i.imgur.com/xGgRWSq.png)
-
Punch-Out !! Fixed :smilie:
(http://i.imgur.com/UmTwfGB.png)
(http://i.imgur.com/xGgRWSq.png)
Nice work! (I meant to say that when you added the fix to the svn)
-
Thanks iq_132 :smilie:
Fix will be commit soon to SVN but before trying to add punchouta and spnchouta clones to the driver ;p
EDIT : Fix submitted to SVN with Punch-Out!! (Rev A) and Super Punch-Out!! (Rev A) added !!!
(http://i.imgur.com/HCYbQlX.png)
Punch-Out!! (Rev A)
(http://i.imgur.com/AV0sXFa.png)
Super Punch-Out!! (Rev A)
Best regards,
JacKc.
-
Thanks iq_132 :smilie:
Fix will be commit soon to SVN but before trying to add punchouta and spnchouta clones to the driver ;p
Nice work Jack. :)
-
Battle Shark (World) (bshark) - game freeze after stage 2.
Dragonball Z (rev B) (dbz)- game freeze
Testing on 64 bit version of FBA
-
Battle Shark (World) (bshark) - game freeze after stage 2.
Dragonball Z (rev B) (dbz)- game freeze
Testing on 64 bit version of FBA
Would you be able to give me save states for these (before the freezes happen?)
-
Would you be able to give me save states for these (before the freezes happen?)
Battle Sharks guaranteed to stall during the mission screen right after level 1 which is quite short i can do ya a savestate
if you want though
-
Battle Shark save on the mission screen it should freeze shortly afterwards
-
gamez fan, bad news re: that savestate, check your pm!
-
Hmm, something has gone horribly wrong with Battle Shark, its locking up now before the game even starts. weird, eh? :)
When I say lock up, I mean, the screen freezes or goes blank, and in the debug window, the cpu#1 spazzes out with "68K #1 Read byte => F00000" and it counts upwards.
Any ideas, guys?
EDIT: things are looking good - I got it to stop hanging! :D
best regards,
- dink
-
FBA 64 BIT version has bad, muffled sound compared to 32 bit - all games. :redface:
-
EDIT: nevermind, see note below!
Avengers (us set 1/parent) in d_lwings.cpp:
Right after beating the last boss, the congratulatory screen is kinda scrambled with the wrong image in FBA. I found a longplay video (by M.Janet) played in MAME, and the screen is OK. Need a little help with this one :)
p.s. after the savestate loads, wait for 2 screens to see what I mean.
From the MAME source:
Notes:
avengers061gre2: corrupted graphics in Avengers' ending not fixed.
This bug is not in the Japanese set "Buraiken".
It might just be a bug in the original: the tiles for the character
image are just not present in the US version, replaced by more tiles
for the title animation. The tile map ROM is the same between the two
versions.
EDIT2: more proof there's nothing wrong, here's a screencap of Buraiken's (Japanese Avengers) ending...
best regards,
- dink
-
well lets see,
Art of Fighting 2
Football Frenzy
are 2 games I noticed it in , it seems to happen at random in the emulator.
In AOF 2 it was when YOU WON or YOU LOST appeared on the screen and with the high pitched TING noise, the noise was crackly.
In Football Frenzy it happened when you were picking out a play.
This happened a few times before with other neogeo games with high pitched noises , cant seem to make it happen at will .
Sorry dink, I'm very bad at describing things well :(
Edit: I THINK I found the problem.........4 point 3rd order FM interpolation was on.........
I finally found the solution to this problem, sorry for taking so long!! Technical info: it didn't check for clipping in the first stage of the PSG mixing with interpolation on.
best regards,
- dink
-
Glitch for Dink- Bad priority sprites on Thunder Dragon 2
Fixed, thanks for reporting!
EDIT: bonus: Fixed the music up a bit, the lead instrument was missing!
best regards,
- dink
-
Thanks for fix. It's my favorite arcade shump all time. What do you think about my cpu AMD FX 9590, it's strong enough for 3D Games on Mame ?? :S
-
Thanks for fix. It's my favorite arcade shump all time. What do you think about my cpu AMD FX 9590, it's strong enough for 3D Games on Mame ?? :S
You're welcome! I stayed up until 3:40 in the morning fixing/playtesting and making it perfect for FBA last night, it was a fun time :)
I don't know about your cpu and mame, sorry! I just have a p4 2.99 ghz.....
best regards,
- dink
-
WiloWar Your PC should be strong enough to run most if not all emulators although Intel processors tend to run programs better than the AMD's. 8 cores really don't mean much specially if a emulator is set up to run on 1, 2 or 4 cores so they won't usually take advantage on what you currently have. The issue might not be your computer though. Mame is designed to run games in their original state if that is the right word which means no speed hacks or other modifications to run specially on older hardware . Usually newer the game or the more powerful the technology was used to run the game the more powerful the computer you will need to run it efficiently. At least what I have taken (I might be wrong) from using Mame the last 12 years Mame "has" games or will potentially have games that you will need a computer that is faster than what we have available today for them to run efficiently. I remember having a single core 1.2GHZ notebook a couple of years ago and I couldn't play a game like Air Gallet because it was too slow but I have since upgraded and the game plays at it's original speed.
-
WILOWAR, Turbo Force is A-OK now, thanks for reporting :)
best regards,
- dink
-
Hi folks. Need to fix Landstalker - treasure of king Nole. The game opens, but don't run!
-
Thx for fix. I played Turbo Force on Mame couple years ago and I think the orange alien form should be under the background stage 2 or maybe I mistake ??
-
Freddy try using a different Rom you might be using a bad one. That happens to me occasionally.
-
One long term bug squashed it seems when the new msm6295 sound core was ported it actually fixed the buzzing sounds
in the USA version of Power Instinct 2 it seems perfect now sound wise so sometimes a fix happens and ya dont know it ;)
but as is often the case sometimes another bug pops up Air Attack by comad in the nmk16.cpp driver now has broken sound
confirmed in FBA while using the new nmk16 sound bios file
-
Freddy, I tried to fix Landstalker, but I think some sort of protection mechanism is kicking in. I can't figure it out right now :( I will come back to it though!
gamezfan, Air Attack is nmk16, it doesn't use nmk004 but 112, the fix I made for Dolmen is what broke this one, I'll have it figured out by the next time we meet up.
best regards,
- dink
-
none of the official MD games have real protection, at most they fail because of missing backup ram.
Actually Pugsy is a bit of an exception, it has 'backup ram' listed in the header, but fails after a few levels if you actually emulate backup ram, that was done to confuse cart copiers.
Only the 3rd party / unlicensed games have actual real protections, although often weak ones.
-
Thanks Haze!!
Landstalker now works fine in FBA :)
best regards,
- dink
-
I'm very new to this entire process, so if this has already been resolved I apologize, but I'm trying to run FightCade on Vista (Service Pack 2) and keep getting Guru Meditation error #c0000005.00557161 any time I try to run ggpofba in any way (through FightCade solo, spectating, a match, outside the FightCade client) and the error report reads as follows:
FB Alpha v0.2.96.74 fatal exception report (Thu Dec 18 03:00:18 2014)
----------------------------------------------------------------------
Exception 0xC0000005 (EXCEPTION_ACCESS_VIOLATION) thrown.
EIP: 0x0055
(attempting to write address 0x00000000)
EAX: 0x00FFFFFF, EBX: 0x0012F990, ECX: 0x0012F988, EDX: 0x00000000
ESI: 0x00000000, EDI: 0x00000002, ESP: 0x0012F960, EBP: 0x00FFFFFF
----------------------------------------------------------------------
...this is very strange, because the ROMs, ROM path, and FightCade client all work perfectly. I also have a standalone GGPO client which runs ggpofba just fine in another folder. The ROMs work perfectly there and it's obvious my computer can run the ggpofba.exe client.
So, I'm kinda stumped as to why FightCade's version of ggpofba.exe and ggpofba-ng.exe that came with the original folder fail to launch at all. If anyone could help me, I'd really love to play some 3rd Strike on GGPO without having to go through the port forwarding process.
-
I have the version which came with the latest version of FightCade. Are they running old software? Can I replace that client with the one that works for me?
-
Found two bugs:
1) Spiderman vs the Kingpin: the game is running too fast.
2) Ayrton Senna's Super Monaco GP 2: Blue line flashing on the top of the sceen.
-
Freddy, thanks again for the bug reports! These 2 are giving me a bit of trouble at the moment, but I will keep trying..
Can I recommend to play the arcade version of Super Monaco GP (smgp.zip) for the time being?
best regards,
- dink
-
Don't worry, I'm just reporting the bugs! I played Super Monaco GP some months ago, good game! At the moment, in some free time, I started to play World Heroes Perfect and Megaman X. Thank's for suggestion, it's always recommended!
-
Tiger Road game have various graphical problems that make it unplayable, the main sprite is not visible and the background are corrupted too.
I use FB Alpha 0.2.97.33 under AmigaOS 4.1 but maybe the same problem are present in the original version aswell
-
Glitch for Dink - Some highlights are bad on Violent Storm (EAC)
Fixed!
-
Tiger Road game have various graphical problems that make it unplayable, the main sprite is not visible and the background are corrupted too.
I use FB Alpha 0.2.97.33 under AmigaOS 4.1 but maybe the same problem are present in the original version aswell
What levels, when? Can you provide a save state?
-
Nice iq! *cartman voice* we're hella cool now :D
-
Regarding those Tiger Road issues, there are a few places where transitioning between rooms causes a huge distortion on the screen, but it goes away pretty quick in the Win version of FBA - the game is fully playable though and doesn't seem to get as bad as the Amiga version. Maybe the problem is related?
best regards,
- dink
-
@iq_132
What levels, when? Can you provide a save state?
It happen at startup, so at the very first level .. game graphics are corrupted since beginning
Will check if i can provide a save state ..
@dink
Regarding those Tiger Road issues, there are a few places where transitioning between rooms causes a huge distortion on the screen, but it goes away pretty quick in the Win version of FBA - the game is fully playable though and doesn't seem to get as bad as the Amiga version. Maybe the problem is related?
Mmm no i don't think it's the case, i just start FB once and then i launch Tiger Road
-
Hmm, Tiger Road works perfect here, the screen-transistion garbage bug I talked about happens around level 3 I think. You might have to talk to the amiga-fba porter.
best regards,
- dink
-
One moe: Chakan is running fast and then slow down!
-
Thanks again to dink and iq_132 for the amazing job done on fixing bugs :smilie:
-
JacKc nice to see you are back :)
best regards,
- dink
-
Hello :cool:,
Yes Samo79
Please post on my forum for the all problems for fixing and add fix at the team FBA ;-)
Here:
http://www.clubevolution4.com/HunoPortSDL/modules/newbb/viewtopic.php?topic_id=117&start=0#forumpost796
This problem is endianess (i think)
I can not test all the games fba because I do not have time, and team FBA does not work on PowerPC machines it is difficult to see what kind of problems related to endianess after me I would undertake to provide them the patch if there because it is can be a problem with the SDK see an error on my part ;-)
Thank you for your understanding and thank you to the team for the team work carried out on this beautiful engine
HunoPPC
-
Yes Dink I manage and take into account the problem, then if I find a problem I'll post here the patch ;-)
HunoPPC
-
Hmm, Tiger Road works perfect here, the screen-transistion garbage bug I talked about happens around level 3 I think. You might have to talk to the amiga-fba porter.
best regards,
- dink
Fixed now ;-)
(http://www.clubevolution4.com/HunoSDL-Port/TIGERROADfix.png)
HunoPPC
-
Thanks HunoPPC, I added the patch to the main source tree.
best regards,
- dink
-
@HunoPPC
Please post on my forum for the all problems for fixing and add fix at the team FBA ;-)
Here:
http://www.clubevolution4.com/HunoPortSDL/modules/newbb/viewtopic.php?topic_id=117&start=0#forumpost796
Great, happy to have a unique place when add my Amiga reports !
And thanks for the fix in Tiger Road :-)
-
After testing it on both 029734 and the Debug
Blast Blade 2 NeogeoCD still wont load past the initial loading screen with the turtle on its back.
KOF 98 NGCD runs on the debug build though, crashes on 029734.
-
Missing blending effects in Riot
FB Alpha:
(https://sites.google.com/site/bisonsas/home/riot-01-03-120051.png)
MAME:
(https://sites.google.com/site/bisonsas/home/riot_latest_svn.png)
In opening demo...
-
Don't know if it's a general bug or not but i have found a little background problem on Garou
(tested on FB Alpha 0.2.97.34 under AmigaOS 4.1 PPC)
-
Hi samo79,
That background glitch doesn't appear to happen in the windows version of FBA.
best regards,
- dink
-
@dink
Thanks for the test, maybe it's another endianess issue ...
I reported the problem to HunoPPC :-)
-
Don't know if it's a general bug or not but i have found a little background problem on Garou
(tested on FB Alpha 0.2.97.34 under AmigaOS 4.1 PPC)
This bug is caused by prototype ROM in EURO and US regions. Play garoup in JAPANESE region to fix this.
-
I have the same problem also with garoup :redface:
-
Missing blending effects in Riot
FB Alpha:
(https://sites.google.com/site/bisonsas/home/riot-01-03-120051.png)
MAME:
(https://sites.google.com/site/bisonsas/home/riot_latest_svn.png)
In opening demo...
Fixed.
-
iq_132, Excellent!! Riot is definitely my favorite Cabal/Blood Bros. style game.
best regards,
- dink
-
Riot: while playing through the game, I noticed some sprite/tile clipping here and there. Here's a state where its the most noticable. (bottom right/train track)
-
Fixed.
nice work
-
Most of the last FBA builds complains about missing .dlls: D3D9.dll and the D3DX9_43.dll; both under my Windows XP (32 bits) and my Windows 8.1 (64 bits). Are they really missing or we are talking about some-kind of bug? When I click ok and ignore the message, FBA runs perfectly anyway.
-
Most of the last FBA builds complains about missing .dlls: D3D9.dll and the D3DX9_43.dll; both under my Windows XP (32 bits) and my Windows 8.1 (64 bits). Are they really missing or we are talking about some-kind of bug? When I click ok and ignore the message, FBA runs perfectly anyway.
download http://www.microsoft.com/en-us/download/details.aspx?id=8109 (This serves always^^)
-
Thank you a lot, 7zxkv! :biggrin: But in any case...
1- What exactly that .dlls do? Why does FBA work perfectly even without them?
2- If everyone suffers of the same problem, so, everyone must download these files separately? Is there a way to avoid such extra download?
3- The background image of the FBA for Windows XP is under very low resolution. This one seems to be a quite easy to solve.
4- I think that, in terms of organization, all the folders for roms of specific hardwares should go as subfolders of the "Roms" directory.
5- Such a wonderful surprise to see the upgraded "Game Info". As a suggestion, I could say that it would be nice to see two lines of tabs instead of one single line with row bar.
Regards and congratulations for the FBA Team! ;p
-
gg_alien3 and gg_alien3j miss game name.
-
gg_alien3 and gg_alien3j miss game name.
Hi money_114,
i guess it is missing for you because it is European fonts :
(http://i.imgur.com/k0o7GP4.jpg)
-
d_konami.cpp :
- G.I. Joe (World, EAB, set 1) & clones :
Game freezes at the end of stage 1 just before the first boss and/or during fight with it.
(http://i.imgur.com/JrWABh3.png)
(http://i.imgur.com/qyTVQoj.png)
NOTE: to be able to play this game, it is recommended to uncheck Use assembly MC68000 Core in Misc/Options
d_sidearms.cpp :
- Turtle Ship (Korea)
Title screen is totally borked
(http://i.imgur.com/z9LAyy0.png)
-
gg_alien3 and gg_alien3j miss game name.
Thanks for letting us know. I changed the description to standard ascii text to avoid localization issues.
-
Friendly PSA: The assembly 68k core has been pretty much phased out at this point - don't use it for playing games, its buggy. We just keep it around for driver development/debugging purposes.
-
Hi money_114,
i guess it is missing for you because it is European fonts :
(http://i.imgur.com/k0o7GP4.jpg)
Yes, it is character encoding issue. It displays null on my OS.
I hope it can be improved.
Some similar issue:
Landstalker - Die Sch鋞ze von K鰊ig Nolo
Stimpy's Invention Starring Starring Ren Ho雓 & Stimpy
-
Friendly PSA: The assembly 68k core has been pretty much phased out at this point - don't use it for playing games, its buggy. We just keep it around for driver development/debugging purposes.
The assembly core should really just go away.
-
"WOW New Fantasia" is upside down. :eek:
-
There are some Capcom, Kaneco, Sega and Taito games inside the "Misc (pre 90)". Should they go inside their respective hardware groups? Besides, there are a lot of Jaleco, Namco and Tecmo games there... perhaps it's time to create their own hardware groups too...
-
tail2nose is a konami game that rely on konamiic.h, it would make sense to have it in the konami folder (that's actually what i'm forced to do on the libretro-fba port)
-
Tail2nose might be using a konami tilegen chip, but it is a V-System game, though.
-
Some people at Pleasuredome have told me they had encounter a bug when scanning their 7zipped merged sets. This bug could be due to Hash collision options of CMP introduced in version 4.016 with pangb and pangb2 sets (merged into pang.7z) which have bb9.bin twice with different crcs (39f47a63 and 3a5883f5).
(http://i.imgur.com/QujMZhO.jpg)
On a side note, there is no problem with regular zipped splits sets.
-
I use 7zipped merged sets without issue.
Is it a torrent 7zip issue as I use standard 7zip?
-
JacKc please link a couple of these 7zipped romsets that cause fba to crash and I'll investigate the crash.
thanks,
- dink
-
Take a look at these dink:
https://www.mediafire.com/?syo0n70mvj0vg44
-
I encountered 2 issues with 0.2.97.35
played Last Resort on NCD and all SoundFX were missing once I got into playable game-mode
switching TOKI (arcade) from window to fullscreen and back resulted in messed up colors.
Anyways, great work!
Thanks!
-
Here is the latest report from kdin6tl3nm at The Pleasuredome, pursuant to my original reports of recent set breakage:
"I found something interesting. I think it explains why 19xx works and 1941 does not.
This is a two part situation.
The first part is that when clrmamepro makes a .7z archive through 7z.exe, it never creates a blank directory record. It only lists items in subfolders by the relative path. When 7z.exe is used in the GUI or command line, and there are subfolders, it always produces some entries that are marked a "D" (a directory) which have no size and no CRC. This is barely apparent in the GUI, but much more noticeable on the command line when using the "l" (list) command.
This means that the 56 block uncompressed 1941.z7 produced by clrmamepro has 56 items in it. If you create the same archive from the 7z.exe GUI or z7.exe command line (or t7z.exe or t7z.exe drag-n-drop), the resulting 1941.7z will have 59 items in it, as there are three directory items with zero length.
The second part is that these items are sorted alphabetically. Since the alphabetically sorted list for 19xx puts these folder items at the bottom of the list, FBA doesn't trip over them by the time it gets there. The alphabetically sorted list for 1941 puts the folder items the top, so FBA trips over them before it gets a chance to see any valid ROMs.
This isn't obvious at all in the GUI of 7z.exe, but is easily noticed when using the command line, such as:
7z.exe l 1941.7z
If someone could advise the FBA team that their .7z processing code is overlooking the fact that not all items in a .7z file are valid file items, it might help with this issue.
This theory explains why all .7z files created by the clrmamepro rebuilder work fine and why some of the t7z.exe processed files do not. The clrmamepro files do not have any folder (only) items listed, so there is no problem. T7z.exe processed items that have either no folder names or all the folder names are at the bottom will not confuse FBA, so no problem there either. But the t7z.exe processed items that have folder names appearing above the bottom of the list confuse the FBA .7z parser - which seems to be the problem."
-
Thanks for all the info, & files guys :) Barry fixed the issue!
-
switching TOKI (arcade) from window to fullscreen and back resulted in messed up colors.
Thanks!
Fixed! Thanks for the bug report.
-
Thank you, may we have a new build to test?
After that I will make noise at PD to get an FBA T7Z Merged set done, because now is the perfect time, esp. considering the recent MESS additions.
-
This is strange,
In the debug build of 0.2.97.35 Last Blade 2 NGCD worked and now in the official build, its back to being broke just as it was in the previous version.
-
Game: Soul Divide; Gunbird 2
Problem: Guru meditation in window mode, enhanced, xBR filter.
-
Hello
I got a bug , I cant play the new fba version
First I was using fba_029712 , then i download fba fba64_029736
So whan i double click the fba.exe archive i got this message below
How can i fix this?
FB Alpha v0.2.97.36 fatal exception report (Sun May 24 18:13:24 2015
)
----------------------------------------------------------------------
Exception 0xC0000005 (EXCEPTION_ACCESS_VIOLATION) thrown.
EIP: 0x0000
(attempting to read address 0xFFFFFFFFFFFFFFFF)
----------------------------------------------------------------------
System information:
OS: Microsoft Windows 8.1 (build 9600)
CPU: Intel(R) Core(TM) i5-3317U CPU @ 1.70GHz
Physical RAM: 4070680 KB (3975 MB) total, 1192784 KB (1164 MB) avail
Total RAM: 5643544 KB (5511 MB) total, 1666788 KB (1627 MB) avail
FB Alpha: 17916 KB in use (17916 KB peak, 23168 KB virtual)
Installed displays and display adapters:
Generic PnP Monitor on Intel(R) HD Graphics 4000 (primary)
----------------------------------------------------------------------
FB Alpha information:
Built on May 1 2015, 20:20:35, using Visual C++ 2013.
Optimised for X64 CPUs.
Using Unicode for all text.
Debug functionality absent.
MMX optimisations disabled.
Musashi emulation core enabled for MC680x0 family emulation.
Not emulating any game.
Video settings:
Interface settings: Video plugin not initialised
Audio settings:
Interface settings: Audio plugin not initialised
Input settings:
Selected module: DirectInput8 input
Interface settings: keyboard 0 System keyboard: Teclado
mouse 0 System mouse: Mouse
joystick 0 Generic USB Joystick
Profiling settings:
Interface settings: Profiling module not initialised
----------------------------------------------------------------------
Detailed process information about fba64.exe:
fba64.exe (base address 0x00007FF7EAD70000, size 36320 KB)
Exception occurred in module dsound.dll:
dsound.dll (base address 0x00007FFF62940000, size 628
Modules loaded by fba64.exe:
ntdll.dll (base address 0x00007FFF7DFB0000, size 1712
KERNEL32.DLL (base address 0x00007FFF7B630000, size 1272
KERNELBASE.dll (base address 0x00007FFF7B2D0000, size 1108
USER32.dll (base address 0x00007FFF7D7E0000, size 1500
GDI32.dll (base address 0x00007FFF7D410000, size 1348
COMDLG32.dll (base address 0x00007FFF7DB00000, size 728
COMCTL32.dll (base address 0x00007FFF781C0000, size 2540
SHELL32.dll (base address 0x00007FFF7BE80000, size 21600
WINMM.dll (base address 0x00007FFF6F430000, size 136
ole32.dll (base address 0x00007FFF7D960000, size 1616
ADVAPI32.dll (base address 0x00007FFF7DE40000, size 680
SETUPAPI.dll (base address 0x00007FFF7BC30000, size 1896
msvcrt.dll (base address 0x00007FFF7DEF0000, size 680
SHLWAPI.dll (base address 0x00007FFF7BBD0000, size 336
combase.dll (base address 0x00007FFF7B770000, size 2116
WINMMBASE.dll (base address 0x00007FFF6F400000, size 168
RPCRT4.dll (base address 0x00007FFF7B9A0000, size 1284
sechost.dll (base address 0x00007FFF7D6D0000, size 356
CFGMGR32.dll (base address 0x00007FFF7B280000, size 316
DEVOBJ.dll (base address 0x00007FFF79F20000, size 160
SHCORE.DLL (base address 0x00007FFF794C0000, size 712
IMM32.DLL (base address 0x00007FFF7D790000, size 216
MSCTF.dll (base address 0x00007FFF7D570000, size 1348
dsound.dll (base address 0x00007FFF62940000, size 628
POWRPROF.dll (base address 0x00007FFF7A750000, size 280
dinput8.dll (base address 0x00007FFF58BE0000, size 264
ddraw.dll (base address 0x00007FFF67EA0000, size 992
DCIMAN32.dll (base address 0x00007FFF694A0000, size 36
d3d9.dll (base address 0x00007FFF59990000, size 2180
VERSION.dll (base address 0x00007FFF74180000, size 40
dwmapi.dll (base address 0x00007FFF78FE0000, size 132
D3DX9_43.dll (base address 0x00007FFF513F0000, size 2444
uxtheme.dll (base address 0x00007FFF79D80000, size 1188
kernel.appcore.dll (base address 0x00007FFF79D30000, size 44
CRYPTBASE.dll (base address 0x00007FFF7B040000, size 44
bcryptPrimitives.dll (base address 0x00007FFF7AFD0000, size 396
HID.DLL (base address 0x00007FFF7A2C0000, size 52
WINTRUST.dll (base address 0x00007FFF7B3F0000, size 324
CRYPT32.dll (base address 0x00007FFF7B450000, size 1916
MSASN1.dll (base address 0x00007FFF7B1B0000, size 68
----------------------------------------------------------------------
-
Is there a way to allow auto-fire in the games: Aliens and 19xx: war against destiny?
-
Is there a way to allow auto-fire in the games: Aliens and 19xx: war against destiny?
For 19XX, it works the same way as all other games - press f5 after the game loads and scroll down near the bottom of the list and find "P1 Auto-Fire .." and set the autofire key(s). On the other hand, Aliens doesn't seem to like autofire because it has its own built-in autofire mechanism.
best regards,
- dink
-
Hey dink... you're gone brow, miss your news!
Here, in the last version of FBA, i can't set auto-fire in the 19xx. I tried, but doesnt works.
In Aliens, only the main weapon has the auto-fire set. The others we have to push the bottons repeatedly :redface:
* upgrade: in this version and can set the normal imput keys, but can't set auto-fire keys at all? any guess?
-
Hmm very strange, can you paste your 19xx.ini to http://pastebin.com and I'll take a look?
I'll see what can be done with aliens, thanks :)
There is no news because well, a few reasons. first, FBA is almost perfect now - the only bugs remaining are the ones that can't be fixed (as far as I can tell..). Another reason - there havn't been many bugs reported. But theres also some terrible stuff that happened regarding real life circumstances and things :/ [I don't really want to talk about it yet] I'm hoping that things clear up and normalize (regarding IRL conditions) by mid-summer - then perhaps the fun can resume..
best regards,
- dink
-
Hope you can overcome this situations and circumstances in the best way. I know some strange bugs in FBA, some of then I put here, but still remains and I don't know why. About the auto-fire macro, it's really strange, because I can't set in any games at all (begins in the "system pause" - above I can change the keys). Here the link: http://pastebin.com/fpsF50qT
-
Freddy, thanks, please try this: sometimes clicking the autofire macro doesn't work, you'll need to click in different places, then it will work. I don't know why. :)
-
The colors in Congo Bongo seem to be messed up with the Enhanced blitter, OK with basic. any ideas?
EDIT: fixed!
-
No dink, doesn't work here. Strange, because in the past I set some auto-fire keys.
By the way, look the game Turbo Outrun, has a strange bug and in the Asura Buster Eternal Warrior the game stuck in the jelly pink boss. Hum, Chakan has a problem in some part of the background stage, which goes very fast!
-
Freddy, the Asura pink boss lockup is known and unfixable at the moment. Could you give more information, pics and savestates regarding Chakan and Turbo Outrun?
thanks & best regards,
- dink
-
1) Chakan goes very fast in some parts of the screem (f.e. goes here):
(http://s15.postimg.org/lxj2wq0p3/chakan.jpg) (http://postimg.org/image/lxj2wq0p3/)
2)Ayrton Senna Monaco GP has some flashes on the screem (you need to play the game):
(http://s1.postimg.org/dp0ns2k4b/senna.jpg) (http://postimg.org/image/dp0ns2k4b/)
3) In Turbo Outrun, sometimes the road doesn't appears:
(http://s15.postimg.org/u89hialmv/outrun.jpg) (http://postimg.org/image/u89hialmv/)
4) Not a bug, but is good to change that: Zombies is not the main rom, it's a clone from the game "zombies ate my nighbors".
5) Can't set auto-fire buttons
Hope can help more!
-
Seems there was a bug in Spark Man basically players 1 and 2 were the same colour this was just fixed in MAME and since
i didn't have much on tonight ive added that fix to FBA. Simple copy n paste about my level really :D
(http://www.mameworld.info/ubbthreads/userfiles/341583-0000.png)
Note the above player 2 should be green
(http://www.fshare.it/image.jpeg?i=gl1434669r3)
static void draw_normal_sprites(INT32 which, INT32 new_style, INT32 gfxbank_type)
- draw_single_sprite(tile + (attr & 0x3)*0x100 + gfxbank + (which*0x8000), (((attr >> 2) & 0xf) | colorbank) + 0x10 * m_palettebank, sx, sy, tile_flipx, tile_flipy);
+ draw_single_sprite(tile + (attr & 0x3)*0x100 + gfxbank + (which*0x8000), (((attr >> 2) & 0xf) ^ colorbank) + 0x10 * m_palettebank, sx, sy, tile_flipx, tile_flipy);
-
Nice catch gamez fan :)
-
Freddy, unfortunately its now to the point where I fix one game in the Megadrive code, and it breaks another one. ugh. I'll keep trying, of course, but for the very few games affected - it doesn't look too good.
-
Freddy, unfortunately its now to the point where I fix one game in the Megadrive code, and it breaks another one. ugh. I'll keep trying, of course, but for the very few games affected - it doesn't look too good.
I thought that, because in old versions of the emulator don't have this bugs. Man, this codes are crazy, huh? Thank god that I just play the games! :biggrin:
-
Hello FBA team and game fans!
I don't know if it is a bug, but I can't allow continues in DoDonPachi DaiOuJou
-
Hello FBA team and game fans!
I don't know if it is a bug, but I can't allow continues in DoDonPachi DaiOuJou
Hi Freddy :)
Try this:
load game, Ctrl-F5 and set "Test Mode" ON and let game boot to the test screen.
Select "6. System", go down to "2. CONTINUE" and press joystick Right to toggle it ON
Press Ctrl-F5 again, and set "Test Mode" Off, and exit the test mode screens.
best regards,
- dink
-
@Freddy You may need to do the same with Esp-Galuda and Ketsui as like DDP3 the continue mode is set to off in those games
by default if i remember correctly
-
@dink Thank you! Worked fine! ;p
@gamez fan Yeah, in some games you can press F3 and configure the game. But in DDP DaiOuJou you need to set test mode "on". :smilie:
I have a suggestion for the FBA time. Lately I played FBA in Kaillera server. Sometimes I recommend the emulator to other players, but unfortunately the first impression is not so good. Maybe because of the standard configuration. For example, the pluggin audio option is set in 22050HZ and this frequency doesn't works in some games. I don't know why, but in the first run, the emulador goes very lagged in Enhanced blitter option. To remove the lag, I need to go in "blitter options" and them in "advanced setting" and turn off the follow option: "use directx texture manegement". Then, I turn on the same option and the lag is removed. Another thing, most of computer has the video configuration set in 1280x800, and the standard video in FBA is set in 640x480.
-
Thanks Freddy - I've been wanting to change these [bad] defaults for ages now, but I keep forgetting!
best regards,
- dink
-
Not really a bug more of an oversight ;)
The ThunderMAME samples for the in level music in ghox are all just short of 5 mins not enough to last the 6+ levels it's used for
they always run out before the boss encounters now depending on how quickly you finish the levels that could be 1 or 2 before
the boss it depends if your progress has been slowed by tricky to hit targets basically the music stops and your left with SFX only
Anyway in ThunderMAME ive confirmed the Samples for Ghox are set to autoloop so ive made this change in the d_ghox.cpp
FBA driver changing the samples to now Autoloop so at least your never without a tune in the game
-
Thanks Freddy - I've been wanting to change these [bad] defaults for ages now, but I keep forgetting!
best regards,
- dink
You have to be careful though.....
The D3D7 blitter is preferred - but not available in 64-bit builds.
xAudio2 is preferred - but IIRC isn't as well supported as DSound.
There are plenty of other caveats as well.
-
Treble Winner, I think the least we need to do is change the default samplerate to 44100hz for DSound. 22khz would have been OK in the mid/early-90s. As I can think of a few games that wouldn't even work at that rate, due to limitations in the sound core (after I wake up a bit more I can name them, too .. lol!)
best regards,
- dink
-
Treble Winner, I think the least we need to do is change the default samplerate to 44100hz for DSound. 22khz would have been OK in the mid/early-90s. As I can think of a few games that wouldn't even work at that rate, due to limitations in the sound core (after I wake up a bit more I can name them, too .. lol!)
best regards,
- dink
Completely agree. We should probably detect the users resolution and colour depth and set the default based on those too.
-
I couldn't see a thread search feature to avoid making a dupe post but here is what I just found.
Kabuki Klash Neogeo has a giant orange obstruction in the opening sequence on the right side of the screen.
-
Hi SNK_Dude, I tried it here with the latest gcc build and this obstruction doesn't appear on my side.
Do you still have the gcc build I had sent you several months ago? Is there any chance you could test with that build to see if it appears?
best regards,
- dink
-
Hi SNK_Dude, I tried it here with the latest gcc build and this obstruction doesn't appear on my side.
Do you still have the gcc build I had sent you several months ago? Is there any chance you could test with that build to see if it appears?
best regards,
- dink
Would that happen to be a debug XP build from the 1st of the year?
FB Alpha [DEBUG] v0.2.97.35
The picture is from the milestone release of 02.97.36
The bug does not appear in the debug build I have .
-
SNK_Dude, thanks for testing, the release builds are made with MSVC. I think it might be a compiler issue, Barry / Treble Winner would have to take a look into it since he's the only dev that uses it. The other issues you had been experiencing with neogeo-cd and such are probably related as well, since I can't reproduce them on my side :(
-
It works fine in the release builds (and it's easy to download a compiled binary to test :)).
For the record, keeping GCC working is a million times harder than MSVC as well. Look at the M68K issues we had with later GCC (and plenty of other issues). MSVC builds tend to be quicker across the board too.
I'm still trying to find the time to do a more modern blitter (by this time DX12) - but it will be a while before that is available in GCC/MinGW headers. Our blitters are hopelessly out of date now sadly. :(
-
Is there way for me to help pinpoint this issue or any future issues?
Any kind of log function or something?
-
It works fine in the release builds (and it's easy to download a compiled binary to test :)).
For the record, keeping GCC working is a million times harder than MSVC as well. Look at the M68K issues we had with later GCC (and plenty of other issues). MSVC builds tend to be quicker across the board too.
I'm still trying to find the time to do a more modern blitter (by this time DX12) - but it will be a while before that is available in GCC/MinGW headers. Our blitters are hopelessly out of date now sadly. :(
We have intf/video/opengl :rolleyes: It's OpenGL >3.3.
Something like MAME and RetroArch shader's would be great.
-
I think I've found a bug with internal rom paths searching engine. I've googled about it but I haven't found anything. I don't undrestand because it hasn't been reported before (?).
I have the full rom sets and the other games for consoles supported, organized in different directories (F:\MAME\roms\fba\coleco;F:\MAME\roms\fba\roms....). I have all the (seven) paths for the differrent system correctly configured in fb (with final slash ant the correct paths).
The problem is related with roms with same zip name for the different systems (i.e. frogger) configured. After running a rom scan only the game found in the first configured path (before the rest), with an same zip rom name, is found for that system. The other same zip roms names are reported as not found for the rest of the systems. Also trying to play that game (not the one for the first system found), it's again reported that the game it's not found (I think that it's trying to read the files from the first zip found with that name, that it hasn't to be the right zip for that system).
I think this is a bug, I haven't found a solution yet. I've found same the problem for both fba.exe and fba64.exe. I've tested it in a Windows 7 x64 system.
-
I think I've found a bug with internal rom paths searching engine. I've googled about it but I haven't found anything. I don't undrestand because it hasn't been reported before (?).
I have the full rom sets and the other games for consoles supported, organized in different directories (F:\MAME\roms\fba\coleco;F:\MAME\roms\fba\roms....). I have all the (seven) paths for the differrent system correctly configured in fb (with final slash ant the correct paths).
The problem is related with roms with same zip name for the different systems (i.e. frogger) configured. After running a rom scan only the game found in the first configured path (before the rest), with an same zip rom name, is found for that system. The other same zip roms names are reported as not found for the rest of the systems. Also trying to play that game (not the one for the first system found), it's again reported that the game it's not found (I think that it's trying to read the files from the first zip found with that name, that it hasn't to be the right zip for that system).
I think this is a bug, I haven't found a solution yet. I've found same the problem for both fba.exe and fba64.exe. I've tested it in a Windows 7 x64 system.
Not a bug - check the Rom Path Dialog entry in the chm manual included or on the site - http://www.fbalpha.com/manual/
-
Not a bug - check the Rom Path Dialog entry in the chm manual included or on the site - http://www.fbalpha.com/manual/
Ok, thank you, I didn't know about this special managing system of ROM paths in Final Burn.
-
Ok, thank you, I didn't know about this special managing system of ROM paths in Final Burn.
Don't feel bad, I just found out about them a few months ago :)
best regards,
- dink
-
Lethal Enforcers (ver EAB, 10/14/92 19:53) (lethalenub) - graphic mess
-
roms were working normally, until suddenly halfway trough loading ANY NeoGeo game I get
Exception 0xC0000005 (EXCEPTION_ACCESS_VIOLATION) thrown.
EIP: 0x0044
(attempting to read address 0x1AED2AC8)
EAX: 0x0AC40020, EBX: 0x00000000, ECX: 0x10292AA8, EDX: 0x02723F08
ESI: 0x0271ED68, EDI: 0x00000000, ESP: 0x0251CEFC, EBP: 0x0251CF04
-
sqxrz2: would you go into the folder with fba*.exe, then into the config sub-folder and rename (don't delete) fba*.ini to fbabroke.ini, then try again with a fresh config? If this does solve the problem, could you please link a copy of fbabroke.ini so I can investigate?
thanks,
- dink
-
Dont know who last worked on lastduel.cpp ive noticed a wee mistake in one of the Rom Defines no big deal but shouldn't Led Storm be like this.......
struct BurnDriver BurnDrvLedstorm = {
"ledstorm", "madgear", NULL, NULL, "1988",
"Led Storm (US)\0", NULL, "Capcom", "Miscellaneous",
NULL, NULL, NULL, NULL,
BDF_GAME_WORKING | BDF_CLONE | BDF_ORIENTATION_VERTICAL, 2, HARDWARE_MISC_PRE90S, GBF_RACING, 0,
NULL, DrvuRomInfo, DrvuRomName, NULL, NULL, DrvInputInfo, DrvDIPInfo,
DrvInit, MadgearExit, DrvFrame, NULL, MadgearScan,
NULL, 0x800, 240, 384, 3, 4
};
Rather than Madgearu as there is no such version
struct BurnDriver BurnDrvMadgearu = {
"ledstorm", "madgear", NULL, NULL, "1988",
"Led Storm (US)\0", NULL, "Capcom", "Miscellaneous",
NULL, NULL, NULL, NULL,
BDF_GAME_WORKING | BDF_CLONE | BDF_ORIENTATION_VERTICAL, 2, HARDWARE_MISC_PRE90S, GBF_RACING, 0,
NULL, DrvuRomInfo, DrvuRomName, NULL, NULL, DrvInputInfo, DrvDIPInfo,
DrvInit, MadgearExit, DrvFrame, NULL, MadgearScan,
NULL, 0x800, 240, 384, 3, 4
};
-
the BurnDriver struct name can be anything really, as long as the data inside it links to the right roms and such.
-
Maybe I spotted two "bugs":
- Kaitei Takara Sagashi (kaitei) starts correctly, but it freezes when the gameplay begins... no problem with the "Namco license" version (kaitein).
- Megadrive's Mickey Mania starts correctly, but it freezes when a demo or gameplay section begins (same issue for all versions: md_mmania, md_mmaniaj, md_mmaniau and md_mmaniaup).
-
Maybe I spotted two "bugs":
- Kaitei Takara Sagashi (kaitei) starts correctly, but it freezes when the gameplay begins... no problem with the "Namco license" version (kaitein).
- Megadrive's Mickey Mania starts correctly, but it freezes when a demo or gameplay section begins (same issue for all versions: md_mmania, md_mmaniaj, md_mmaniau and md_mmaniaup).
Thanks,
For kaitei, see the "Notes:" in the rom info box at the bottom of the screen when the game is highlighted :)
Regarding MD games with issues, the MD driver is at its limit for which fixes can be made. Fixing one game breaks another every time. But luckily, most games work good, its just a handfull that don't. Some day I would like to take out our Megadrive driver and replace it with Charles MacDonald's brilliant Genesis Plus emulator code. Either that, or I might try to create alternate frame-code selectable via dip (mega-hack).
thanks and best regards,
- dink
-
Thanks,
For kaitei, see the "Notes:" in the rom info box at the bottom of the screen when the game is highlighted :)
Ops... sorry for the "false alarm"... I was a little absent-minded and I didn't read the notes: "freezes up, use kaitein instead!" :p
Regarding MD games with issues, the MD driver is at its limit for which fixes can be made. Fixing one game breaks another every time. But luckily, most games work good, its just a handfull that don't. Some day I would like to take out our Megadrive driver and replace it with Charles MacDonald's brilliant Genesis Plus emulator code. Either that, or I might try to create alternate frame-code selectable via dip (mega-hack).
Obviously, there's no hurry... but in a next FBA release I will be happy to see an improvement of Megadrive driver, perhaps with the Mega CD/SEGA CD support! :biggrin: (just kidding, I'm really a "bad guy"!) :p
thanks and best regards,
- dink
Thanks to you for the excellent work! :)
-
I noticed a "strange thing"...
if I load Dig Dug (digdug) and then I try to load another game (I tried with various titles) the emulator crashes... same thing if I try with the option "Exit game".
For test I used FBA 0.2.97.37 32 bit on Win7 64 bit.
-
Gab75, hmm, I can't seem to make problem happen here with the release build or my development/gcc build of FBA. will you please attach your fba*.ini in "fba folder"\config so I can see if it might be due to a certain setting? You'll have to zip it first.
(generally I need to reproduce the problem on my side to be able to fix a problem like this)
thanks,
- dink
-
Gab75, hmm, I can't seem to make problem happen here with the release build or my development/gcc build of FBA. will you please attach your fba*.ini in "fba folder"\config so I can see if it might be due to a certain setting? You'll have to zip it first.
(generally I need to reproduce the problem on my side to be able to fix a problem like this)
Of course! hereunder my fba.ini file in zipped format... :wink:
-
Hi, please see your PM - I sent a debug version of FBA, unzip/copy the included exe over your current FBA exe and try digdug. When it crashes, please tell me what the guru message says. (screencap if possible).
thanks,
- dink
-
Hi, please see your PM - I sent a debug version of FBA, unzip/copy the included exe over your current FBA exe and try digdug. When it crashes, please tell me what the guru message says. (screencap if possible).
Hi dink,
I sent you a PM with the screen capture of "crash event"... but the guru message doesn't say much...
-
PS: I tried to use Dig Dug Rom with FBA 64 bit, but nothing is changed with respect to FBA 32 bit...
-
I need a screen capture of the guru message, including all the numbers! :)
-
I need a screen capture of the guru message, including all the numbers! :)
Ok, I sent you all guru message infos by PM ! :)
-
Has Haunted Castle been reported yet?
The sound (BGM) in it got borked in the latest FBA, in the previous build it was accurate.
-
I noticed a "strange thing"...
if I load Dig Dug (digdug) and then I try to load another game (I tried with various titles) the emulator crashes... same thing if I try with the option "Exit game".
For test I used FBA 0.2.97.37 32 bit on Win7 64 bit.
The "bug" was related to an issue in the sound core... dink fixed it, great! ;)
PS: in addition to Dig Dug, (IIRC) the issue was also in Galaga, Pac-man, Pac-land, Pengo and Rally-X.
-
is this bug: [http://imgur.com/m0O4n07 (http://imgur.com/m0O4n07)]
Puck Man (Japan set 1)
puckman
Namco (1980, Pac-man hardware)
its "upside down" even the control is also inverted.
-
is this bug: [http://imgur.com/m0O4n07 (http://imgur.com/m0O4n07)]
its "upside down" even the control is also inverted.
Use a different video blitter.
-
Use a different video blitter.
thanks! it works on softFX and directX9.
-
SNK Main Event (mainsnk) doesn't work... black screen after the boot...
In SNK TouchDown Fever (all versions) the "rotate vertically aligment games" option doesn't work properly... in few words, the positioning of virtual screen doesn't rotate... update: moreover (as far as I could ascertain) the game controls don't work, it's impossible move the players or make a choice in the menu screens.
-
Use a different video blitter.
Or use the option to force flip on the DDraw blitter using software rather than GPU.
We should really remove the DDraw blitter though. :)
-
In SNK TouchDown Fever (all versions) the game controls don't work, it's impossible move the players or make a choice in the menu screens.
TD Fever and Fighting Soccer need the controls for both games hooked up in the driver it might get done at some point or maybe
not either way there's a few games like this in FBA maybe they need to be marked with the debug flag in the driver untill such time
as they're fully playable..???
-
notice a little graphic issue in grdians since it was added to FBA (this issue is not in MAME)
(http://i.imgur.com/h8bz0h5.png)
-
I'm pretty sure that is protection-related. We either need to implement the protection or disable the game.
-
I'm pretty sure that is protection-related. We either need to implement the protection or disable the game.
iq_132, the game actually works perfectly except for that part. its just a little transitional piece from before the game starts, and doesn't matter or detract from the game at all - no need to disable this great game :)
-
Except for the bug reports about it every few months. :/
-
I'll mark the game notes: "graphic issue @ game start". When someone reports the bug, reply "read the game notes" ;)
-
I think it is better to mark as game notes "imperfect protection emulation @ game start" and keep it as working.
So for people (like me) reporting every few months this little graphic issue (which seems to not affect the playability of the game), this issue will be known and the devs also will know that something in the future can be improved regarding protection emulation (and maybe improve protection for others games in d_seta2.cpp)
http://git.redump.net/mame/tree/src/mame/drivers/seta2.cpp
grdians:
- the map screen after the character selection needs zooming. There is a global
zoom register that should affect the background map and the level picture but
not the frontmost frame. This latter should use color 7ff (the last one) and
ignore the individual color codes in the tiles data. Note: the frontmost frame
has the shadow bit set, and has become invisible after implementing it.
-
The reason I don't think it should be marked as such is that it has nothing to do with protection. "Unemulated graphic mode @ game start" or something like graphic glitch @ game start is better suiting IMO. I got it, how about "Imperfect graphics @ game start cut-scene", that makes sense and would not confuse our fans.
best regards,
- dink
I think it is better to mark as game notes "imperfect protection emulation @ game start" and keep it as working.
So for people (like me) reporting every few months this little graphic issue (which seems to not affect the playability of the game), this issue will be known and the devs also will know that something in the future can be improved regarding protection emulation (and maybe improve protection for others games in d_seta2.cpp)
http://git.redump.net/mame/tree/src/mame/drivers/seta2.cpp
-
While I was working on the Game Gear PNGs I noticed that three games (operating in SMS mode) had issues with the screen size and the colors scheme... I suppose the bugs are connected to the different video resolution and colors indexing between Master System and Game Gear (SMS 256x192px and 6 bit RGB palette, GG 160x144px and 12 bit RGB palette).
The "implicated" games are:
the Excellent Dizzy Collection (gg_excdizzy, gg_excdizzyp);
Street Hero (gg_shero, gg_sheroa);
Super Tetris (gg_stetris).
Attached two pics as example...
-
Thanks, they'll be fixed in the next version ;)
-
While I was testing some ROMs I noticed the following bugs:
1) In the Data East Mutant Fighter intro (all versions) the image of colosseum appears partially corrupted (tearing in the upper side), probably an issue with the scrolling of background clouds;
2) the graphic tiles of SEGA Heavyweight Champ "decrypted" (hwchampjd) are totally corrupted.
Attached two "explanatory" pics...
-
These "issues" are not related to the games emulation, but to the translation template...
while I was working on the UI translation I noticed that the voice "state: nothing to undo" hasn't a counterpart in the translation template. Moreover the voices "Angle" and "Fishbowl Factor" (3D projection properties) remain in english language indipendently from the UI language template.
-
While I was testing some ROMs I noticed the following bugs:
...
2) the graphic tiles of SEGA Heavyweight Champ "decrypted" (hwchampjd) are totally corrupted.
Attached two "explanatory" pics...
2) It has been fixed. A simple copy/paste error when syncing FBA with MAME.
-
Maybe I found a little bug in Kaneko's Gals Panic 4 (galpani4) and Gals Panic S (galpanis).
1) Gals Panic 4: in the "continue screen" the countdown numbers seem to have an imperfect adherence of the sprites during the scaling-in effect (see the attached png)
2) Gals Panic S: similar issue with the scaling-out effect of the "boss" explosion (see the attached png)
-
The zooming code in the suprnova driver has issues, every few zooms it puts in a transparent line (or lines) due to some scaling maths being off. I tried once or twice to fix it with no luck, maybe iq_132 has an idea? :)
notes for debug / 100% recreate: load vs. block breaker, after bootup and the insert coins screen it will have a funny screen with a monkey and hamster on each side of the screen, one squishing the other (causing them to scale), hitting pause during this and f1 several times will sooner or later cause the lines I'm talking about.
best regards,
- dink
-
notes for debug / 100% recreate: load vs. block breaker, after bootup and the insert coins screen it will have a funny screen with a monkey and hamster on each side of the screen, one squishing the other (causing them to scale), hitting pause during this and f1 several times will sooner or later cause the lines I'm talking about.
The attached snapshots are esplicative of the "issue" that you are talking about!
-
The zooming code in the suprnova driver has issues, every few zooms it puts in a transparent line (or lines) due to some scaling maths being off. I tried once or twice to fix it with no luck, maybe iq_132 has an idea? :)
Maybe I'm wrong, but, through a visual inspection, the issue seems to be related with an imperfect sprites/objects scaling synchronization... in other words, sometimes, during the scaling, the sprites don't stick among them. An example can be give by the four sprites/objects that they assemble each countdown number; to be more explicit I attached two snapshots, I edited the second one to match the sprites (the green lines highlight the changes of position).
-
Maybe I'm wrong, but, through a visual inspection, the issue seems to be related with an imperfect sprites/objects scaling synchronization... in other words, sometimes, during the scaling, the sprites don't stick among them. An example can be give by the four sprites/objects that they assemble each countdown number; to be more explicit I attached two snapshots, I edited the second one to match the sprites (the green lines highlight the changes of position).
That's exactly what is happening. The math for sprite zooming is usually a little fuzzy and can be off like this when the rounding isn't quite correct. I may look at this later... or I could always just start banging my head on my keyboard now.
-
... or I could always just start banging my head on my keyboard now.
Probably, in this case, the edges of your head and keyboard will match perfectly, but the issue with the Supernova scaling code would remain unchanged !!! :P
-
I take this opportunity to "highlight" two Sega Super Hang-on (all ROMs) bugs:
1) the exhaust flames (overboost mode) and tires screeching effects are "audible", but not visible;
2) the title screen freezes after few seconds, stop of music/sound (if present) and video animation. In any case the game is not really freezed, indeed if you use P1 coin or P2 coin inputs it's possible to begin a game session.
-
I came up with a fix for the graphics issues, but the hang @ the titlescreen is still present. Maybe Barry has an idea for this one? (I tried tweaking the timing & irq's... no luck...)
best regards,
- dink
I take this opportunity to "highlight" two Sega Super Hang-on (all ROMs) bugs:
1) the exhaust flames (overboost mode) and tires screeching effects are "audible", but not visible;
2) the title screen freezes after few seconds, stop of music/sound (if present) and video animation. In any case the game is not really freezed, indeed if you use P1 coin or P2 coin inputs it's possible to begin a game session.
-
I came up with a fix for the graphics issues, but the hang @ the titlescreen is still present. Maybe Barry has an idea for this one? (I tried tweaking the timing & irq's... no luck...)
Thanks! :) I'm glad to know that the "first point" is solved ! The exhaust flames and screeching tire are "small" visual effects, but (IMHO) these are the "little things" that make the difference! ;)
-
Dragon Egg (pce_dragnegg) - game freeze on start
-
While I was testing SEGA Out Run I noticed that the bgm playback is too slow (all ROMs)... (about half of normal speed).
Moreover Out Run Japan Deluxe sitdown (outrundxj) doesn't run (black screen on start-up) and the musics in the "car audio" select menu are audible only in the Out Run Enhanced Edition (outruneh)... I don't know if this last point is "normal".
-
That was a side effect of debug code left in from debugging super hangon, sorry about that - nothing to see here :)
p.s. those aren't black helicopters flying over you right now, either. ;)
best regards,
- dink
-
Thanks for the clarification! :)
-
Hello
when i was playing C.O.W Boys of moo mesa. There is a sound problem. İn the 2nd stage when the train door open the girl is screaming is '' HELP, HELP'' there is a sound problem. I checked at the mame, but it was good wasnt problem
thanks....
-
Hello
when i was playing C.O.W Boys of moo mesa. There is a sound problem. İn the 2nd stage when the train door open the girl is screaming is '' HELP, HELP'' there is a sound problem. I checked at the mame, but it was good wasnt problem
thanks....
I just tried in the latest FBA, and it sounds exactly the same as MAME. Some things to check though: Moo Mesa's sound chip operates internally at 48khz, so if you have the your samplerate under 44100hz there might be problems. Check Audio -> Plugin Options -> and make sure 44100 or 48000hz is checked (depending on what you want / what your soundcard can handle) and let me know if that makes a difference.
best regards,
- dink
-
Hello
i tried at 48000 and still same.
-
Hursit, what does it sound like? Can you make a recording?
-
Hello, I just wanted to report a bug on Soldier Blade for the Turbografx 16 and the PC Engine version. The game freezes on the first level after the mid level boss. Thank You and have a nice day.
-
BaconIsGood16, thanks for letting us know!
best regards,
- dink
-
When running Fatal Fury 2 on the Neo Geo, if you LOAD any saved states it causes the game to put the clock always on "00" into the next round (if you load before Round 2 stars) making you losing your credit after that or the attacks don't hit the enemy (if you load before Round 1 starts), making only the grapping moves available.
Already tested several different ROMs and this bug happened with all of them, including different BIOS from different regions. That doesn't happen with Fatal Fury Special, only with Fatal Fury 2.
-
Noticed some graphical problems with Tail 2 Nose it does use Konami Gfx so i guess it got broke with the Konami changes of late
as you can see no BG when playing the game.
-
Thanks, Tail2Nose is fixed now.
Also another issue in .39 release, Konami's The Main Event was accidentally left out. sorry guys!
best regards,
- dink
-
Charlie Ninja (Simple156.cpp) seems to have some dip/service related issues in FBA it's defaulting to..........
Coin 1 3 coin 1 credit
Coin 2 3 coin 1 credit
Screen Rotation On (Although the Flip doesn't seem to be active)
Demo Sound Off
Continue Off
Game Level V Hard
Player Level 1
Life Number 2
Title Mode Japan
When it should be according to how it is in MAME
Coin 1 1 coin 1 credit
Coin 2 1 coin 1 credit
Screen Rotation Off
Demo Sound On
Continue On
Game Level Normal
Player Level 3
Life Number 3
Title Mode English
All the other games in the driver are correct for their dip/service mode default settings, obviously these settings for Charlie Ninja
are wrong and making the game play less enjoyable in FBA.
-
thanks, its fixed.
best regards,
- dink
-
thanks, its fixed.
best regards,
- dink
Good job big man cheers!!!
-
Confirmed in latest FBA i guess a fix done for was it Bonk's Adventure..??? cant mind must have broke Great Thousand Mile Rally 2
-
GTMR2 is fixed now. Also while hunting this bug down I found another one which caused a crash when exiting any kaneko16 game which didn't use nvram when a kaneko16 game with nvram was loaded before it. eek :)
best regards,
- dink
-
GTMR2 is fixed now. Also while hunting this bug down I found another one which caused a crash when exiting any kaneko16 game which didn't use nvram when a kaneko16 game with nvram was loaded before it. eek :)
best regards,
- dink
Good job and thanks for taking the time to sort it cheers!!!
-
Some days ago mafer asked for an issue with the "single PCB versions" of Demon Front and SVG, they need the bios files (pgm_t01s.rom and pgm_m01s.rom) inside the zipped ROM to work properly... but in the "FB Alpha ClrMame" dat file there's an apparent contradiction: if I understood correctly, the pgm bios files should be merged and not directly in the zipped ROM... maybe does the dat file need an adjustment !?
Demon Front (V107, Korea, Single PCB Version)
<game name="dmnfrntpcb" cloneof="dmnfrnt" romof="dmnfrnt">
<description>Demon Front (V107, Korea, Single PCB Version) [Insert coin to get past ERROR]</description>
<year>2002</year>
<manufacturer>IGS</manufacturer>
<rom name="demonfront_v107-u43.bin" size="2097152" crc="671d8a31"/>
<rom name="t04501.u29" merge="t04501.u29" size="8388608" crc="900eaaac"/>
<rom name="a04501.u3" merge="a04501.u3" size="8388608" crc="9741bea6"/>
<rom name="a04502.u4" merge="a04502.u4" size="8388608" crc="e104f405"/>
<rom name="a04503.u6" merge="a04503.u6" size="8388608" crc="bfd5cfe3"/>
<rom name="b04501.u9" merge="b04501.u9" size="8388608" crc="29320b7d"/>
<rom name="b04502.u11" merge="b04502.u11" size="2097152" crc="578c00e9"/>
<rom name="w04501.u5" merge="w04501.u5" size="8388608" crc="3ab58137"/>
<rom name="dmnfrnt_igs027a.bin" merge="dmnfrnt_igs027a.bin" size="16384" status="nodump"/>
<rom name="demonfront_v107-u62.bin" size="4194304" crc="cb94772e"/>
<rom name="pgm_t01s.rom" merge="pgm_t01s.rom" size="2097152" crc="1a7123a0"/>
<rom name="pgm_m01s.rom" merge="pgm_m01s.rom" size="2097152" crc="45ae7159"/>
<rom name="pgm_p02s.u42" size="131072" crc="78c15fa2"/>
S.V.G. - Spectral vs Generation (V100, Japan, Single PCB Version)
<game name="svgpcb" cloneof="svg" romof="svg">
<description>S.V.G. - Spectral vs Generation (V100, Japan, Single PCB Version)</description>
<year>2005</year>
<manufacturer>IGS</manufacturer>
<rom name="svg_v100jp.u50" size="524288" crc="8d0405e4"/>
<rom name="t05601w016.bin" merge="t05601w016.bin" size="2097152" crc="03e110dc"/>
<rom name="a05601w064.bin" merge="a05601w064.bin" size="8388608" crc="ea6453e4"/>
<rom name="a05602w064.bin" merge="a05602w064.bin" size="8388608" crc="6d00621b"/>
<rom name="a05603w064.bin" merge="a05603w064.bin" size="8388608" crc="7b71c64f"/>
<rom name="a05604w032.bin" merge="a05604w032.bin" size="4194304" crc="9452a567"/>
<rom name="b05601w064.bin" merge="b05601w064.bin" size="8388608" crc="35c0a489"/>
<rom name="b05602w064.bin" merge="b05602w064.bin" size="8388608" crc="8aad3f85"/>
<rom name="w05601b064.bin" merge="w05601b064.bin" size="8388608" crc="bfe61a71"/>
<rom name="w05602b032.bin" merge="w05602b032.bin" size="4194304" crc="0685166d"/>
<rom name="svgpcb_igs027a_execute_only_area" size="392" status="nodump"/>
<rom name="svgcpb_igs027a_v100_japan.bin" size="15992" crc="7a59da5d"/>
<rom name="svg_v100jp.u64" size="4194304" crc="399d4a8b"/>
<rom name="svg_v100jp.u65" size="4194304" crc="6e1c33b1"/>
<rom name="pgm_t01s.rom" merge="pgm_t01s.rom" size="2097152" crc="1a7123a0"/>
<rom name="pgm_m01s.rom" merge="pgm_m01s.rom" size="2097152" crc="45ae7159"/>
<rom name="svg_bios.u49" size="131072" crc="3346401f"/>
-
FBA changing the image format of all the games.
FBA plays all games in 4: 4 format (Whereas games are in 4: 3 format).
Also Street Fighter 2 is a game in widescreen format (Street Fighter 2 is a little narrow format than 16: 9).
But Even FBA playing Street Fighter 2 game in 4: 4 format.
We want to play in the original format of all the games.
-
FBA changing the image format of all the games.
FBA plays all games in 4: 4 format (Whereas games are in 4: 3 format).
Also Street Fighter 2 is a game in widescreen format (Street Fighter 2 is a little narrow format than 16: 9).
But Even FBA playing Street Fighter 2 game in 4: 4 format.
We want to play in the original format of all the games.
Street Fighter 2 is intended to be rendered as 4:3, it is NOT A WIDE SCREEN GAME
multiple sources within Capcom have confirmed that, even if it's just common sense.
99.99% of arcade games from that period are 4:3, there are only a tiny number of exceptions and those were all games with large scale dedicated deluxe cabinets like Virtua Racing.
The only traditional Street Fighter series entry (aside the modern games) to have a wide-screen mode is Street Fighter 3 - 2nd Impact, where the option can be turned on in service mode. That option wasn't even included in 3rd Strike as it wasn't popular enough.
-
FBA changing the image format of all the games.
FBA plays all games in 4: 4 format (Whereas games are in 4: 3 format).
Also Street Fighter 2 is a game in widescreen format (Street Fighter 2 is a little narrow format than 16: 9).
But Even FBA playing Street Fighter 2 game in 4: 4 format.
If you select the right settings in the "Video" menu ("Stretch" and "Monitor properties") the games should be emulated in the correct aspect ratio/format.
We want to play in the original format of all the games.
Here's a "tip" for next time... the verb "to wish" is better than the verb "to want" ! ;)
-
Hi, friends.
I know thats a known issue, but I wanted to mark that Asura Busters (asurabus.zip) is still freezing at first boss.
Is there any cause? I've made a query but it didn't show any result about the reason.
I like this game, but is a pitty to not be able to play and complete it.
-
Hi, friends.
I know thats a known issue, but I wanted to mark that Asura Busters (asurabus.zip) is still freezing at first boss.
Is there any cause? I've made a query but it didn't show any result about the reason.
I like this game, but is a pitty to not be able to play and complete it.
we know about this, it's the legendary 68*020 bug that is keeping us from porting taito f3. I've been doing some research on this issue the past couple of days and hopefully will have a cure some day.
best regards,
- dink
-
Thanks, dink, for the info.
I know you'll find the solution, no matter if sooner or later :)
I wanted to tell you that each time I read your posts, as individuals or as FBA team, all of you have been getting deeper and deeper into the code, solving the hardest cases, with passion and always doing the best for everyone.
Don't know.. It's what I feel from this side.
For me you've grown as formidable developers and I wanted to thank you all!
Cheers! :)
-
Thanks, dink, for the info.
I know you'll find the solution, no matter if sooner or later :)
I wanted to tell you that each time I read your posts, as individuals or as FBA team, all of you have been getting deeper and deeper into the code, solving the hardest cases, with passion and always doing the best for everyone.
Don't know.. It's what I feel from this side.
For me you've grown as formidable developers and I wanted to thank you all!
Cheers! :)
+1
Thanks!
-
Thanks guys!
Here's the current progress with Asura Buster, you'll have to watch this video :)
https://www.youtube.com/watch?v=RMbYiFMupBY
best regards,
- dink
-
Here's the current progress with Asura Buster, you'll have to watch this video :)
https://www.youtube.com/watch?v=RMbYiFMupBY
Excellent work! :)
PS: are you making progress with 68*020 core?
-
getting close to having it fixed. Dream World and Asura BUster are fixed. We're about 3/4 the way there to a fixed 020 :)
-
getting close to having it fixed. Dream World and Asura BUster are fixed. We're about 3/4 the way there to a fixed 020 :)
Great news! Now Taito F3 driver can be seen on the horizon! ;)
PS: a big thanks for your hard work (dink and iq_132) ! :)
-
Thanks guys!
Here's the current progress with Asura Buster, you'll have to watch this video :)
https://www.youtube.com/watch?v=RMbYiFMupBY
best regards,
- dink
Impressive!
I knew you could do it :D
Thanks dink!
-
Not really a bug, but does anyone else stumble into this trap every now and then? ROM dirs need to end with a "\". When editing them manually, I tend to forget it. This could be handled automatically.
Also, the UI to manage ROM dirs is kinda awkward. Why not have a single window with 20 paths, rather than 20 tabs? It'd improve visibility and avoid extra clicks. Or better yet, have a dynamic / flexible number of paths if you want more. I reckon this is more of a feature request than a bug report, but I didn't think it deserved its own thread.
-
That's a great idea, and I'm totally with you on that. It really could use some work.
best regards,
- dink
-
Also, why not have a recursive directory search and then have less paths (for arcade systems)?
-
That'd be great. I actually checked whether this was supported last year, out of curiosity, but alas... :p
-
Hi!
I've just reported a new bug in DDTOD: http://www.fbalpha.com/bugs/00040/
Sound disappears. What a strange thing! Have you experience it in that or other game?
Thanks.
-
Hi!
I've just reported a new bug in DDTOD: http://www.fbalpha.com/bugs/00040/
Sound disappears. What a strange thing! Have you experience it in that or other game?
Thanks.
This is a common issue, and can easily be fixed. Press F5 and go down about 4 entries under the last P4 control, there's a volume up and volume down assignable button. Double click & choose the "clear lock" button then "clear input" for both of them, this way while you're playing the game, you don't accidentally hit the volume up/down buttons. It basically disables those buttons.
Now if you want those 2 buttons disabled for every CPS2 game, go into the config/games folder and copy the 2 volume lines from the ddtod.ini and put it in your preset.ini in the config/presets subdir...assuming you have this set up - see http://http://neo-source.com/index.php?topic=2800.0 for more info :)
best regards,
- dink
-
Thanks Dink. But you say that this issue is caused by pressing buttons accidentally. I don't understand how it can be happening always in the same point of the game and simultaneously in two netplay players.
Best regards.
-
Thanks Dink. But you say that this issue is caused by pressing buttons accidentally. I don't understand how it can be happening always in the same point of the game and simultaneously in two netplay players.
Best regards.
If that's the case and there is a bug or something more sinister "behind the scenes" happening I'll need a little more help from you, like some savestates before during and after the volume goes down.
But... I have actually seen it happen in netplay also, and it was because someone was pressing a key that was set as the cps2-volume down key. The next time the volume goes down assign a key/button to "volume up" and use that to try to restore the volume. If that doesn't work I'll definitely need the savestates.
best regards,
- dink
-
If that's the case and there is a bug or something more sinister "behind the scenes" happening I'll need a little more help from you, like some savestates before during and after the volume goes down.
But... I have actually seen it happen in netplay also, and it was because someone was pressing a key that was set as the cps2-volume down key. The next time the volume goes down assign a key/button to "volume up" and use that to try to restore the volume. If that doesn't work I'll definitely need the savestates.
best regards,
- dink
We will check it. Thanks.
Best regards.
-
I don't know if it's really a bug, but "Strikers 1945 Plus" got a lot of slowdown with current fba (on my i7 with standalone fbalpha, not on my raspberry with libretro-fbalpha ;)). I heard it was a lot smoother with the 0.2.97.37 release (i didn't test it myself though, but it was mentioned by the guy who reported this to me on github), so it could be a regression.
-
I don't know if it's really a bug, but "Strikers 1945 Plus" got a lot of slowdown with current fba.
That game has a speedhack, which is normally disabled. IIRC the slowdown is correct. You can simply overclock its CPU to make it run smooth.
-
I don't know if it's really a bug, but "Strikers 1945 Plus" got a lot of slowdown with current fba (on my i7 with standalone fbalpha, not on my raspberry with libretro-fbalpha ;)). I heard it was a lot smoother with the 0.2.97.37 release (i didn't test it myself though, but it was mentioned by the guy who reported this to me on github), so it could be a regression.
I don't know if in the previous FBAs Strikers 1945 Plus was smoother, but the slowdowns are present with the full emulation speed (all frames rendered, 59,18 fps), so, probably, the slowdowns are intrinsic of the original hardware (I never played with the real cabinet, so I can't be sure).
PS: the S1945Plus (original hardware) is based on the Neo Geo MVS, so also your raspberry Pi should emulate it without problems... ;)
-
Thanks for the quick answer, i just tried it with the cpu overclock feature (first time i use it, it never crossed my mind i could solve original hardware slowdown with it) and it feels a lot better.
-
Do you know where in the FBA Src the speedhack for Strikers 1945 Plus can be found..???
Probably 1625 to 1646 in neo_run.cpp (s1945p is mentioned) ? But it doesn't seem to help (it is enabled by default on libretro-fbalpha), perhaps this speedhack doesn't work properly anymore and that's why people said it was smoother in older fba releases.
-
Even with CPU oc, the game is not as smooth as it should be.
The .37 FBA revision played it way smoother (i could spot no differences in speed compared with the arcade machine) without any OC. It is definitelly an issue with latest FBA.
-
Even with CPU oc, the game is not as smooth as it should be.
The .37 FBA revision played it way smoother (i could spot no differences in speed compared with the arcade machine) without any OC. It is definitelly an issue with latest FBA.
It seems like the fix for the sound in karnov's revenge, haunted castle and some other games is the cause of this regression from .37 to .38
-
Thanks guys, I'll look into it. :)
-
Ouch, I now see and realize how bad the slowdown is in strikers 1945 plus. Will try to do something about it asap!
best regards,
- dink
-
Big difference with my builds too (no speedhacks). I just compared, my build is fluid with no slowdown until a couple of levels in. Looks like the sound sync between the cpus is going sideways.
-
In .37 I "fixed" ZetRunEnd(), and that had the side effect of breaking a ton of a games.. and probably more that weren't discovered yet. I then realized there was a reason it was broken in the first place and changed it back.
Stikers 1945Plus! benefitted from this immensely, though. So now its enabled just for this game. problem solved :)
barbudreadmon, I'll pm you the changes
best regards,
- dink
-
In .37 I "fixed" ZetRunEnd(), and that had the side effect of breaking a ton of a games.. and probably more that weren't discovered yet. I then realized there was a reason it was broken in the first place and changed it back.
Stikers 1945Plus! benefitted from this immensely, though. So now its enabled just for this game. problem solved :)
barbudreadmon, I'll pm you the changes
best regards,
- dink
It didn't work on the xbox it seems as the game still slows to a crawl when the screen fills up with enemies and bullets, but im sure
it will on other platforms which have more power under the hood.
-
Greatt!!!!
Now to wait for someone to compile it for android :D
-
It didn't work on the xbox it seems as the game still slows to a crawl when the screen fills up with enemies and bullets, but im sure
it will on other platforms which have more power under the hood.
If the sound becomes bad as well, then it is surely lack of power, if the sound stays good, then it's an issue with the emu.
-
It didn't work on the xbox it seems as the game still slows to a crawl when the screen fills up with enemies and bullets, but im sure
it will on other platforms which have more power under the hood.
I'm kinda surprised. It's using around 35% of 1 cpu core on my raspberry pi 3.
Greatt!!!!
Now to wait for someone to compile it for android :D
If you are using retroarch, it should be available on libretro's buildbot.
-
Yes, got the core and it works flawlessly. Thanks guys, you are great!
I didn't even know that .39 FBA was ported to android. It works great and it has a ton of options in the core, like Kaneko speed up and dips for most games :biggrin:
-
King of Fighters 98 Anniversary Edition (2016) >kof98ae2016< has some sound issues like sound overlapping and some distortions.
Examples: end of Iori's desperation move sounds weird (down, forward, down backwards + punch), the beginning of Kyo's hidden desperation move (half circle backwards X 2 + both punches).
Actually, depending on the rom, sound is weird in different spots. Revision .37 was better in this game as far as i know.
-
We added code in .39 to give voices to the new characters in this game, and that part seemed to work fine. I will look into this a little deeper and see if the other issues you mention can hopefully be fixed.
King of Fighters 98 Anniversary Edition (2016) >kof98ae2016< has some sound issues like sound overlapping and some distortions.
Examples: end of Iori's desperation move sounds weird (down, forward, down backwards + punch), the beginning of Kyo's hidden desperation move (half circle backwards X 2 + both punches).
Actually, depending on the rom, sound is weird in different spots. Revision .37 was better in this game as far as i know.
-
Revision .37 was better in this game as far as i know.
Could be related to the z80 cpu core issue ?
-
I don't ever play fighting games, so bear with me...
How do I do half circle back? I tried everything I can think of but can't get him to make that move.
King of Fighters 98 Anniversary Edition (2016) >kof98ae2016< has some sound issues like sound overlapping and some distortions.
Examples: end of Iori's desperation move sounds weird (down, forward, down backwards + punch), the beginning of Kyo's hidden desperation move (half circle backwards X 2 + both punches).
Actually, depending on the rom, sound is weird in different spots. Revision .37 was better in this game as far as i know.
-
el_rika, you seem to use libretro-fbalpha, did you update it to the latest version ?
-
I don't ever play fighting games, so bear with me...
How do I do half circle back? I tried everything I can think of but can't get him to make that move.
Half circle back = half a circle from forward to back, passing by the lower side (ie: forward, forward/down, down, down/back, back).
Note that you can't do desperation moves whenever you want... Depending on the game and selected settings, you either need a full super bar and/or be low on life. So it's tricky if you're not into fighting games. I guess it'd be easier if you're provided with a savestate and/or a macro.
-
el_rika, you seem to use libretro-fbalpha, did you update it to the latest version ?
yes, i did.
both .38 and .39 have strange sound issues in this game. Could be that i have an incompatible rom, though i tried 3 different versions.
-
If the sound becomes bad as well, then it is surely lack of power, if the sound stays good, then it's an issue with the emu.
Yeah the sound is fine, so as you say it's very likely a problem with the Neogeo code in FBL when it comes to this game as
we have alot of VMM loading code wrapped around the Gfx code so the big Neogeo roms dont run out of mem on boot
due to the 64mb Ram limitations of the Xbox console, maybe something in that extra VMM code which causes issues with
Strikers 1945 Plus.
-
I'm kinda surprised. It's using around 35% of 1 cpu core on my raspberry pi 3.
If you are using retroarch, it should be available on libretro's buildbot.
Yeah i dont think it's the hardware as such more than likely a Emulator problem as we have older MAME ports on the xbox and
Strikers 1945 Plus plays like a treat when using those.
-
Battle Garregga issue.
I tested on .37, .38, .39, and after a while (pretty random, from a couple of minutes to fifteen minutes) the fire button (which is autofire in this game) behaves strangely:
1. it changes the autofire rate, with huge gaps appearing between salvos (it should normally be a pretty constant rate of fire).
2. stops working completely (the ship simply stops firing till i drop a bomb or die, or it just comes back after a few seconds, or not al all).
It is usually 1 followed by 2.
Tested mostly with Golden Bat (fourth ship chose by pressing A+B+C).
In Mame this doesn't happen.
Cheers, and have a beautiful weekend! :biggrin:
-
Hi el_rika,
Its really important to fix this one. Please check your PM's here, I need some more information/help to track down this bug.
thanks for the report! :)
best regards,
- dink
Battle Garregga issue.
I tested on .37, .38, .39, and after a while (pretty random, from a couple of minutes to fifteen minutes) the fire button (which is autofire in this game) behaves strangely:
1. it changes the autofire rate, with huge gaps appearing between salvos (it should normally be a pretty constant rate of fire).
2. stops working completely (the ship simply stops firing till i drop a bomb or die, or it just comes back after a few seconds, or not al all).
It is usually 1 followed by 2.
Tested mostly with Golden Bat (fourth ship chose by pressing A+B+C).
In Mame this doesn't happen.
Cheers, and have a beautiful weekend! :biggrin:
-
el_rika:
I talked to a friend regarding our Battle Garegga emulation and things, he says that you probably accidentally put your ship into "program autofire" mode. This will record the fire buttons you press and play them back. Its built into the game itself, you can read about it here: http://shmups.system11.org/viewtopic.php?t=351 search for autofire. It causes problems exactly like you mentioned when it is accidentally activated.
I think I never ran into this in the million times I've played Battle Garegga because I always use the FBAlpha Autofire instead which can easily be configured: Input -> Autofire Rate, and Input -> Map Game Inputs and scroll way down to map the autofire buttons.
If you think its a bug and not this, please do the recording stuff I PM'd you about and I will investigate :)
best regards,
- dink
-
thanks for answering, i'm checking the link right now!
edit:
Ok, so i tested the game for the last few days, and also asked the guys at shmup forums about the fire rate. I swear the developers couldn't have made this game more complicated if the wanted, this is the MGS of shmups right here :p
It is still a bit random the change in fire rate, but i guess it is how it is.
One small other thing though. I noticed that, unless i overclock the internal CPU of the emu to 200 (max), Garegga has slowdowns here and there (like when you lose your last life of the credit and all the power ups fall from you, or when the 2-nd stage boss appears, etc). As far as i know, the original PCB has almost zero slowdown, so i guess this is the way to experience it.
Thanks for you patience, have a great week!
-
Oops, I didn't see your reply until now because you edited instead of replied. :redface:
I agree, Battle Garegga is seriously advanced in all respects, every time I play it I learn something new from the game. and also most perfect shooter ever :biggrin:
I think our emulation of it is near-perfect, but I will compare our emulation of B.Garegga with MAME and PCB to find those timing problems you have mentioned, and try my best to fix them properly without needing any sort of overclocking.
best regards,
- dink
thanks for answering, i'm checking the link right now!
edit:
Ok, so i tested the game for the last few days, and also asked the guys at shmup forums about the fire rate. I swear the developers couldn't have made this game more complicated if the wanted, this is the MGS of shmups right here :p
It is still a bit random the change in fire rate, but i guess it is how it is.
One small other thing though. I noticed that, unless i overclock the internal CPU of the emu to 200 (max), Garegga has slowdowns here and there (like when you lose your last life of the credit and all the power ups fall from you, or when the 2-nd stage boss appears, etc). As far as i know, the original PCB has almost zero slowdown, so i guess this is the way to experience it.
Thanks for you patience, have a great week!
-
So far I tried Battle Garegga in MAME and FBA, and the slowdown when you die at the first and second boss is exactly the same for both emulators. I don't see any slowdown when the 2nd level boss appears in at all, can you possibly make a video of this slowdown or record a input recording(using the windows version) to show me how and where this happens so I can figure out where the problem is? Are you using the windows version? If not, can you see if the same problem happens in the windows version?
Thanks & best regards,
- dink
-
The hardware that game runs on has a register that reflects the CRT status in detail, and that's probably wrong to some degree in most or all emulators. The games heavily depend on it, and will sit in loops to wait for certain conditions.
e.g. here's one from battle garegga (there are more): 002236 -- -- -- -- 3f07 move.w D7, -(A7)
002238 -- -- 3e39 0021 c03c move.w $21c03c.l, D7
00223e -- -- -- 6b00 0008 bmi 2248
002242 -- -- 3e39 0021 c03c move.w $21c03c.l, D7
002248 -- -- -- -- 5307 subq.b #1, D7
00224a -- -- -- 0c07 00ee cmpi.b #-$12, D7
00224e -- -- -- -- 64e8 bcc 2238
002250 -- -- -- -- 3e1f move.w (A7)+, D7
002252 -- -- -- -- 4e75 rts
Get that register right, and it affects both the slowdown and the sprites moving out of sync (my current source code is a bit better with the CPU at 100%). These games are a part of the reason for the CPU overclocking feature in fba.
Batrider also has some weirdness with the sound communication, though that only seems to affect the attract mode. FBA's drivers need an update to synchronise the cpus and sound better, but that probavly won't affect the slowdown at all.
-
Where does the issue with Batrider's sound happen? I'd like to fix this asap but need to know what to listen for.
best regards,
- dink
-
Ahh right, so after listening to Batrider on MAME and PCB, I realized how sluggish ours is. Bumping the interleave up from 8 to 64 gets ours on par with everything :) Happy day!
EDIT: after many many many more comparisons, I found a couple spots where certain sounds lagged still. Increasing the interleave a bit more and removing a check that only ran the soundcpu every other frame interleave did the trick.
I wonder if any other toaplan games suffer from this as well? I guess I will have to investigate. :cool:
best regards,
- dink
-
Regarding the recent toaplan bughunt / from the "all video effects matters" and "I was always told FBAlpha's Toaplan emulation was infallible" departments.
Gab75 pointed out earlier that the mosaic effect is missing from FixEight during scene transistions. I made a fix for this, but we need to do some more testing before committing - since some major games are involved (battle garegga/bakraid, mahoudai, shippumd & truxton2).
best regards,
- dink
-
Oops, I didn't see your reply until now because you edited instead of replied. :redface:
I agree, Battle Garegga is seriously advanced in all respects, every time I play it I learn something new from the game. and also most perfect shooter ever :biggrin:
It is indeed a pretty good game. I just discovered it a year ago and i love it. It is an outstanding achievement from the developers seeing how it is so complex and difficult yet the game is so balanced and addictive :)
The hardware that game runs on has a register that reflects the CRT status in detail, and that's probably wrong to some degree in most or all emulators. The games heavily depend on it, and will sit in loops to wait for certain conditions.
Get that register right, and it affects both the slowdown and the sprites moving out of sync (my current source code is a bit better with the CPU at 100%). These games are a part of the reason for the CPU overclocking feature in fba.
So my assessment was correct. I think with default CPU clock, Garegga stutters a bit compared to PCB (when you die, at the levels transition, some bombs involving a lot of sprites, etc). With the clock to 160 though, it is smooth. I foud Progear to benefit from this too, though a bit more testing is needed.
Sadly i can not use the emu on PC atm, it is antiquated :(
My newly acquired Note 3 should be strong enough to keep me in the loop. Except for the Cave SH3 games (in mame of course), everything flies ;)
-
Sadly i can not use the emu on PC atm, it is antiquated :(
Which means you are using fba4droid ? libretro-fbalpha ? You should mention it when you report an issue, it could be an issue specific to the port.
-
Get that register right, and it affects both the slowdown and the sprites moving out of sync (my current source code is a bit better with the CPU at 100%). These games are a part of the reason for the CPU overclocking feature in fba.
Could you please be a little more specific, and/or share the code?
-
Which means you are using fba4droid ? libretro-fbalpha ? You should mention it when you report an issue, it could be an issue specific to the port.
libretro-fba in Retroarch. I will be more detailed from now on in my findings. Thanks! :smilie:
-
libretro-fba in Retroarch. I will be more detailed from now on in my findings. Thanks! :smilie:
If there is any difference between fba standalone and libretro-fba in regards to those slowdowns, perhaps looking at the new z80 core would be a good idea (it was pre-released for libretro-fba a few weeks ago to fix the issue for s1945p, and there shouldn't be any major difference except this one for battle garegga)
-
Could you please be a little more specific, and/or share the code?
I simply meant what I said... There's a register that reflects the CRT's status (electron beam position, blanking...) that the games depend on to avoid graphics glitches. It's undocumented, and needs to be emulated correctly or you run into various problems.
Also, when forced interleave between the CPUs affects the timing, it's time to get rid of it, and just sync the CPUs whenever they access a shared resource, which is how things generally *should* work anyway... batrider and battle bakraid seem to let the Z80 trigger interrupts on the 68K as well as the other way around, so you can't get it perfect unless you have per-instruction interleave, but you can get close enough.
I did both of the above for batrider and battle garegga (battle bakraid aways had proper cpu sync), and the games run a bit better. I'll do the same for the rest when I have time.
-
Also, when forced interleave between the CPUs affects the timing, it's time to get rid of it, and just sync the CPUs whenever they access a shared resource, which is how things generally *should* work anyway... batrider and battle bakraid seem to let the Z80 trigger interrupts on the 68K as well as the other way around, so you can't get it perfect unless you have per-instruction interleave, but you can get close enough.
Thanks, that did the trick. The interleave for batrider is back to 8 (from 128) and the sound timing is perfect now. I synchronized the z80 cycles to the 68k before setting the soundlatches, and got rid of those silly zetrun's after the nmi.
edit: regarding bgaregga and the ToaScanlineRegister, I did many tests between your more advanced version and fba's current simplified version of the function. I also tested a super-simplified version of this from Raine, which basically just returns 0x8000 | (counter++&0x1ff), and another ported from MAME which somewhat resembles yours. Between them all there really is no difference at all in the lag at the second boss. Then I tested different emulators: yours vs. FBA vs. MAME and many-a-time, and really there is no difference in the lag when the ship blows up. I found that the laser attack of the second level boss causes the most lag when your ship blows up, so I used that as the testcase.
... so.. at this point I think the best thing for people not willing to deal with that lag is to bump up the cpu speed. Or buy the PCB :)
best regards,
- dink
-
Found a problem with Bio-Ship Paladin NMK16.cpp it's not specific to this build as it's been this way for a while, check
level two and you'll notice the BG is missing not sure about the later levels as i've not got that far as of yet.
-
gamez fan thanks for the bug report, problem solved. :)
-
Not a new problem, but something I've had since the beginning: vertical games are displayed upside down on my computer (but are displayed correctly on my Pi).
I have the "Rotate vertically aligned games" option checked, but it seems the rotation is done in the wrong direction.
Any ideas? I have a NVIDIA GeForce GTX 660, for what it's worth.
-
Change blitters.
-
ddraw blitter and manufacturers dropping 2D support is not a good combination. That thing should be considered to have legacy AKA "strictly there for compatibility to run on old rubbish" status.
-
gamez fan thanks for the bug report, problem solved. :)
Nice one dink!!! i knew you'd get to the bottom of it ;)
-
Hmm okay, thanks.
By the way, I just tried playing Super Cup Finals (scfinals.zip). Trying to insert credits triggers a sound, but credits are not added (whether I add P1 or P2 coins).
Edit: also, Pururun sound emulation is imperfect. It seems there are missing music channels or something. I guess it should sound the way it does on MAME (https://youtu.be/9bs8GsVPmjs). For example, when you're in the first level, there is some kind of beat channel in the tune that is missing in FBA. Available if you need more info.
-
By the way, I just tried playing Super Cup Finals (scfinals.zip). Trying to insert credits triggers a sound, but credits are not added (whether I add P1 or P2 coins).
The game has some issues... to add coins you can use the "service coin" input (9 by default)
-
Thanks for the reports,
For super cup finals, use 9 to insert a coin. This one is weird, and I can't figure out why it does that. I will keep trying though.
The problem with Pururun has been corrected!
best regards,
- dink
Hmm okay, thanks.
By the way, I just tried playing Super Cup Finals (scfinals.zip). Trying to insert credits triggers a sound, but credits are not added (whether I add P1 or P2 coins).
Edit: also, Pururun sound emulation is imperfect. It seems there are missing music channels or something. I guess it should sound the way it does on MAME (https://youtu.be/9bs8GsVPmjs). For example, when you're in the first level, there is some kind of beat channel in the tune that is missing in FBA. Available if you need more info.
-
Thanks for the reports,
For super cup finals, use 9 to insert a coin. This one is weird, and I can't figure out why it does that. I will keep trying though.
The problem with Pururun has been corrected!
Nice, thanks! :smilie:
-
GP Rider (World, FD1094 317-0163) ------------ gpriders.zip
is listed as a clon of:
GP Rider (World, FD1094 317-0163) (Twin setup) ------- gprider.zip
However both parent and clon need exactly the same 30 files, and then are exactly the same game.
Maybe the clon files should be different?
-
I know that this is not very important, but this game:
Knuckle Bash (Korean PCB
has a missing ")" at the end of its name.
-
I know that this is not very important, but this game:
Knuckle Bash (Korean PCB
has a missing ")" at the end of its name.
Not sure what you mean. I do see the ending ")" in the game list.
-
Rasek, Thanks for letting us know!
Stifu, it's in the non-Japanese game name, enable "Latin text only" mode to see.
best regards,
- dink
I know that this is not very important, but this game:
Knuckle Bash (Korean PCB
has a missing ")" at the end of its name.
-
I know that this is not very important, but this game:
Knuckle Bash (Korean PCB
has a missing ")" at the end of its name.
Good eye! It's fixed now (dink).
-
Rasek, try this gpriders.zip, its just basically an empty zip file so you don't have to have a duplicate romset for gpriders. On the other hand, gpriders should play fine without a gpriders.zip at all as long as the parent set is good. I think :)
best regards,
- dink
GP Rider (World, FD1094 317-0163) ------------ gpriders.zip
is listed as a clon of:
GP Rider (World, FD1094 317-0163) (Twin setup) ------- gprider.zip
However both parent and clon need exactly the same 30 files, and then are exactly the same game.
Maybe the clon files should be different?
-
Hi dink,
There is no need for a duplicate file.
Just having the parent is enough.
As the files that the "clon" need are the same 30 files that the "parent", all of them are already available.
Also it is not necessary to have an empty gpriders.zip
Game can be launched without it.
After all, it is the same game that the "parent".
-
Stratof's level 3 "pink glob" are not rendered properly, the following video shows how they should be rendered :
https://www.youtube.com/watch?v=VB3QiJPjkkY&feature=youtu.be&t=454
Actually this issue was reported to fbalpha-libretro a few months ago, but at first i thought it was a port issue.
-
Stratof's level 3 "pink glob" are not rendered properly, the following video shows how they should be rendered :
https://www.youtube.com/watch?v=VB3QiJPjkkY&feature=youtu.be&t=454
Actually this issue was reported to fbalpha-libretro a few months ago, but at first i thought it was a port issue.
Indeed, watching your video, the pink globs should have a "translucent effect"... I'm not sure the real hardware supports the translucent effect (probably not)... maybe the translucent effect is simulated with a "flashing effect" (like the title in the intro).
-
I've been getting error messages on the first Land Maker startup. A red text blinks 3 times over a black background (something error, can't read it, it's too fast). A little while after those error messages, the game still starts properly. The following Land Maker startups do not show those errors, but I can repro the problem if I delete landmakr.ini and landmakr.nv and try again.
Also, I noticed one pixel row seems off at the bottom of the game intro. See screenshot.
-
Thanks for the report!
At the present time, our Raiga emulation doesn't support alpha blending or raster effects. (that's why the title flickers horribly).
best regards,
- dink
Stratof's level 3 "pink glob" are not rendered properly, the following video shows how they should be rendered :
https://www.youtube.com/watch?v=VB3QiJPjkkY&feature=youtu.be&t=454
Actually this issue was reported to fbalpha-libretro a few months ago, but at first i thought it was a port issue.
-
F3 games will get kinda in a "bad mood" if their nvram goes corrupt, and this will happen if you close the game while its writing to the nvram. (which could be any time during the game...) I think that might be the cause of the error message @ boot - try going into config/games and rename landmakr.nv to landmakr.nvbak or something, and see if that clears them up. If it does, please send me your landmakr.nvbak file and I'll take a peek just to be sure.
I know about that line in the intro, I'm not sure that it can be fixed, but I'll give it another try.
thanks,
- dink
I've been getting error messages on the first Land Maker startup. A red text blinks 3 times over a black background (something error, can't read it, it's too fast). A little while after those error messages, the game still starts properly. The following Land Maker startups do not show those errors, but I can repro the problem if I delete landmakr.ini and landmakr.nv and try again.
Also, I noticed one pixel row seems off at the bottom of the game intro. See screenshot.
-
Well, I get those errors (on PC and Pi) only when I have no nv file yet. I don't get the errors if a nv file is present. Don't you get the same behavior?
-
I miss some scenes in Operation Wolf.
For example, at the end of stage 2 (Jungle), there should be a final boss (a bad guy covering behind a woman).
Seems that this scene is not emulated.
Also, the game should have some randomly generated extra levels "The enemy has located you".
I have never seen them.
Here is a video of the arcade:
https://www.youtube.com/watch?v=cQqaw_Vf9wk
-Jungle Boss can be seen at 19:25
-"The enemy has located you" extra level at 21:38
-
I found this, just in case it can help:
http://www.bryanmcphail.com/wp/?p=794
-
And this:
http://forum.arcadecontrols.com/index.php?topic=150248.0
-
Using the "Set Dipswitches" to run the game in Japanese, those 2 issues are solved, and two other I havent noticed yet are also solved:
-At the end of stage 4, there is now a nice explosion effect. I had forgotten that from my Arcades days...
-At the end of stage 5, the pov is raised to show the sky, and the battle with the remaining helicopters is held there, not at ground level.
So all 4 bugs are solved... but only if you play in Japanese :-(
-
Rasek,
I played and compared Operation Wolf in MAME and FBA, and they operate exactly the same in both emu's. I afraid that until someone dumps the protection MCU, there really isn't much we can do about those issues. In the meantime, check out these excellent games from my top lightgun games list: Alligator Hunt, Beast Busters, Blood Bros., and Cabal.
best regards,
- dink
-
It seems that since 0.2.97.11 the memory card option is broken (doesn't work at all) on Neo Geo games, any ideas about this?
-
Rasek,
I played and compared Operation Wolf in MAME and FBA, and they operate exactly the same in both emu's. I afraid that until someone dumps the protection MCU, there really isn't much we can do about those issues. In the meantime, check out these excellent games from my top lightgun games list: Alligator Hunt, Beast Busters, Blood Bros., and Cabal.
best regards,
- dink
Hi dink, thank you very much for your answer and your tips ;p
I will try those games... Cabal specially.
And of course Operation Wolf in Japanese, which has the Jungle Boss (and the weak point he had at the Arcades, still works in FBA :biggrin: ).
Just wondering why setting that dipswitch makes the game work properly... Something related to protection, I suppose.
-
Stratof's level 3 "pink glob" are not rendered properly, the following video shows how they should be rendered :
https://www.youtube.com/watch?v=VB3QiJPjkkY&feature=youtu.be&t=454
Actually this issue was reported to fbalpha-libretro a few months ago, but at first i thought it was a port issue.
Fixed.
-
Fixed.
Excellent! :)
Out of curiosity: does the real hardware manage a true semi-transparent layer or use some tricks to create the translucent effect?
-
Excellent! :)
Out of curiosity: does the real hardware manage a true semi-transparent layer or use some tricks to create the translucent effect?
According to MAME it does an additive blend to achieve this, so a true semi-transparent layer.
-
According to MAME it does an additive blend to achieve this, so a true semi-transparent layer.
Remarkable for an arcade game of early 90s! Thanks for the clarification! :)
-
Very nice!! :D
-
Hi,
In windows 7, FBA UI doesn't work when icons function is activated.
Best regards.
-
In windows 7, FBA UI doesn't work when icons function is activated.
Maybe you can find the solution in the the following post:
http://http://neo-source.com/index.php?topic=2838.msg24212#msg24212
-
Maybe you can find the solution in the the following post:
http://http://neo-source.com/index.php?topic=2838.msg24212#msg24212
Hi Gab75,
I tried that solution but doesn't worked for my.
Best regards.
-
Hi Gab75,
I tried that solution but doesn't worked for my.
Hi IPLA,
I'm sorry that it wasn't helpful... Did you try with all FBA versions?
I use Windows 7 64 bit and FBA 32 bit (Windows Binary and Windows XP Binary) and I haven't issues with the icons driver (I set GDI handle quota to 65,536)...
-
The reason FBA ui goes bonkers when icons is enabled: the system is running out of gdi handles from loading all of the icons. I have a couple ideas I'm going to try to implement to reduce the number of handles needed.
best regards,
- dink
-
I found this segfault/crash launching kov2p, in versions 0.2.97.40 and 0.2.97.42. The stack trace is from the libretro version.
I reported it to the libretro port maintainer but I think I confused them, and it might be helpful if people who are more familiar with the codebase can look at this, since it's likely a memory safety issue or overflow in the emulation code.
Thread 1 "retroarch" received signal SIGSEGV, Segmentation fault.
0x00007fffe6778497 in Arm7ReadByte (addr=939524096) at src/cpu/arm7_intf.cpp:208
208 return membase[READ][addr >> PAGE_SHIFT][addr & PAGE_BYTE_AND];
(gdb) bt
#0 0x00007fffe6778497 in Arm7ReadByte (addr=939524096) at src/cpu/arm7_intf.cpp:208
#1 0x00007fffe677fb1d in arm7_cpu_read8 (addr=939524096) at src/cpu/arm7/arm7core.c:187
#2 0x00007fffe6788eaa in Arm7Run (cycles=277) at src/cpu/arm7/arm7exec.c:802
#3 0x00007fffe61a09c0 in pgmFrame () at src/burn/drv/pgm/pgm_run.cpp:873
#4 0x00007fffe5d4eade in BurnDrvFrame () at src/burn/burn.cpp:719
#5 0x00007fffe68ea5b0 in ForceFrameStep () at src/burner/libretro/libretro.cpp:795
#6 0x00007fffe68ebcb0 in retro_run () at src/burner/libretro/libretro.cpp:1317
#7 0x000000000040d819 in core_run () at core_impl.c:415
#8 0x0000000000411e35 in runloop_iterate (sleep_ms=sleep_ms@entry=0x7fffffffe8ec) at retroarch.c:3051
#9 0x000000000040c732 in rarch_main (argc=1, argv=<optimized out>, data=0x0) at frontend/frontend.c:129
#10 0x00007ffff55f64ca in __libc_start_main () from /usr/lib/libc.so.6
#11 0x0000000000409fba in _start ()
I can reproduce this consistently on one of my machines, but I've found that reproducing it in general is highly dependent on the operating system/host CPU, among other factors. kov and kovplus don't trigger it, but kov2p, kovsh, and theglad do.
-
I found this segfault/crash launching kov2p, in versions 0.2.97.40 and 0.2.97.42. The stack trace is from the libretro version.
I reported it to the libretro port maintainer but I think I confused them, and it might be helpful if people who are more familiar with the codebase can look at this, since it's likely a memory safety issue or overflow in the emulation code.
I see the problem, thanks for the bug report :)
code change:
Index: arm7_intf.cpp
===================================================================
--- arm7_intf.cpp (revision 4897)
+++ arm7_intf.cpp (working copy)
@@ -391,6 +391,7 @@
for (INT32 i = 0; i < 3; i++) {
membase[i] = (UINT8**)malloc(PAGE_COUNT * sizeof(UINT8*));
+ memset(membase[i], 0, PAGE_COUNT * sizeof(UINT8*));
}
CpuCheatRegister(nCPU, &Arm7CheatCpuConfig);
best regards,
- dink
-
Marvellous, thank you very much.
(Unfortunately I can't get a password reset email from this forum. :redface: )
-
Hi. I'm still having some sound and graphical issues in Strikers 1945 plus. I only have the issue when using the "Zero Fighter" plane. The super shot explosion is distorted. Also the music in some levels will skip around or loop or just stop. You also get some random white noise pops. Unsure why but it seems limited to using that plane.
Thanks for the help and all you have done with the emulator.
-
Wildman1024, I played the first 2 levels with Zero Fighter on my p4/winxp system and could not reproduce any glitches to audio/video as you mentioned.
Did you watch the FPS counter while playing the game? it can be toggled w/backspace. Does the FPS stay at or above 58? Does it dip down a bit when the glitches occur? If it dips down below the upper 50's, try enabling Auto Frameskip under Video. Also try the XAudio2 audio plugin under the Audio tab, and if it still doesn't help, try increasing the
"buffer size" (under Audio tab again)
Since I can't reproduce the problem on my side, and if some settings don't help - I'll need you to give me some more information and an input recording (Game -> Replay -> Record Input), then attach it to a message here and let me know around the frame #'s the glitches happen - After you record it, play it back and watch the frame counter.
best regards,
- dink
-
Wildman1024, I played the first 2 levels with Zero Fighter on my p4/winxp system and could not reproduce any glitches to audio/video as you mentioned.
Did you watch the FPS counter while playing the game? it can be toggled w/backspace. Does the FPS stay at or above 58? Does it dip down a bit when the glitches occur? If it dips down below the upper 50's, try enabling Auto Frameskip under Video. Also try the XAudio2 audio plugin under the Audio tab, and if it still doesn't help, try increasing the
"buffer size" (under Audio tab again)
Since I can't reproduce the problem on my side, and if some settings don't help - I'll need you to give me some more information and an input recording (Game -> Replay -> Record Input), then attach it to a message here and let me know around the frame #'s the glitches happen - After you record it, play it back and watch the frame counter.
best regards,
- dink
Dink. Thank you for the reply. Its greatly appreciated.
I tried the tings you mentioned and it did not help me. I have a recording to send you. Thank you
***Not sure how to send you the file***
-
Hi. I'm still having some sound and graphical issues in Strikers 1945 plus. I only have the issue when using the "Zero Fighter" plane. The super shot explosion is distorted. Also the music in some levels will skip around or loop or just stop. You also get some random white noise pops. Unsure why but it seems limited to using that plane.
I did various tests (with different audio settings) and I found some audio issues when I disabled the "FM interpolation".
Have you the FM interpolation enabled? If not, you can enable it in the audio menu (Audio => 4-point 3rd order FM interpolation)
-
Dink. Thank you for the reply. Its greatly appreciated.
I tried the tings you mentioned and it did not help me. I have a recording to send you. Thank you
***Not sure how to send you the file***
Just click "attachments and other options" when replying (under the message box), or upload the file to http://www.sendspace.com and post the link here.
thanks,
- dink
-
I did various tests (with different audio settings) and I found some audio issues when I disabled the "FM interpolation".
Have you the FM interpolation enabled? If not, you can enable it in the audio menu (Audio => 4-point 3rd order FM interpolation)
Quite right. FM interpolation is actuially a bit of a misnomer -- only if that is enabled will fba emulate the chip at the correct clock frequency / sample rate. You should always have it enabled.
-
Just click "attachments and other options" when replying (under the message box), or upload the file to http://www.sendspace.com and post the link here.
thanks,
- dink
I do not have that option. I will upload to your link.
https://www.sendspace.com/file/zq99er (https://www.sendspace.com/file/zq99er)
Delete link
https://www.sendspace.com/delete/zq99er/7aa2df2d0080a9426f2e073d1970d01b (https://www.sendspace.com/delete/zq99er/7aa2df2d0080a9426f2e073d1970d01b)
-
I do not have that option. I will upload to your link.
If you expand the underlying picture you can see where is the "Attachments and other options" function... PS: I used this function to attach the picture! ;)
-
If you expand the underlying picture you can see where is the "Attachments and other options" function... PS: I used this function to attach the picture! ;)
OK I see now. I saw trying to do it through a PM not directly here.
Thank you for the "how to"
-
Here is the file
-
This is the explosion that I was talking about. This is your super shot with the Zero. Is this normal? Doesn't seem right to me.
-
This is the explosion that I was talking about. This is your super shot with the Zero. Is this normal? Doesn't seem right to me.
Unfortunately I can't try the game with the original PCB... I tried the game with MAME and the "super shot explosion" doesn't change (attached two MAME screenshots)
-
Unfortunately I can't try the game with the original PCB... I tried the game with MAME and the "super shot explosion" doesn't change (attached two MAME screenshots)
That must be normal then. Thank you for checking.
-
PS: maybe I found a small bug... if I use the XAudio2 audio plugin in combination with 44.1 kHz samplerate, after a while I began a game session, the sound effects stop working (differently the music goes on without problems)... the sound effects don't stop to work in an exact spot, approximately shortly before to meet the level boss.
-
PS: maybe I found a small bug... if I use the XAudio2 audio plugin in combination with 44.1 kHz samplerate, after a while I began a game session, the sound effects stop working (differently the music goes on without problems)... the sound effects don't stop to work in an exact spot, approximately shortly before to meet the level boss.
Yes it is kinda random. But that's pretty similar to my issue. Gameplay is fine though.
-
Thanks for the testing and bug report, the problem is now fixed :) The "speedhack" was running too many cycles on the z80 causing certain soundcommands to get skipped, and if the wrong one gets skipped - like a timer re-init, music stops. (or worse)
best regards,
- dink
-
Thanks for the testing and bug report, the problem is now fixed :) The "speedhack" was running too many cycles on the z80 causing certain soundcommands to get skipped, and if the wrong one gets skipped - like a timer re-init, music stops. (or worse)
Thanks for your invaluable support ! :)
-
Thanks for the testing and bug report, the problem is now fixed :) The "speedhack" was running too many cycles on the z80 causing certain soundcommands to get skipped, and if the wrong one gets skipped - like a timer re-init, music stops. (or worse)
best regards,
- dink
Thank you for the good work :).
-
Thank you very much. Your hard work is much appreciated.
-
Hi, not sure if this is really a bug, but maybe you can help. I found that if you load a game directly via a savestate (tried with CPS2 ssf2xj), that had some settings modified (EEPROM settings, F2), it loads them correctly, but if you press F3 to reset the driver, then these settings are wiped. But if you modify settings again, then press F3 again, settings are kept.
It seems like if the savedata load doesn't set an eeprom flag or something, so when you press F3, it justs resets the EEPROM. But if you go to settings (F2) then save, that flag is set so when pressing F3 it doesn't wipe the data. I've run through the source code but couldn't figure out by myself. Any idea?
Thanks!
-
Aquajack - The water changes color/palette, similar to 19XX but more easily reproducable. (after savestate load) Now I'm having trouble reproducing this one.. Maybe it was a fluke?
It happens here, colour going quite monochrome but on a different colour full screen (red) or windowed (green) in-game, while mainly greyish-brown in intermission.
-
In the game Arbalester, music playback is slow (about 50% slower).
The whole game also seems to run a bit slower, but I'm not completely sure.
Here's my system info:
FB Alpha v0.2.97.42 system information (Sun Sep 3 12:13:15 2017
).
----------------------------------------------------------------------
System information:
OS: Microsoft Windows 7 Service Pack 1 (build 7601)
CPU: AMD Athlon(tm) II X3 455 Processor
Physical RAM: 4193464 KB (4095 MB) total, 1982644 KB (1936 MB) avail
Total RAM: 8385068 KB (8188 MB) total, 5803464 KB (5667 MB) avail
FB Alpha: 41056 KB in use (46440 KB peak, 101072 KB virtual)
Installed displays and display adapters:
Generic PnP Monitor on NVIDIA GeForce 9500 GT (primary)
----------------------------------------------------------------------
FB Alpha information:
Built on Jul 9 2017, 20:25:03, using Visual C++ 2017.
Optimised for X64 CPUs.
Using Unicode for all text.
Debug functionality absent.
MMX optimisations disabled.
Musashi emulation core enabled for MC680x0 family emulation.
Emulating arbalest (Arbalester)
Vertical refresh is 60.00 Hz.
CPU running at 100% of normal frequency.
Video settings:
Selected module: DirectDraw7 video output
Interface settings: Running in windowed mode, $ix763, 980bpp
Source image 384x224, 32bpp
Module settings: Using Blt() to transfer the image
Using video memory for the final blit
Using software rotation,
Audio settings:
Selected module: DirectSound3 audio output
Interface settings: Playback at 44100Hz, 100% volume
Module settings: Audio is delayed by approx. 83ms
Input settings:
Selected module: DirectInput8 input
Interface settings: keyboard 0 System keyboard: Keyboard
mouse 0 System mouse: Mouse
joystick 0 2-Axis,8-Button
Profiling settings:
Interface settings: Profiling module not initialised
----------------------------------------------------------------------
Detailed process information about fba64.exe:
fba64.exe (base address 0x000000013F530000, size 55188 KB)
Modules loaded by fba64.exe:
ntdll.dll (base address 0x0000000077840000, size 1704
kernel32.dll (base address 0x0000000077720000, size 1148
KERNELBASE.dll (base address 0x000007FEFD8E0000, size 424
USER32.dll (base address 0x0000000077620000, size 1000
GDI32.dll (base address 0x000007FEFDED0000, size 412
LPK.dll (base address 0x000007FEFEAB0000, size 56
USP10.dll (base address 0x000007FEFEAE0000, size 812
msvcrt.dll (base address 0x000007FEFE5E0000, size 636
COMDLG32.dll (base address 0x000007FEFDB70000, size 604
SHLWAPI.dll (base address 0x000007FEFE1A0000, size 452
COMCTL32.dll (base address 0x000007FEFBE30000, size 2000
SHELL32.dll (base address 0x000007FEFEC90000, size 13868
WINMM.dll (base address 0x000007FEFA820000, size 236
ole32.dll (base address 0x000007FEFE220000, size 2032
RPCRT4.dll (base address 0x000007FEFE4B0000, size 1204
ADVAPI32.dll (base address 0x000007FEFEBB0000, size 876
sechost.dll (base address 0x000007FEFE9F0000, size 124
WININET.dll (base address 0x000007FEFDC10000, size 1192
urlmon.dll (base address 0x000007FEFDD40000, size 1552
XmlLite.dll (base address 0x000007FEFD950000, size 212
OLEAUT32.dll (base address 0x000007FEFFA70000, size 872
CRYPT32.dll (base address 0x000007FEFD990000, size 1460
MSASN1.dll (base address 0x000007FEFD7F0000, size 60
[size=12pt][size=8pt] iertutil.dll (base address 0x000007FEFDF40000, size 2428
AVIFIL32.dll (base address 0x000007FEF47F0000, size 124
MSACM32.dll (base address 0x000007FEFBE00000, size 96
MSVFW32.dll (base address 0x000007FEF4730000, size 164
IMM32.DLL (base address 0x000007FEFE420000, size 184
MSCTF.dll (base address 0x000007FEFE680000, size 1060
api-ms-win-core-synch-l1-2-0.DLL (base address 0x000007FEF3450000,
dsound.dll (base address 0x000007FEF8490000, size 544
POWRPROF.dll (base address 0x000007FEFC310000, size 176
SETUPAPI.dll (base address 0x000007FEFE790000, size 1884
CFGMGR32.dll (base address 0x000007FEFD8A0000, size 216
DEVOBJ.dll (base address 0x000007FEFDB40000, size 104
dinput8.dll (base address 0x000007FEF2480000, size 240
ddraw.dll (base address 0x000007FEEFBF0000, size 964
DCIMAN32.dll (base address 0x000007FEF8620000, size 32
dwmapi.dll (base address 0x000007FEFB740000, size 96
d3d9.dll (base address 0x000007FEEF200000, size 2044
VERSION.dll (base address 0x000007FEFC880000, size 48
d3d8thk.dll (base address 0x000007FEF86B0000, size 28
D3DX9_43.dll (base address 0x000007FEEEA10000, size 2444
uxtheme.dll (base address 0x000007FEFBD80000, size 344
HID.DLL (base address 0x000007FEFB7A0000, size 44
WINTRUST.dll (base address 0x000007FEFDB00000, size 236
ntmarta.dll (base address 0x000007FEFB9A0000, size 180
WLDAP32.dll (base address 0x000007FEFE450000, size 328
CRYPTBASE.dll (base address 0x000007FEFD680000, size 60
CLBCatQ.DLL (base address 0x000007FEFEA10000, size 612
MMDevApi.dll (base address 0x000007FEFC160000, size 300
PROPSYS.dll (base address 0x000007FEFC030000, size 1200
AUDIOSES.DLL (base address 0x000007FEF6B70000, size 316
RICHED20.DLL (base address 0x000007FEF1F30000, size 632
nvd3dumx.dll (base address 0x000007FEE9260000, size 17756
PSAPI.DLL (base address 0x0000000077A10000, size 28
avrt.dll (base address 0x000007FEFC2E0000, size 36
----------------------------------------------------------------------
-
In the game Arbalester, music playback is slow (about 50% slower).
The whole game also seems to run a bit slower, but I'm not completely sure.
I did a comparison with the MAME version and, as you reported, the sound playback is slower in the FBA version (moreover, probably as a consequence of the "wrong timing", the music sounds a bit distorted)... otherwise I have not found any noticeable difference in the game speed...
-
I did a comparison with the MAME version and, as you reported, the sound playback is slower in the FBA version (moreover, probably as a consequence of the "wrong timing", the music sounds a bit distorted)... otherwise I have not found any noticeable difference in the game speed...
You are right. I tested it in MAME, and the game speed is as it should be.
The only issue is the slow music playback in FBA.
I hope this can be fixed, I really like using FBA.
-
Arbalester music is fixed :)
best regards,
- dink
-
It happens here, colour going quite monochrome but on a different colour full screen (red) or windowed (green) in-game, while mainly greyish-brown in intermission.
Unfortunately the video emulation for Aquajack isn't completely working. So this game has wrong colors for some things. I'll try to figure out what is going on here - but no promises that this status will change any time soon. This isn't just a FBAlpha problem, it happens in MAME and Raine as well.
best regards,
- dink
-
.... later that night...
How does this look? :biggrin: (second picture is how it looked before)
-
.... later that night...
How does this look? :biggrin: (second picture is how it looked before)
Waah, it looks accurate according to flyers around : http://www.flyerfever.com/post/93471170263/aqua-jack
Good job again dink!
-
Arbalester music is fixed :)
best regards,
- dink
Wow, that was fast.
Thanks for fixing it.
-
.... later that night...
How does this look? :biggrin: (second picture is how it looked before)
Nice fix @dink proper correct colours now in Aqua Jack for the first time in any Emulator.
-
We're not out of the dark yet, the colors in level 4 of Aquajack are messed up, but can't be un-messed up at this point by shuffling data around - but I have another idea that I'm going to explore tonight :)
best regards,
- dink
-
We're not out of the dark yet, the colors in level 4 of Aquajack are messed up, but can't be un-messed up at this point by shuffling data around - but I have another idea that I'm going to explore tonight :)
best regards,
- dink
given that the old colours also look 'ok' (hence nobody noticing they were wrong) I'd suspect it's some kind of pen addition effect, and that maybe both are valid and the game at some point does a 'time of day' effect.
-
Bad news regarding Aquajack color hack:
Unfortunately its not possible to get all levels to look OK without some rather severe and ugly palette hacks. I've decided to just leave it as-is/as it was for now until the problem is a bit better understood. Who knows :rolleyes:
best regards,
- dink
-
While fixing some default controls in fbalpha-libretro i ended up searching for information on the Chase HQ cabinet :
http://www.ukvac.com/forum/taito-chase-hq-fix-tidy-up_topic357660_post994952.html
I'm not 100% sure but it seems to me like the gear is supposed to be some kind of analog control, or at least something with 2 positions. Someone on github also mentioned mame is handling it as an analog control.
Controls in fba look like this :
A("Steering" , BIT_ANALOG_REL, &TaitoAnalogPort0 , "p1 x-axis" ),
{"Brake" , BIT_DIGITAL , TC0220IOCInputPort0 + 5, "p1 fire 1" },
{"Accelerate" , BIT_DIGITAL , TC0220IOCInputPort1 + 5, "p1 fire 2" },
{"Turbo" , BIT_DIGITAL , TC0220IOCInputPort1 + 0, "p1 fire 3" },
{"Gear" , BIT_DIGITAL , TC0220IOCInputPort1 + 4, "p1 fire 4" },
Edit : another issue with controls : p2 coin in backfire doesn't seem to work.
-
barbudreadmon: some games have analog or digital controls, and sometimes we chose to go with the digital control only because it was easier to code in. Tonight when I have more time I'll check to see if this digital control is somehow impeeding gameplay, and if it is I'll add the analog control.
Regarding Backfire, p2 doesn't work properly in any emulated environment. I added a single-screen setting for this game a few days ago. (Wrally2 also has this option, avail. in next version - but both 1p and 2p work)
best regards,
- dink
-
barbudreadmon: some games have analog or digital controls, and sometimes we chose to go with the digital control only because it was easier to code in. Tonight when I have more time I'll check to see if this digital control is somehow impeeding gameplay, and if it is I'll add the analog control.
If i understand well, the Gear control had 2 positions (High and Low : https://www.arcade-museum.com/game_detail.php?game_id=7318), so i'm actually just wondering how the current fba controls handle it : Gear button pressed is High and unpressed is Low ? That would be totally fine, having analog control while the control only support 2 positions seems kinda overboard anyway.
-
If i understand well, the Gear control had 2 positions (High and Low : https://www.arcade-museum.com/game_detail.php?game_id=7318), so i'm actually just wondering how the current fba controls handle it : Gear button pressed is High and unpressed is Low ? That would be totally fine, having analog control while the control only support 2 positions seems kinda overboard anyway.
Now I remember!
Chase HQ's gear shifter uses the new burn_shift device, when the gear button is pressed it toggles between high and low gear, and also writes a little H / L to the bottom right of the screen. It was added in .40.
best regards,
- dink
-
Thanks for the explanation :).
I noticed what appeared as some kind of graphical glitch on "VS Block Breaker" title screen while fixing libretro-fbalpha's default mapping. Didn't check on fba standalone yet though, i'll try to confirm and make a proper report with a screenshot when i get back from work.
-
barbudreadmon, try deleting the nvram / config/games/vblok*.fs, I think that might help in some instances of weirdness with vblokbrk, but if there is still an issue please let me know,
best regards,
- dink
-
This one almost drove me crazy as i wasn't able to reproduce it and was starting to wonder if i dreamed it, but i was finally able to reproduce it.
It is not a nvram issue (got it before and after deleting my nvram), but it only happens if you insert coin at a specific time (see my second screenshot).
I don't think it is worth investigating, as it doesn't seem to impair gameplay.
Edit : the screenshots are from fbalpha-libretro but i'm able to reproduce it on fba standalone : just press coin button during the sfx after my second screenshot
-
barbudreadmon, thanks for the reproduce, I've been trying to track down the cause of that one for ages :)
best regards,
- dink
-
Hi everyone,
I think those games have some input issues :
- deroon/deroona : button 1-4 doesn't seem to do anything, both in menu and in game
- lordgun : i couldn't find a way to reload, i suppose it makes sense since i can't shoot out of the screen.
- most of the "hunting" series (deer, trophy and turkey) seems to have issues with the shot not following the crosshair.
PS : to be honest, i'm mostly interested in deroon, since puzzle game is one of my favorite genre :p.
-
Thanks, derodero and lordofgun have been fixed. Lordofgun reloads when you shoot the left side, but I changed it to reload when you shoot any side.
For the hunting series, you'll have to go into service mode and calibrate the guns (IIRC). If that's not it, please let me know :)
When things aren't so busy I'll probably hook up default calibrations for those games, so they can be played right away.
best regards,
- dink
-
Nice job, thanks :)
-
I took another look at the hunting series, trying to access service mode end up in adding coins.
-
barbudreadmon, toggle service mode in the dips. most of the time "service" means "service coin" and is by default mapped to the 9 key. If a game has a button service mode, it's usually mapped to the f2 key (like in Taito F3). Unfortunately though, there are several games that use a service coin button to enter service mode, or even worse, a toggle dip to set the service mode when a button is needed. I try to fix them as I find them, but its a slow process..
best regards,
- dink
-
Hi friends...
Just wanted to tell that Double Dragon (Special 2017) can't be loaded. Seems to be it is not listed, as mentioned for last .42 release.
Don't know if it is due to that problem reported by chinese forums against Jackc and FBAs clone releases.
-
KaaMoS, there is only 1 version of the romset that will work, so try other versions until you get the right one ;)
The next version of FBA won't support any of this guys hacks, for a few reasons - if you'd like to keep playing them, keep a copy of .42 around just for them. (kof* special, dd special by gsc2007). While he says nothing but bullshit about us that isn't true (we didn't remove any copyright stuff etc). He intentionally changed opcodes and put in bad opcodes so that his hacks only work in his hacked version of MAME. Usually this isn't a problem - but he doesn't supply the source changes to this version of MAME, so like anything else that comes from that end of the world - use extreme caution when running supplied executables. Also, he violates MAME license by not releasing the source - so it's even more pathetic that he complains about us, when all we did was change the bad opcodes so the game would boot.
best regards,
- dink
-
Hi friends...
Just wanted to tell that Double Dragon (Special 2017) can't be loaded. Seems to be it is not listed, as mentioned for last .42 release.
Don't know if it is due to that problem reported by chinese forums against Jackc and FBAs clone releases.
I'm curious what he said. ^^ I haven't heard anything about this.
*edit*
Here's the source to emulate these in FBA
// Double Dragon (Special 2017, hack)
// Modified by: GSC2007
// Version number: Ver 1.0.0311
static struct BurnRomInfo doubledrspRomDesc[] = {
{ "082-p1sp.p1", 0x100000, 0x8ea8ee3d, 1 | BRF_ESS | BRF_PRG }, // 0 68K code
{ "082-p2sp.p2", 0x100000, 0x0e2616ab, 1 | BRF_ESS | BRF_PRG }, // 0 68K code
{ "082-p3sp.p3", 0x020000, 0x8b4839c4, 1 | BRF_ESS | BRF_PRG }, // 0 68K code
{ "082-s1.s1", 0x020000, 0xbef995c5, 2 | BRF_GRA }, // 1 Text layer tiles / TC531000
{ "082-c1.c1", 0x200000, 0xb478c725, 3 | BRF_GRA }, // 2 Sprite data / TC5316200
{ "082-c2.c2", 0x200000, 0x2857da32, 3 | BRF_GRA }, // 3 / TC5316200
{ "082-c3.c3", 0x200000, 0x8b0d378e, 3 | BRF_GRA }, // 4 / TC5316200
{ "082-c4.c4", 0x200000, 0xc7d2f596, 3 | BRF_GRA }, // 5 / TC5316200
{ "082-c5sp.c5", 0x200000, 0xb9c799fe, 3 | BRF_GRA }, // 6 / TC5316200
{ "082-c6sp.c6", 0x200000, 0x11569bc9, 3 | BRF_GRA }, // 7 / TC5316200
{ "082-c7.c7", 0x100000, 0x727c4d02, 3 | BRF_GRA }, // 8 / TC538200
{ "082-c8.c8", 0x100000, 0x69a5fa37, 3 | BRF_GRA }, // 9 / TC538200
{ "082-m1.m1", 0x020000, 0x10b144de, 4 | BRF_ESS | BRF_PRG }, // 10 Z80 code / TC531001
{ "082-v1.v1", 0x200000, 0xcc1128e4, 5 | BRF_SND }, // 11 Sound data / TC5316200
{ "082-v2.v2", 0x200000, 0xc3ff5554, 5 | BRF_SND }, // 12 / TC5316200
};
STDROMPICKEXT(doubledrsp, doubledrsp, neogeo)
STD_ROM_FN(doubledrsp)
static UINT8 *doubledrspExtraROM;
static INT32 DoubledrspInit()
{
INT32 nRet = NeoInit();
if (nRet == 0) {
doubledrspExtraROM = (UINT8*)BurnMalloc(0x20000);
if (BurnLoadRom(doubledrspExtraROM, 2, 1)) return 1;
UINT16 *rom = (UINT16*)doubledrspExtraROM;
for (INT32 i = 0; i < 0x20000/2; i++) {
if (rom[i] == 0x4e7d) rom[i] = 0x4e71;
if (rom[i] == 0x4e7c) rom[i] = 0x4e75;
}
rom = (UINT16*)Neo68KROMActive;
for (INT32 i = 0; i < 0x100000/2; i++) {
if (rom[i] == 0x4e7d) rom[i] = 0x4e71;
if (rom[i] == 0x4e7c) rom[i] = 0x4e75;
}
rom[0xbff2] = 0x2b7c; // 4ef9
rom[0xbff3] = 0x0001; // 0091
rom[0xbff4] = 0x7fee; // 0206
rom[0xbff5] = 0xa26a; // 4e7d
SekOpen(0);
SekMapMemory(doubledrspExtraROM, 0x900000, 0x91ffff, MAP_ROM);
SekClose();
}
return nRet;
}
static INT32 DoubledrspExit()
{
BurnFree (doubledrspExtraROM);
return NeoExit();
}
struct BurnDriver BurnDrvdoubledrsp = {
"doubledrsp", "doubledr", "neogeo", NULL, "2017",
"Double Dragon (Special 2017, hack)\0", NULL, "hack", "Neo Geo MVS",
NULL, NULL, NULL, NULL,
BDF_GAME_WORKING | BDF_CLONE, 2, HARDWARE_PREFIX_CARTRIDGE | HARDWARE_SNK_NEOGEO, GBF_VSFIGHT, 0,
NULL, doubledrspRomInfo, doubledrspRomName, NULL, NULL, neogeoInputInfo, neogeoDIPInfo,
DoubledrspInit, DoubledrspExit, NeoFrame, NeoRender, NeoScan, &NeoRecalcPalette,
0x1000, 320, 224, 4, 3
};
// The King of Fighters '95 (Special 2017)
// Modified by: GSC2007
// Version number: Ver 1.0.0627
static struct BurnRomInfo kof95spRomDesc[] = {
{ "084-p1sp.p1", 0x100000, 0x8bcca012, 1 | BRF_ESS | BRF_PRG }, // 0 68K code
{ "084-p2sp.p2", 0x100000, 0x5cb1af9e, 1 | BRF_ESS | BRF_PRG }, // 1
{ "084-p3sp.p3", 0x020000, 0x030e71cb, 1 | BRF_ESS | BRF_PRG }, // 2
{ "084-s1sp.s1", 0x020000, 0x83cbae60, 2 | BRF_GRA }, // 3 Text layer tiles / TC531000
{ "084-c1.c1", 0x400000, 0xfe087e32, 3 | BRF_GRA }, // 4 Sprite data / TC5332202
{ "084-c2.c2", 0x400000, 0x07864e09, 3 | BRF_GRA }, // 5 / TC5332202
{ "084-c3.c3", 0x400000, 0xa4e65d1b, 3 | BRF_GRA }, // 6 / TC5332202
{ "084-c4.c4", 0x400000, 0xc1ace468, 3 | BRF_GRA }, // 7 / TC5332202
{ "084-c5.c5", 0x200000, 0x8a2c1edc, 3 | BRF_GRA }, // 8 / TC5316200
{ "084-c6.c6", 0x200000, 0xf593ac35, 3 | BRF_GRA }, // 9 / TC5316200
{ "084-c7.c7", 0x100000, 0x9904025f, 3 | BRF_GRA }, // 10 / TC538200
{ "084-c8.c8", 0x100000, 0x78eb0f9b, 3 | BRF_GRA }, // 11 / TC538200
{ "084-m1.m1", 0x020000, 0x6f2d7429, 4 | BRF_ESS | BRF_PRG }, // 12 Z80 code / TC531001
{ "084-v1.v1", 0x400000, 0x84861b56, 5 | BRF_SND }, // 13 Sound data / TC5332201
{ "084-v2.v2", 0x200000, 0xb38a2803, 5 | BRF_SND }, // 14 / TC5316200
{ "084-v3.v3", 0x100000, 0xd683a338, 5 | BRF_SND }, // 15 / TC538200
};
STDROMPICKEXT(kof95sp, kof95sp, neogeo)
STD_ROM_FN(kof95sp)
static UINT8 *kof95spExtraROM;
static INT32 Kof95spInit()
{
INT32 nRet = NeoInit();
if (nRet == 0) {
kof95spExtraROM = (UINT8*)BurnMalloc(0x20000);
if (BurnLoadRom(kof95spExtraROM, 2, 1)) return 1;
UINT16 *rom = (UINT16*)kof95spExtraROM;
for (INT32 i = 0; i < 0x20000/2; i++) {
if (rom[i] == 0x4e7d) rom[i] = 0x4e71;
if (rom[i] == 0x4e7c) rom[i] = 0x4e75;
}
rom = (UINT16*)Neo68KROMActive;
for (INT32 i = 0; i < 0x100000/2; i++) {
if (rom[i] == 0x4e7d) rom[i] = 0x4e71;
if (rom[i] == 0x4e7c) rom[i] = 0x4e75;
}
rom[0x1f3a8] = 0x2b7c; // 4ef9
rom[0x1f3a9] = 0x0003; // 0090
rom[0x1f3aa] = 0xe7fa; // 16be
SekOpen(0);
SekMapMemory(kof95spExtraROM, 0x900000, 0x91ffff, MAP_ROM);
SekClose();
}
return nRet;
}
static INT32 Kof95spExit()
{
BurnFree (kof95spExtraROM);
return NeoExit();
}
struct BurnDriver BurnDrvKof95sp = {
"kof95sp", "kof95", "neogeo", NULL, "2017",
"The King of Fighters '95 (Special 2017, hack)\0", NULL, "hack", "Neo Geo MVS",
NULL, NULL, NULL, NULL,
BDF_GAME_WORKING | BDF_CLONE | BDF_HACK, 2, HARDWARE_PREFIX_CARTRIDGE | HARDWARE_SNK_NEOGEO, GBF_VSFIGHT, FBF_KOF,
NULL, kof95spRomInfo, kof95spRomName, NULL, NULL, neogeoInputInfo, neogeoDIPInfo,
Kof95spInit, Kof95spExit, NeoFrame, NeoRender, NeoScan, &NeoRecalcPalette,
0x1000, 304, 224, 4, 3
};
As you can see, all that was done to "protect" these was change all instances of 0x4e71 to 0x4e7d, 0x4e75 to 0x4e7c and mess up a jump to a sub that I patched out.
Anywho, I admire the gall of someone (GSC2007) that takes someone else's work (SNK) makes minor changes and then passes it off as their own and gets pissed off when other people use it how they want (ME).
-
Hi everyone,
Got a report on libretro-fbalpha mentioning player 3 can't add coins on ctribe, and it seems there is no input info indeed in the ddragon3 driver for this button :
static struct BurnInputInfo DrvInputList[] =
{
{"P1 Coin" , BIT_DIGITAL , DrvInputPort1 + 0, "p1 coin" },
{"P1 Start" , BIT_DIGITAL , DrvInputPort0 + 7, "p1 start" },
{"P1 Up" , BIT_DIGITAL , DrvInputPort0 + 2, "p1 up" },
{"P1 Down" , BIT_DIGITAL , DrvInputPort0 + 3, "p1 down" },
{"P1 Left" , BIT_DIGITAL , DrvInputPort0 + 1, "p1 left" },
{"P1 Right" , BIT_DIGITAL , DrvInputPort0 + 0, "p1 right" },
{"P1 Fire 1" , BIT_DIGITAL , DrvInputPort0 + 4, "p1 fire 1" },
{"P1 Fire 2" , BIT_DIGITAL , DrvInputPort0 + 5, "p1 fire 2" },
{"P1 Fire 3" , BIT_DIGITAL , DrvInputPort0 + 6, "p1 fire 3" },
{"P2 Coin" , BIT_DIGITAL , DrvInputPort1 + 1, "p2 coin" },
{"P2 Start" , BIT_DIGITAL , DrvInputPort2 + 7, "p2 start" },
{"P2 Up" , BIT_DIGITAL , DrvInputPort2 + 2, "p2 up" },
{"P2 Down" , BIT_DIGITAL , DrvInputPort2 + 3, "p2 down" },
{"P2 Left" , BIT_DIGITAL , DrvInputPort2 + 1, "p2 left" },
{"P2 Right" , BIT_DIGITAL , DrvInputPort2 + 0, "p2 right" },
{"P2 Fire 1" , BIT_DIGITAL , DrvInputPort2 + 4, "p2 fire 1" },
{"P2 Fire 2" , BIT_DIGITAL , DrvInputPort2 + 5, "p2 fire 2" },
{"P2 Fire 3" , BIT_DIGITAL , DrvInputPort2 + 6, "p2 fire 3" },
{"P3 Start" , BIT_DIGITAL , DrvInputPort3 + 7, "p3 start" },
{"P3 Up" , BIT_DIGITAL , DrvInputPort3 + 2, "p3 up" },
{"P3 Down" , BIT_DIGITAL , DrvInputPort3 + 3, "p3 down" },
{"P3 Left" , BIT_DIGITAL , DrvInputPort3 + 1, "p3 left" },
{"P3 Right" , BIT_DIGITAL , DrvInputPort3 + 0, "p3 right" },
{"P3 Fire 1" , BIT_DIGITAL , DrvInputPort3 + 4, "p3 fire 1" },
{"P3 Fire 2" , BIT_DIGITAL , DrvInputPort3 + 5, "p3 fire 2" },
{"P3 Fire 3" , BIT_DIGITAL , DrvInputPort3 + 6, "p3 fire 3" },
{"Reset" , BIT_DIGITAL , &DrvReset , "reset" },
{"Service" , BIT_DIGITAL , DrvInputPort1 + 2, "service" },
{"Dip 1" , BIT_DIPSWITCH, DrvDip + 0 , "dip" },
{"Dip 2" , BIT_DIPSWITCH, DrvDip + 1 , "dip" },
{"Dip 3" , BIT_DIPSWITCH, DrvDip + 2 , "dip" },
{"Dip 4" , BIT_DIPSWITCH, DrvDip + 3 , "dip" },
};
-
For the Combatribes there are only 2 coin inputs. For 3 player, set the dips to "# of players:" 3, reset, put coins into any slot and press 1p 2p 3p start to start the players :)
best regards,
- dink
-
Hello folks,
In Kaiser Knuckle game, the kick buttons doesn't work, at least here, in my computer.
-
Bad Dude,
For Kaiser Knuckle, you'll have to press F5 and map buttons 4 5 and 6 to buttons you wish to use for kicking.
best regards,
- dink
-
KaaMoS, there is only 1 version of the romset that will work, so try other versions until you get the right one ;)
The next version of FBA won't support any of this guys hacks, for a few reasons - if you'd like to keep playing them, keep a copy of .42 around just for them. (kof* special, dd special by gsc2007). While he says nothing but bullshit about us that isn't true (we didn't remove any copyright stuff etc). He intentionally changed opcodes and put in bad opcodes so that his hacks only work in his hacked version of MAME. Usually this isn't a problem - but he doesn't supply the source changes to this version of MAME, so like anything else that comes from that end of the world - use extreme caution when running supplied executables. Also, he violates MAME license by not releasing the source - so it's even more pathetic that he complains about us, when all we did was change the bad opcodes so the game would boot.
best regards,
- dink
Good job! I really feel digusted with those bastards. They can not politely talk at all. It's good to remove all drivers of their crap.
-
Hi everyone,
i think moomesa, bucky and their clones got some issues with 2+ players, i couldn't get player 3 and 4 to join whatever the dipswitch setting and the coin/start buttons i pressed.
-
Good job! I really feel digusted with those bastards. They can not politely talk at all. It's good to remove all drivers of their crap.
Thank you for your support :)
-
Thanks for the bug report, here's a fix. Just change the 8 in line #1210 to a 16 - "for (INT32 i = 0; i < 16; i++) {"
Hi everyone,
i think moomesa, bucky and their clones got some issues with 2+ players, i couldn't get player 3 and 4 to join whatever the dipswitch setting and the coin/start buttons i pressed.
-
Thanks for the bug report, here's a fix. Just change the 8 in line #1210 to a 16 - "for (INT32 i = 0; i < 16; i++) {"
You mean like this in d_moo.cpp ?
for (INT32 i = 0; i < 16; i++) {
DrvInputs[0] ^= (DrvJoy1[i] & 1) << i;
DrvInputs[1] ^= (DrvJoy2[i] & 1) << i;
DrvInputs[2] ^= (DrvJoy3[i] & 1) << i;
DrvInputs[3] ^= (DrvJoy4[i] & 1) << i;
}
It was on line 1101 for me though, i guess there are new things in this driver for next release then :)
-
barbudreadmon: that'll do :)
-
Kaiser Knuckle kick inputs have wrong mappings (d_taitof3.cpp), that's why he can't bind the buttons. These are the correct ones
{"P1 Button 4", BIT_DIGITAL, DrvJoy4 + 0, "p1 fire 4"},
{"P1 Button 5", BIT_DIGITAL, DrvJoy4 + 1, "p1 fire 5"},
{"P1 Button 6", BIT_DIGITAL, DrvJoy4 + 2, "p1 fire 6"},
{"P2 Button 4", BIT_DIGITAL, DrvJoy3 + 8, "p2 fire 4"},
{"P2 Button 5", BIT_DIGITAL, DrvJoy3 + 9, "p2 fire 5"},
{"P2 Button 6", BIT_DIGITAL, DrvJoy3 + 10, "p2 fire 6"},
Bad Dude,
For Kaiser Knuckle, you'll have to press F5 and map buttons 4 5 and 6 to buttons you wish to use for kicking.
best regards,
- dink
-
shine, thanks for the fix info!
best regards,
- dink
-
Game: Ex-Mutants.
Problem: The screen is freezing at the begining of the game.
-
Hi,author,it seemed this emulation can't play rom which named kovplus consummately,even if patched the corresponding repire ips. Specific bug is the secret room of bamboo grove can not go out .May i ask how to resolve it?
-
Game: Ex-Mutants.
Problem: The screen is freezing at the begining of the game.
This game was fixed when I did some Megadrive improvements a little while back, it will be available in the next version :)
best regards,
- dink
-
Hi :smilie:
No Sound in "Toukidenshou - Angel Eyes".
(https://www.canalicara.com/upload/photos/eventos2013/20130130_touki/8.jpg)
Playing today this game and no have sound :(
Make little test in MAME and sound works.
-
Hi :smilie:
No Sound in "Toukidenshou - Angel Eyes".
(https://www.canalicara.com/upload/photos/eventos2013/20130130_touki/8.jpg)
Playing today this game and no have sound :(
Make little test in MAME and sound works.
Not a bug as such more a case of FBA does not support the sound core this game and Deron DeroDero require.
-
Not a bug as such more a case of FBA does not support the sound core this game and Deron DeroDero require.
Ok :( Thanks for info :)
Sound core is YMF262 and YMZ280B. Right?.
-
Ok :( Thanks for info :)
Sound core is YMF262 and YMZ280B. Right?.
Yes... and, to be more accurate, you need to add an OKI6295 (an ADPCM sound chip) and a Z80 (the CPU that manages the audio sub-system)... :)
-
re: derodero/angel eyes,
the reason we can't get it partially working with the soundchips we have is that the 262 generates the interrupts for the z80.
-
This game was fixed when I did some Megadrive improvements a little while back, it will be available in the next version :)
best regards,
- dink
So, maybe you know that the game Galahad (Legend of Galahad) have some issues too. Check it out.
Dink, the next version of FBA will be release sooner?
-
The good news is Galahad is completely fixed :) Usually a new exe is released every 3 or 6 months, depending on what is going on.
best regards,
- dink
So, maybe you know that the game Galahad (Legend of Galahad) have some issues too. Check it out.
Dink, the next version of FBA will be release sooner?
-
The good news is Galahad is completely fixed :) Usually a new exe is released every 3 or 6 months, depending on what is going on.
best regards,
- dink
Good news. A new FBA is coming up. Seizing the opportunity, I think will be a great thing for the emulador the
addition of the arcade system used to preview Super NES called Nintendo Super System (NSS). List of games:
Thanks. Keep the work!
-
Well, we aren't taking requests right now, sorry about that.
-
There could be some issue with savestates in the konami mystwarr driver, probably some pointer included in it. See https://github.com/libretro/fbalpha/issues/159
What i did to reproduce this issue in fba standalone is saving state while in ram rom check, then reload with the state.
-
I spent some time making serveral savestates at different points in time while the ram/rom check screen was up in Violent Storm, but every time I loaded the state it checked out fine afterwords.
[A few moments later] I also tried with the release version in a fresh setup. hrmm, maybe I'm missing something? :)
best regards,
- dink
There could be some issue with savestates in the konami mystwarr driver, probably some pointer included in it. See https://github.com/libretro/fbalpha/issues/159
What i did to reproduce this issue in fba standalone is saving state while in ram rom check, then reload with the state.
-
Check "restore full state on load" and use this one.
-
Hello people.
I realise interesting Bug on Cow Boys of Moo Mesa.
I just tried the game (moomesa.zip ) in mame. After i tried in FBA. In the FBA there is some scene missing.
This screen is from the game's intro
https://i.hizliresim.com/nJdpY0.png (https://i.hizliresim.com/nJdpY0.png)
But in FBA there is only black screen in this screen.
-
Hello people.
I realise interesting Bug on Cow Boys of Moo Mesa.
I just tried the game (moomesa.zip ) in mame. After i tried in FBA. In the FBA there is some scene missing.
This screen is from the game's intro
https://i.hizliresim.com/nJdpY0.png (https://i.hizliresim.com/nJdpY0.png)
But in FBA there is only black screen in this screen.
That should be correct. It's part of the FBA team's policy of censoring giant Lazer beams.
-
Hi hursit,
Thanks for the bug report!
This problem in Moo Mesa has been fixed along with the end-level boss scene, the problems were fixed last week but I didn't announce it yet :)
best regards,
- dink
Hello people.
I realise interesting Bug on Cow Boys of Moo Mesa.
I just tried the game (moomesa.zip ) in mame. After i tried in FBA. In the FBA there is some scene missing.
This screen is from the game's intro
https://i.hizliresim.com/nJdpY0.png (https://i.hizliresim.com/nJdpY0.png)
But in FBA there is only black screen in this screen.
-
barbudreadmon, thanks for all the info regarding that bug in violent storm, please check your pm for a possible solution :)
best regards,
- dink
-
Hi everyone,
Someone at github reported an issue with sounds on neogeo : it seems some games have missing sounds when you select a mvs usa bios or the usa+arcade setting with the unibios. The games reported are bstars (no music when the game start), nam1975 (no sound at all ingame), karnovr (no sound in selection screen + ingame), and some clones, i don't know if it affects other games. I can confirm it happens with official fbalpha.
Is that a bug ? I've been wondering if those sounds aren't just missing from the usa version of those games.
-
Is that a bug ? I've been wondering if those sounds aren't just missing from the usa version of those games.
Hmm i'd go with a USA Bio's bug, as surely too many sounds are missing to have been a regional change.
-
barbudreadmon, its a bug in fba. if you uncomment the #undef speedhacks at the top of neo_run.cpp (see // #undef SPEED_HACKS) it will work. I think it would be a good idea to convert that to a dip setting like you did in the other drivers, if you want. (or, if the games work fine on rpi with speedhacks disabled, just leave them disabled?) What do you think?
best regards,
- dink
-
barbudreadmon, its a bug in fba. if you uncomment the #undef speedhacks at the top of neo_run.cpp (see // #undef SPEED_HACKS) it will work. I think it would be a good idea to convert that to a dip setting like you did in the other drivers, if you want. (or, if the games work fine on rpi with speedhacks disabled, just leave them disabled?) What do you think?
best regards,
- dink
The raspberry 3 should be fine with those speedhacks disabled, i'm wondering about older/slower devices though. I'll make some test and let you know by pm the results.
-
The raspberry 3 should be fine with those speedhacks disabled, i'm wondering about older/slower devices though. I'll make some test and let you know by pm the results.
That Neo Geo speedhack is nothing but trouble. It doesn't even really do anything noticeable for most games. Maybe it should just be removed.
-
The raspberry 3 should be fine with those speedhacks disabled, i'm wondering about older/slower devices though. I'll make some test and let you know by pm the results.
Excellent, yea, it would be nice to remove them or at least give an option. The thing is, people have been raving about our neogeo support for years, and I really hope disabling this speedhack doesn't uncover anything more serious or cause problems down the line. Would you try some previously problematic games, like s1945p, and see if there is any regression with the speedhack off?
thanks,
- dink
-
I'm observing a ~2% performance decrease without the speedhack (i underclocked my rpi3, checked the fps with and without the speedhack, it went from ~50 to ~49 fps), and s1945p is running fine when i'm back at normal clock, i don't think this speedhack is worth keeping.
-
IIRC the speed hack was put in for s1945p specifically. in the stone age when cpu speed was still measured in MHz. (That game talks to the sound cpu differently from other neo geo games, and requires the emulation to switch back and forth between the emulated cpus a lot).
-
Actually, with or without the speedhack, karnovr doesn't seem to have music in selection screen and ingame when using mvs bios. aes and unibios seems fine though.
-
Actually, with or without the speedhack, karnovr doesn't seem to have music in selection screen and ingame when using mvs bios. aes and unibios seems fine though.
I remember fixing this problem in karnovr, but, I don't know where the fix is. I checked the entire neogeo svn history and I don't see it. IIRC it was something to do with irq's getting skipped. If it comes back to me, I'll say something :) but verify: 2610 soundcore & interface, z80 interface and cpu core are all up to date. It seems like the only thing I can find relating to it is: http://http://neo-source.com/index.php?topic=2879.msg21278#msg21278
best regards,
- dink
-
I remember fixing this problem in karnovr, but, I don't know where the fix is. I checked the entire neogeo svn history and I don't see it. IIRC it was something to do with irq's getting skipped. If it comes back to me, I'll say something :) but verify: 2610 soundcore & interface, z80 interface and cpu core are all up to date. It seems like the only thing I can find relating to it is: http://http://neo-source.com/index.php?topic=2879.msg21278#msg21278
best regards,
- dink
Could it be related to those 2 commit ?
https://github.com/libretro/fbalpha/commit/c197019873817680fda27d8e8fbdb9b5e5129e5c
https://github.com/libretro/fbalpha/commit/310d3e2a8d9eb65e2495faae28302f497266ecb2
That's the 2 hotfix you gave me for s1945p in february, Karnov's irq issue was mentioned in the code before the changes.
-
barbudreadmon, ahh yes - now I remember - after properly fixing the Z80's ZetRunEnd() also fixed karnovr (without having to disable ZetRunEnd() of course). I can confirm karnovr is working in FBA with or without speedhacks using the default dips, though. I wish I had some helpfull information for you, but, this is a strange one.
best regards,
- dink
-
barbudreadmon, ahh yes - now I remember - after properly fixing the Z80's ZetRunEnd() also fixed karnovr (without having to disable ZetRunEnd() of course). I can confirm karnovr is working in FBA with or without speedhacks using the default dips, though. I wish I had some helpfull information for you, but, this is a strange one.
best regards,
- dink
Hmm the neogeo driver in lr-fbalpha should be in sync with the one in standalone fbalpha, so perhaps some kind of issue related to the compiler/cflags ? Did you use gcc or msvc for your test ? And if gcc, which version ? I'll also try without the -O3 cflag when i'm back home.
-
Hmm the neogeo driver in lr-fbalpha should be in sync with the one in standalone fbalpha, so perhaps some kind of issue related to the compiler/cflags ? Did you use gcc or msvc for your test ? And if gcc, which version ? I'll also try without the -O3 cflag when i'm back home.
The ZetRunEnd fix would be in the Z80 Interface/Core though - so external to the driver itself....
-
Hmm the neogeo driver in lr-fbalpha should be in sync with the one in standalone fbalpha, so perhaps some kind of issue related to the compiler/cflags ? Did you use gcc or msvc for your test ? And if gcc, which version ? I'll also try without the -O3 cflag when i'm back home.
I'm using gcc 4.8.1 with -O3 here, I wonder about this :/ do any errors/warnings come up when compiling? can you try asan-ing it?
-
My bad, it seems an old karnovr.fs was at fault here, i should have tried removing it first.
-
No problem, I'm relieved to know it was something simple :)
best regards,
- dink
-
I was wrong, removing karnovr.fs didn't solve this issue, and i think i was finally able to reproduce this issue consistently :
1. start the game (any mvs or unibios arcade)
2. insert coin so that the game go to the title screen
3. wait for title screen music to start, immediately insert coin => music stop
4. process further by pressing start, there is no music
-
barbudreadmon, I can't reproduce that here at all using MVS USA ver.5 2slot, MVS Asia Europe ver.6 2slot, Universe Bios v3.0. I tried it 10 times for each bios just to make sure. These tests were made with "#undef USE_SPEEDHACKS". I also tried twice with Unibios and the default "MVS Asia/Europe ver.6" with the speedhacks enabled, and still no problem. I'm really not sure what's going on :/
best regards,
- dink
-
Weirdest issue ever :
- The guy who reported this get no sound when in non-aes mode, whatever he does
- I get music as long as i don't press coin when the title screen music is running
- You don't have this issue at all.
I'll try to get to the bottom of this, perhaps using different gcc version with various sanitizer will help.
Also, this is the other thing that comes to mind, perhaps it is worth mentioning the audio settings described here (http://http://neo-source.com/index.php?topic=2487.msg22811#msg22811) are hardcoded in the libretro port.
-
- Same issue with all audio sample rate and interpolation settings i tried
- Same issue with every arch and gcc i tried (arm x86_64, gcc-4 gcc-5 gcc-6, sadly i don't have a x86_32 test bed)
- Same issue with any c(xx)flags i tried
- Asan didn't say anything
- Ubsan said the following but i remember you told me it was normal behavior from the m68k (every m68k game output the same iirc) :
src/cpu/m68000_intf.cpp:394:48: runtime error: load of misaligned address 0x0000021e151a for type 'UINT32', which requires 4 byte alignment
0x0000021e151a: note: pointer points here
02 00 f9 4e c1 00 02 00 f9 4e c1 00 02 00 f9 4e c1 00 02 00 f9 4e c1 00 02 00 f9 4e c1 00 02 00
^
src/cpu/m68000_intf.cpp:457:39: runtime error: store to misaligned address 0x000002294ede for type 'UINT32', which requires 4 byte alignment
0x000002294ede: note: pointer points here
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
^
src/cpu/m68000_intf.cpp:413:47: runtime error: load of misaligned address 0x0000021f148a for type 'UINT32', which requires 4 byte alignment
0x0000021f148a: note: pointer points here
04 00 d0 4e c1 00 96 03 c1 00 32 04 c1 00 9a 04 c1 00 52 04 c1 00 c8 04 c1 00 e2 04 c1 00 fc 04
^
src/burn/snd/fm.c:1526:65: runtime error: left shift of negative value -577
src/burn/snd/fm.c:1147:68: runtime error: left shift of negative value -581
src/cpu/m68k/m68kops.c:8228:16: runtime error: left shift of 1 by 31 places cannot be represented in type 'int'
The first 5 appear on bios screen, the last one when you enter select character screen, none appear on the title screen
All tests with the speedhack disabled. Deleted my karnovr.fs file a couple of times too.
At this point i don't know what to do anymore.
For what it's worth, i captured a video at default settings, in case something comes to mind.
-
barbudreadmon, Just a guess, but this might be the problem:
src/cpu/m68k/m68kops.c:8228:16: runtime error: left shift of 1 by 31 places cannot be represented in type 'int'
code: uint mask = 1 << (OPER_I_8() & 0x1f);
uint is defined as "unsigned int" in m68kcpu.h. that should be ok. 1<<31 or 1<<0x1f is the top bit in a 32bit int, so that should also be ok. But if it can't set the top bit when it needs to - there's going to be a problem.
try to get rid of that error. Some things I can think of... uint mask = ((uint) 1 << (OPER_I_8() & 0x1f)); .. maybe? Does it work if you disable optimization for the 68k cpu? if not, try disabling optimization for everything that neogeo requires (68k, z80, snd/*2610*, fm.c, drv/neogeo/*) and see if that helps.
bye for now,
- dink
-
barbudreadmon, Just a guess, but this might be the problem:
src/cpu/m68k/m68kops.c:8228:16: runtime error: left shift of 1 by 31 places cannot be represented in type 'int'
code: uint mask = 1 << (OPER_I_8() & 0x1f);
uint is defined as "unsigned int" in m68kcpu.h. that should be ok. 1<<31 or 1<<0x1f is the top bit in a 32bit int, so that should also be ok. But if it can't set the top bit when it needs to - there's going to be a problem.
try to get rid of that error. Some things I can think of... uint mask = ((uint) 1 << (OPER_I_8() & 0x1f)); .. maybe? Does it work if you disable optimization for the 68k cpu? if not, try disabling optimization for everything that neogeo requires (68k, z80, snd/*2610*, fm.c, drv/neogeo/*) and see if that helps.
bye for now,
- dink
It removed the ubsan message, but it didn't solve the issue. I already tried disabling all optimization. I also made sure differences in codebase between lr-fbalpha and fbalpha were not the root of this, and there was no difference apart from the "s1945pmode" hotfix and a change to "if (nNeoSMARNGAddress[nNeoActiveSlot][0] > 0 || nNeoSMARNGAddress[nNeoActiveSlot][1] > 0) {" in "NeoSMABankswitch", which i think is part of your current svn version (tried reverting to original .42 code, didn't help), there were no change in 68k, z80, 2610 and fm code.
At this point, 2 things comes to mind :
- An issue in the libretro frontend part of the code, not finding similar issues in other games is weird though
- An issue in the codebase for non standard x86_32 platform, for cross-platform compatibility reasons lr-fbalpha is built without fastcall (works only on x86_32), x86_asm (same ?), mmx (x86 only), and c++11 (no compatible compiler for ios, ps3, ...)
I'll try building fbalpha standalone without speedhack, fastcall, x86_asm, mmx and c++11 to see if i can reproduce this issue.
-
FWIW, my older builds have the same problem in karnovr with a different Z80 core. But blazstar's intro music is only missing in the new ones, just as the regular fba release.
So it looks like there are several problems. My builds have the same neo geo code as the older ones. The timers haven't changed, but the Z80 and YM2610 cores have, as has the Z80 interface code. So, karnovr's problem is probably in neo_run.cpp somewhere. but blazstar's isn't, at least not directly.
c++11 (no compatible compiler for ios, ps3, ...)
Say what? IOS has had a current version version of CLang for years, included in XCode. PS3 is another matter entirely, though unofficial devkits definitely support C++11, and much of C++14.
-
Say what? IOS has had a current version version of CLang for years, included in XCode. PS3 is another matter entirely, though unofficial devkits definitely support C++11, and much of C++14.
Well, i am kinda sure being asked to remove c++11 dependency for ios compatibility, it was probably 1 or 2 years ago though, perhaps it was for older device (support for newer ios version was dropped for iphone 4 and below a few years ago if i remember well).
Considering my experience as a web developper with apple devices (safari doesn't support a lot of standard w3c stuff other browser have been supporting for 5+ years), and the fact opengl 4.2+ support (2011) was still missing a few months ago (and perhaps is still missing) on mac os x, i never thought it was strange.
There are probably other platforms not compatible with c++11, iirc lr-fbalpha is also used on psvita, 3ds, wiiu, ...
-
barbudreadmon, if there is some way you could make this problem happen in the standalone version, I could help better.
p,s, also add xbox to the list of those that don't do current c++.
best regards,
- dink
-
Well, i am kinda sure being asked to remove c++11 dependency for ios compatibility, it was probably 1 or 2 years ago though, perhaps it was for older device (support for newer ios version was dropped for iphone 4 and below a few years ago if i remember well).
Yeah, apple is fairly ruthless where backward compatibility is concerned (c++11 support arrived before that cutoff, though).
Considering my experience as a web developper with apple devices (safari doesn't support a lot of standard w3c stuff other browser have been supporting for 5+ years), and the fact opengl 4.2+ support (2011) was still missing a few months ago (and perhaps is still missing) on mac os x, i never thought it was strange.
Don't get me talking about that... I do mostly back-end web stuff, but still sometimes the refusal of the browser makers (not just Apple) to even acknowledge big parts of the standards makes me want to throw stuff.
There are probably other platforms not compatible with c++11, iirc lr-fbalpha is also used on psvita, 3ds, wiiu, ...
Right, I'm not expecting platforms that were discontinued and unsupported by 2011 to ever support c++11. My approach is to not use c++11 in burn, but some stuff I write for other parts of fba relies on it extensively.
Anyway, good luck with the sound problem.
-
I couldn't reproduce it on standalone, but i noticed something while testing : on my video, when i reach the title screen by "inserting coin", the sound from inserting this first coin is weird, i hear it twice (i pressed coin button only once to reach the title screen). On standalone, i hear it once. I tried a couple of other neogeo games on lr-fbalpha and didn't notice anything similar, so i was wondering if it's not a hint to the issue ?
I was wondering if it couldn't be related to something in the code defining audio segment length and audio buffer in my libretro port.
My audio buffer is declared like this :
static int16_t *g_audio_buf;
(that's the type the libretro api audio callback is asking for)
I have a macro defined like this :
#define AUDIO_SAMPLERATE 48000
Before running BurnDrvInit i do this :
nAudSegLen = (AUDIO_SAMPLERATE * 100 + (6000 >> 1)) / 6000;
g_audio_buf = (int16_t*)malloc(nAudSegLen<<2 * sizeof(int16_t));
pBurnSoundOut = g_audio_buf;
nBurnSoundRate = AUDIO_SAMPLERATE;
nBurnSoundLen = nAudSegLen;
After running it i do this :
nAudSegLen = (AUDIO_SAMPLERATE * 100 + (nBurnFPS >> 1)) / nBurnFPS;
g_audio_buf = (int16_t*)malloc(nAudSegLen<<2 * sizeof(int16_t));
nBurnSoundRate = AUDIO_SAMPLERATE;
nBurnSoundLen = nAudSegLen;
(redefining nBurnSoundRate was probably not necessary)
Anything in this is likely to cause an issue ?
-
Your calculation for the number of samples in a frame is fine (identical to mine).
In your case, I would see if the coin sound is triggered twice because it's getting the coin insert inputs twice, the 68k is sending the sound command twice, and so on.
-
Hi,
I'm part of libretro, on 0.2.97.39 I did a lot of research with aliaspider to fix savestates between popular platforms for some popular machines (because netplay).
Turns out several states included pointers and we added some padding which in turn fixed the issue.
barbudreadmon backported 0.2.97.40 and now everything seems broken again:
- CPSx states get fudged between 64-bit and 32-bit platforms (music glitches mostly)
- NeoGeo states are different size again
Do you have any ideas regarding these?
I can try to fix them again but I wonder if we can avoid these in the future.
If at all possible, do you think we can pull from SVN directly? is the SVN public? (at least for reading?)
Thanks in advance for your responses
-
radius,
I think we did some padding using a compiler macro to prevent different sized states between 32/64bit systems. I only have a 32bit system, so I can't really test it right now. I'd better let barbudreadmon respond here, as he knows more about the libretro port :) Our svn is private right now, but that may change in the future.
best regards,
- dink
-
I'm not sure it is related to the bump to .40 release, as i mentioned in https://github.com/libretro/fbalpha/issues/166, it is more likely an issue with changing int to intptr_t in m68kcpu.c's m68k_context_size_no_pointers function in https://github.com/libretro/fbalpha/commit/0d21221fe9030ca8c482ca9042f92babf0a52dde .
From my understanding of "intptr_t", it doesn't return the same in 32/64 bits, and i kinda remember there was a warning related to this function and dink explicitly told me not to mess with it if i didn't want to break savestates, but someone else did and i overlooked this.
Edit : it would also be worth checking what was done on http://neo-source.com/index.php?topic=3162.0 , jan's special version mention being fully cross-platform for savestates. I don't think i'll have time to deal with bugs this week (wait for next week, i took leave), so please be patient, or try reverting the 2 changes i mentioned in https://github.com/libretro/fbalpha/issues/166 (the one with intptr_t and the one with ACB_FULLSCAN) and tell me if it fixes this issue.
-
Correct -- intptr_t returns an int the size of a pointer on the current platform.
Unfortunately what I meant by cross-platform there is that the code in burner is cross-platform. It still gets the same blobs of data from burn so that old savestates can be loaded or converted. Any 32/64bit or endianness issues that remain in burn need to be fixed there, which may break savestates (input recordings need the endianness converted on BE systems but should otherwise work).
The new fileformat code itself can deal with endianness and has string conversions (all strings in the files are utf8), so that all files can be read and understood on any platform. It also uses the boost filesystem library (now part of the C++ standard, since C++17) to deal with file system in a cross-platform manner. It also has C++14 template stuff. The fixes burn needs are easy enough to do with the same C++14 template stuff, but portability...
-
I'd like to help, but I don't have a 64bit system handy for testing.
Also, got some bad news: the next version has lots of updated cores, and with updates come more critical variable scanning. This means .42 states won't be compatible with .43. I had a dream about a new savestate format that would overcome problems like this, but we would probably have to do some collaboration to make it happen.
I'm guessing that right now everything is saved in a binary blob, and if something is added or missing in-between, everything below it goes a-skew.
The idea is, state data could be saved in a format like:
var_name_token,size_of_data,data
if the size doesn't match the current size, it would be skipped. Also, if a new variable is added but the data in the savestate doesn't exist, it would get skipped.
Of course, Jan's special version could do this in the latest incarnation of C++, with a ton of add-on libraries, but, that would leave a lot of people out. We need a global solution that will work with archaic compilers alike. C++99 would be the limit.
Jan, I'm really surprized you can't think of a solution that doesn't require the latest C++ and a bunch of libraries, I'm not even going to debate this with you, but.. give me a break!
best regards,
- dink
-
Actually, most of what you describe is already how the savestates work. And using templates to store some types in a cross-platform manner requires zero add-on libraries, just a few small templates, and the use of ptrdiff_t and inptr_t/uinptr_t for pointer arithmetic. That also assumes you (or at least the compiler) know the types involved, so dumping structs with these types in them into a single blob won't work (but templates can still help).
You can do that in plain C, but you'd need to explicitly call special functions for each affected type, which is error-prone and goes against the way fba has handled savestates since forever.
Also, just for the record, my builds have exactly 2 external dependencies that yours doesn't: libjpeg and boost. Go be a luddite and keep being stuck in the past, I really don't care. I'm just completely uninterested in joining you there. Modern C++ and Python help me do stuff that make fba better, so I use them. Go figure.
-
I'm not sure it is related to the bump to .40 release, as i mentioned in https://github.com/libretro/fbalpha/issues/166, it is more likely an issue with changing int to intptr_t in m68kcpu.c's m68k_context_size_no_pointers function in https://github.com/libretro/fbalpha/commit/0d21221fe9030ca8c482ca9042f92babf0a52dde .
From my understanding of "intptr_t", it doesn't return the same in 32/64 bits, and i kinda remember there was a warning related to this function and dink explicitly told me not to mess with it if i didn't want to break savestates, but someone else did and i overlooked this.
Edit : it would also be worth checking what was done on http://neo-source.com/index.php?topic=3162.0 , jan's special version mention being fully cross-platform for savestates. I don't think i'll have time to deal with bugs this week (wait for next week, i took leave), so please be patient, or try reverting the 2 changes i mentioned in https://github.com/libretro/fbalpha/issues/166 (the one with intptr_t and the one with ACB_FULLSCAN) and tell me if it fixes this issue.
It's not the fullscan thing (that one helped me with other issues like the coin counter not being saved properly in neogeo games) and I did that waaaay later.
I tried reverting that change and it didn't help though.
It's definitely the bump to .40 but that is one HUGE COMMIT so it will be difficult to narrow it down.
-
Ok, i think this is my fault, when i told dink about the changes for padding the savestates, it seems i overlooked some of them, so they weren't included in .40 release, here is the patch for the things i overlooked : https://github.com/libretro/fbalpha/commit/9722254c6b3789b0936e883d5b63a27a7c20b81a.patch
I also reverted intptr_t to int in m68kcpu.c .
There is just one last thing in sh2.cpp, in libretro-fbalpha .39 we did this for savestates cross-platform compatibility : https://pastebin.com/0cmkDAAi
But the current code is slightly different :
ScanVar(& ( Sh2Ext[i].sh2 ), sizeof(SH2) - 4, szText);
was replaced by
ScanVar(& ( Sh2Ext[i].sh2 ), sizeof(SH2), szText);
Sh2Ext[i].sh2.irq_callback = irq_callback;"
So i'm not sure how to apply the changes from https://pastebin.com/0cmkDAAi to the current version (honestly i don't even know why they were necessary for savestates cross-platform compatibility). Any suggestion would be welcome.
-
barbudreadmon,
sh2 was modified so that it can scan the entire struct while preserving the pointer, so the new changes are ok.
I don't understand the need to pad this in capcom/qs_c.cpp:
- INT32 nPlayStart; // Start of being played
+ ALIGN_VAR(8) INT32 nPlayStart; // Start of being played
or this in snd/ay8910.c:
- INT32 register_latch;
+ ALIGN_VAR(8) INT32 register_latch;
-
barbudreadmon,
sh2 was modified so that it can scan the entire struct while preserving the pointer, so the new changes are ok.
I don't understand the need to pad this in capcom/qs_c.cpp:
- INT32 nPlayStart; // Start of being played
+ ALIGN_VAR(8) INT32 nPlayStart; // Start of being played
or this in snd/ay8910.c:
- INT32 register_latch;
+ ALIGN_VAR(8) INT32 register_latch;
I'm not sure this is the issue, but those are the only things radius and aliaspider did in libretro-fba .39 that disappeared when i rebased the code against the fba .40 release (which seems to be the first commit where the savestates compatibility issue reappeared). I'll let you know if it fixes their issue as soon as they test it, or i'll try testing this myself (my raspberry is 32 bits and my desktop is 64 bits, i should be able to test this).
-
I can't really recommend those 2 changes, though, because an int32 doesn't need to be padded on any system.
best regards,
- dink
-
I can't really recommend those 2 changes, though, because an int32 doesn't need to be padded on any system.
best regards,
- dink
I'm not so sure, here is a quote from an article :
GNU g++ has a language extension that allows you to query a type?s alignment. The following program prints fundamental type sizes and alignment requirements of a platform for which it was compiled:
#include <iostream>
using namespace std;
template <typename T>
void print (char const* name)
{
cerr << name
<< " sizeof = " << sizeof (T)
<< " alignof = " << __alignof__ (T)
<< endl;
}
int main ()
{
print<bool> ("bool ");
print<wchar_t> ("wchar_t ");
print<short> ("short int ");
print<int> ("int ");
print<long> ("long int ");
print<long long> ("long long int ");
print<float> ("float ");
print<double> ("double ");
print<long double> ("long double ");
print<void*> ("void* ");
}
The following listing shows the result of running this program on a 32-bit x86 GNU/Linux machine. Notice the size and alignment of the long long, double, and long double types.
bool sizeof = 1 alignof = 1
wchar_t sizeof = 4 alignof = 4
short int sizeof = 2 alignof = 2
int sizeof = 4 alignof = 4
long int sizeof = 4 alignof = 4
long long int sizeof = 8 alignof = 4
float sizeof = 4 alignof = 4
double sizeof = 8 alignof = 4
long double sizeof = 12 alignof = 4
void* sizeof = 4 alignof = 4
[Actually, the above program shows that the alignment of long long and double is 8. This is, however, not the case since the IA32 ABI specifies that their alignments should be 4. Also, if you wrap long long or double in a struct and take the alignment of the resulting type, it will be 4, not 8.]
And the following listing is for 64-bit x86-64 GNU/Linux:
bool sizeof = 1 alignof = 1
wchar_t sizeof = 4 alignof = 4
short int sizeof = 2 alignof = 2
int sizeof = 4 alignof = 4
long int sizeof = 8 alignof = 8
long long int sizeof = 8 alignof = 8
float sizeof = 4 alignof = 4
double sizeof = 8 alignof = 8
long double sizeof = 16 alignof = 16
void* sizeof = 8 alignof = 8
source : https://www.codesynthesis.com/~boris/blog/2009/04/06/cxx-data-alignment-portability/
If i understand well (i'm still a noob at c/c++), the INT32 is a "long int", right ? This article is saying it has an alignment of 4 in 32 bits and 8 in 64 bits, the author could be saying bullshit though.
-
Must be it, the issue with neogeo/cps was fixed. I just need to fix cps3 now.
-
Sizes may vary (normally only long int these days I think). You should use int32_t etc. when your int needs to be a specific width. To check at runtime what you have, use std::numeric_limits.
-
If i understand well (i'm still a noob at c/c++), the INT32 is a "long int", right ? This article is saying it has an alignment of 4 in 32 bits and 8 in 64 bits, the author could be saying bullshit though.
INT32 is a standard 32bit/4 byte integer (on all platforms), this is why the alignment to 8 bytes doesn't make much sense.
best regards,
- dink
-
INT32 is a standard 32bit/4 byte integer (on all platforms), this is why the alignment to 8 bytes doesn't make much sense.
best regards,
- dink
Well, if i understand what "sizeof" do, it definitely seems INT32 is a 8 byte integer on 64 bits.
-
long is 64 bits on some systems, 32bits on others. gcc isn't lying to you. See http://en.cppreference.com/w/cpp/language/types
-
long is 64 bits on some systems, 32bits on others. gcc isn't lying to you. See http://en.cppreference.com/w/cpp/language/types
Yes, actually i also heard different compilers could produce different size/alignment on the same arch.
-
In theory, absolutely. I don't know if it's the case in practice. One thing to watch is Windows' BOOL type (which is really an int) vs C/C++ bool.
-
barbudreadmon, oh my... INT32 isn't a long! you have your define for this type set wrongly, or perhaps, need to define it differently on those platforms. it should be an unsigned int / 32bit / 4 bytes :) Obviously, the 32 after INT means something, right? :)
read here:
http://en.cppreference.com/w/cpp/language/types (thanks jan for the link)
burn.h:
typedef unsigned int UINT32;
typedef signed int INT32;
okidoki?
bye for now,
- dink
-
I found some broken dip switches:
galaga - setting it to freeplay breaks the game and it no longer loads past "ROM OK"
toaplan GP9001 hardware, toaplan-developed games (batsugun, batsugunsp, dogyuun, truxton2, possibly others, but not Raizing games on the same hardware) - region cannot be changed from the default (Korea for Batsugun, Japan for others), changes have no effect.
Setting these dips seems to work ok in MAME. :S
-
barbudreadmon, oh my... INT32 isn't a long! you have your define for this type set wrongly, or perhaps, need to define it differently on those platforms. it should be an unsigned int / 32bit / 4 bytes :) Obviously, the 32 after INT means something, right? :)
read here:
http://en.cppreference.com/w/cpp/language/types (thanks jan for the link)
burn.h:
typedef unsigned int UINT32;
typedef signed int INT32;
okidoki?
bye for now,
- dink
You are right, i assumed int32 was a long but it's a short.
Well, hopefully i'm using the vanilla defines, but this is a fact that adding the alignment macro for those 2 int32 ended up solving the cross platform savestates issue for neogeo/cps, so there definitely seems to be something different with "signed short" on 32/64 bits. Since this is really puzzling, and could be an important information for writing cross platform code in the future, i'll write a little program to test size/alignment of those on both 32 and 64 bits system.
Edit : i confirm size/alignment of INT32 don't differ between 32 and 64 bits system (at least on gcc linux), i overlooked this one but there is also an ALIGN_VAR(8) for the void in ay8910.h, so it is probably this one which solved the issue for those drivers.
-
i assumed int32 was a long but it's a short.
I hope not -- that would make it 16 bits. Again, check using std::numeric_limits if you're not sure.
-
I hope not -- that would make it 16 bits. Again, check using std::numeric_limits if you're not sure.
My bad, i mean an int.
-
The cross-platform issue with sh2 savestates is that opbase in sh2ext is 4 byte on 32 bits and 8 byte on 64 bits. In libretro-fba 39 someone commented the
SCAN_VAR (Sh2Ext[i].opbase)
in sh2.cpp. It seems commenting it again solve the cross-platform issue, but i suppose it is there for a reason.
After reading jan's posts, i tried replacing the type by uint8_t, the size would still differ.
So i tried replacing the declaration/code by this :
diff --git a/src/cpu/sh2/sh2.cpp b/src/cpu/sh2/sh2.cpp
index a257404..1649fa2 100644
--- a/src/cpu/sh2/sh2.cpp
+++ b/src/cpu/sh2/sh2.cpp
@@ -179,7 +179,7 @@ typedef struct
pSh2ReadLongHandler ReadLong[SH2_MAXHANDLER];
pSh2WriteLongHandler WriteLong[SH2_MAXHANDLER];
- unsigned char * opbase;
+ ALIGN_VAR(8) unsigned char * opbase;
int suspend;
} SH2EXT;
@@ -3508,7 +3508,7 @@ int Sh2Scan(int nAction)
Sh2Ext[i].sh2.irq_callback = irq_callback;
SCAN_VAR (Sh2Ext[i].suspend);
- SCAN_VAR (Sh2Ext[i].opbase);
+ ScanVar(& (Sh2Ext[i].opbase), 8, "Sh2Ext[i].opbase" );
#if FAST_OP_FETCH
// Sh2Ext[i].opbase
It seems to work but i've got a feeling my code is horrible. Any suggestion ?
-
That's because Sh2Ext.opbase's type is not uint8_t, but uint8_t*.
Pointers have no business being in any savestate. In fact at first glance doesn't look like opbase needs to be included at all.
More generally, in cases where you do need to include pointers, which will be rare, you will need to find a way to convert them to something that's not a pointer, even if that's by using pointer arithmetic and forced alignment.
-
Just remove/comment this line: SCAN_VAR (Sh2Ext.opbase);
it's recalculated on state load, and doesn't need to be scanned in the first place.
best regards,
- dink
-
In order to detect that sort of problem, here's a bit of code that replaces the SCAN_VAR macro; insert it into state.h replacing the #define SCAN_VAR line:
#if !defined(__cplusplus) || ((_MSC_VER < 1900) && (__cplusplus <= 199711L))
#define SCAN_VAR(x) ScanVar(&x, sizeof(x), #x)
#else
class scan_exception
{
const char* const _what_arg;
public:
const char* const variable;
scan_exception(const char* const what_arg, const char* const variable) : _what_arg(what_arg)
, variable(variable)
{
return;
}
const char* const what()
{
return _what_arg;
}
};
#define SCAN_VAR(x) { if (std::is_pointer<decltype(x)>::value) throw scan_exception("variable is a pointer!", #x); else ScanVar(&x, sizeof(x), #x); }
#endif
Also insert #include <type_traits> in burnint.h:
#if defined(__cplusplus) && !((_MSC_VER < 1900) && (__cplusplus <= 199711L))
#define _USE_MATH_DEFINES
#include <type_traits>
#endif
If you use this, trying to include a pointer in a savestate will cause an exception to be thrown. Note that it's only compiled if your compiler has C++11 support enabled, does not work for code that's compiled as plain C, and your compiler may complain about exceptions being thrown where it doesn't expect them.
Other than that, see screenshot.
-
I make a nicer version. The new one uses static asserts and sfinae to detect problems at compile time and do the right thing in 32/64 bit versions. I also built in support for pointers into some part of the game's memory or an array (only neo geo uses those atm).
Direct support for ptrdiff_t and (u)intptr_t to come.
Compiling threw up errors in two drivers that dump pointers into savestates:
d_powerins.cpp:
SCAN_VAR(RamCurPal);
d_galpanic.cpp:
SCAN_VAR(RamCurPal);
SCAN_VAR(RamCTB64k);
They should not use SCAN_VAR, since they need to save the contents of RAM and not a pointer.
-
d_powerins.cpp:
SCAN_VAR(RamCurPal);
d_galpanic.cpp:
SCAN_VAR(RamCurPal);
SCAN_VAR(RamCTB64k);
Thanks for the heads-up.
best regards,
- dink
-
Final update on area/variable scanning:
- included a macro and c++ functions for scanning memory, much like the one for variables
- better const-correctness, full const-correctness will require changes to all drivers, but as a start...
- burnArea struct has constructors easily allowing use of const char* const etc, wich requires...
- all of burn is compiled as c++, which allows...
- no more c linkage except for the burn <-> burner interface and external libraries
- both c++97 and c++11 versions of scanning functions
- c++11: separate header file
- c++11: type-safe support for pointers pointing inside a driver's data (e.g. memory banking, used by neo geo)
- c++11: compile time test for incompatible types (except in structs which is impossible to detect in the current standard)
- c++11: 64bit builds have fixups for some types (pointers, ptrdiff_t, size_t, (u)intptr_t)
- c++11: 64bit builds have runtime checks that throw exceptions (indicating driver bugs)
Tested and working with both visual studio and gcc (except 64bit, so until that code is tested it fails with a static assert).
-
Nice, use this to find more bugs for us, and leave our burn / drivers alone! edit: please.
-
Thanks, Dink. Christmas spirit in action, I guess.
Is this how you treat everyone who contributes, or am I I just special?
-
Thanks, Dink. Christmas spirit in action, I guess.
Is this how you treat everyone who contributes, or am I I just special?
Just curious Jan, you got your own version of FBA or it's another emulator based on FBA?
-
Just curious Jan, you got your own version of FBA or it's another emulator based on FBA?
It's FBA, though it has a lot of new things for the Windows UI and is slowly diverging otherwise. It's also a work in progress. That wasn't my choice, but as you can see this situation is probably going to remain the status quo.
-
It's FBA, though it has a lot of new things for the Windows UI and is slowly diverging otherwise. It's also a work in progress. That wasn't my choice, but as you can see this situation is probably going to remain the status quo.
try to make a jan's WIP topic, it will be easier to follow :)
-
Yes, It would be nice to stay on topic --> "FB Alpha 0.2.97.42 Bug Reports"
best regards,
- dink
-
I'm not sure it should be called a bug, but there doesn't seem to be multitap support in fbalpha megadrive.
Also, perhaps it is worth mentioning there were several models of multitap, with different compatibilities (see Team Player (https://segaretro.org/Team_Player) and 4 Way Play (https://segaretro.org/4_Way_Play))
-
It's called an unimplemented feature. I have a great idea, check your pm's regarding this.
best regards,
- dink
-
Hi everyone,
It seems Tournament Arkanoid is stuck with a "bad hardware" screen.
-
Thanks for reminding me, all of the sets in arkanoid are kinda screwed up. We need to update our mcu interface or the port handlers a little bit, it seems. Hopefully another dev will help because I failed at fixing this.
-
hello all my respects to the dev teams
i am just making certain that this bug isnt a personal one..
it is about the games alpine racer 2& armadillo racing.
would be nice to have in fbalpha. the thing about it, it
seems as it is marked green long time ago by mame.
but the sound is very buggy. so my hopes of playing
these two isnt coming it seems. the gfx scroll is
perfect it seems but the music is garbled by bad routines
in the sample playback, which stops right in the middle
of a riff/section & plays it over again many times so it is
sort of a stuttering type only it stutters the music. so
we are talking about alot here & it comes over & over
again & is indeed a lot of headache. i have to stop after
a while with sound on to take a pill & eventually turn off
the sound which is a loss, cause it gets boring. whenever
you fba devs get into this driver remember this. i also
noticed that the audio driver in windows can get dirty &
cause even more noise/slowness clicks etc. due to unknown
reasons, probably a hack though. i use fixwmp utility for
this if anyone messes with it. released by
so best of luck with the 0299 release or maybe it is a 03
release. i have been waiting for it for a long time, but i aint
complaining :)
-
Hi h-enigm,
the games you mentioned are based on NAMCO System 22 hardware (the same system of Ridge Racer) and I doubt that the DEVs would add the related emulation on FBA... but I could be wrong! ;)
-
hello all my respects to the dev teams
i am just making certain that this bug isnt a personal one..
it is about the games alpine racer 2& armadillo racing.
would be nice to have in fbalpha. the thing about it, it
seems as it is marked green long time ago by mame.
but the sound is very buggy. so my hopes of playing
these two isnt coming it seems. the gfx scroll is
perfect it seems but the music is garbled by bad routines
in the sample playback, which stops right in the middle
of a riff/section & plays it over again many times so it is
sort of a stuttering type only it stutters the music. so
we are talking about alot here & it comes over & over
again & is indeed a lot of headache. i have to stop after
a while with sound on to take a pill & eventually turn off
the sound which is a loss, cause it gets boring. whenever
you fba devs get into this driver remember this. i also
noticed that the audio driver in windows can get dirty &
cause even more noise/slowness clicks etc. due to unknown
reasons, probably a hack though. i use fixwmp utility for
this if anyone messes with it. released by
so best of luck with the 0299 release or maybe it is a 03
release. i have been waiting for it for a long time, but i aint
complaining :)
Definitely not the sort of game well-suited for being in FBA. If you look in the games list there are precisely ZERO 3D [polygon-based] games in FBA.
-
when you check autoframeskip in the program , in the fba.ini it change "bAlwaysDrawFrames" to 0, but the explanation in the ini says that "//If non-zero, skip frames when needed to keep the emulation running at full speed"
:confused: :confused: :confused: :confused: :confused: :confused: :confused:
-
when you check autoframeskip in the program , in the fba.ini it change "bAlwaysDrawFrames" to 0, but the explanation in the ini says that "//If non-zero, skip frames when needed to keep the emulation running at full speed"
:confused: :confused: :confused: :confused: :confused: :confused: :confused:
I always thought that was funny, thanks reminding me to fix it :)
It will read (next version): // If zero, skip frames when needed to keep the emulation running at full speed
-
after several days of insanity, arkanoid (and all games it's driver) are finally working with the real/non-bootleg prot mcu's. yay!
best regards,
- dink
-
hello all my respects to the dev teams
i am just making certain that this bug isnt a personal one..
it is about the games alpine racer 2& armadillo racing.
would be nice to have in fbalpha. the thing about it, it
seems as it is marked green long time ago by mame.
but the sound is very buggy. so my hopes of playing
these two isnt coming it seems. the gfx scroll is
perfect it seems but the music is garbled by bad routines
in the sample playback, which stops right in the middle
of a riff/section & plays it over again many times so it is
sort of a stuttering type only it stutters the music. so
we are talking about alot here & it comes over & over
again & is indeed a lot of headache. i have to stop after
a while with sound on to take a pill & eventually turn off
the sound which is a loss, cause it gets boring. whenever
you fba devs get into this driver remember this. i also
noticed that the audio driver in windows can get dirty &
cause even more noise/slowness clicks etc. due to unknown
reasons, probably a hack though. i use fixwmp utility for
this if anyone messes with it. released by
so best of luck with the 0299 release or maybe it is a 03
release. i have been waiting for it for a long time, but i aint
complaining :)
Well, i have been dreaming of seeing some stv games in fba myself (mainly my favorite shmup Radiant Silvergun, which was crashing around stage 10 last time i tried it in mame), but that won't happen.
after several days of insanity, arkanoid (and all games it's driver) are finally working with the real/non-bootleg prot mcu's. yay!
best regards,
- dink
Nice work !
-
@dink nice job ya did there!!
-
Great release as always!
I tested a bit the libretro .43 core and found a small issue. Cyvern (Kaneko system) has very choppy video with "Speed Hack:On". The framerate number shows no fluctuations and the sound has no issues, but the image is slightly stuttery every two seconds or so. Before, with the hack, it had perfectly smooth image with some missing and delayed sounds and music.
Also, the buttons seem to be inverted/differently placed (what used to be fire is now nothing, what was nothing is now bomb and what was bomb is now fire....if that makes any sense), with no option in the ""Core options" to set them to either Gamepad or Arcade, like before.
-
Cyvern is one of the games that uses nearly all of its cycles doing something (unlike other skns games), therefore a speedhack really isn't recommended on this game.
Although, if it's a case of using a speedhack or not playing the game at all, maybe barbudreadmon (libretro guy) can change it back to use the old speedhack mode (there's a few of them for sh-2) just for cyvern?
best regards,
- dink
-
Great release as always!
I tested a bit the libretro .43 core and found a small issue. Cyvern (Kaneko system) has very choppy video with "Speed Hack:On". The framerate number shows no fluctuations and the sound has no issues, but the image is slightly stuttery every two seconds or so. Before, with the hack, it had perfectly smooth image with some missing and delayed sounds and music.
Also, the buttons seem to be inverted/differently placed (what used to be fire is now nothing, what was nothing is now bomb and what was bomb is now fire....if that makes any sense), with no option in the ""Core options" to set them to either Gamepad or Arcade, like before.
I guess i could add a third option in the speedhack dipswitch to enable the old "cps3speedhack", saying "some missing and delayed sounds and music" is a big understatement though : the sound was a complete mess.
About the lr-fbalpha mapping, it was migrated to "Quick Menu > Controls" a year ago. There is no equivalent to the old "Arcade" setting though, because it relied on not having your retropad mapped as documented there (https://docs.libretro.com/specs/retropad/), which is bad thing.
-
Ah i get it.
Really, don't bother adding the old hack. I mean, one could use the old core anyway for a different experience.
I guess i'll modify the touch controls to keep the button placements 😊
-
More bug reports, please!!
Remember, we rely on you: friends and fans of FBAlpha, to report any bugs or glitches in games that you like to play.
best regards,
- dink
-
More bug reports, please!!
I saw a report this morning about green beret & rush'n attack running extremely fast since last release, considering they are both from the same driver there could be some truth behind this issue, i didn't have time to test it before going to work though.
-
I saw a report this morning about green beret & rush'n attack running extremely fast since last release, considering they are both from the same driver there could be some truth behind this issue, i didn't have time to test it before going to work though.
gberet is OK here, I did the side-by-side comparison test :)
best regards,
- dink
-
gberet is OK here, I did the side-by-side comparison test :)
best regards,
- dink
gberet needs a re-write in a bad way. One of these days i'll get around to finishing that.
-
I saw a report this morning about green beret & rush'n attack running extremely fast since last release, considering they are both from the same driver there could be some truth behind this issue, i didn't have time to test it before going to work though.
Since there were no changes to the driver, I have a theory:
first: this was probably always broke for the reporter, unless there was a core change to the fps handling in retroarch/libretro etc.
second: gberet runs at 30fps. If the system they are running on is 60fps only, or fps handling is broken somehow, the game will run twice the speed.
best regards,
- dink
-
Since there were no changes to the driver, I have a theory:
first: this was probably always broke for the reporter, unless there was a core change to the fps handling in retroarch/libretro etc.
second: gberet runs at 30fps. If the system they are running on is 60fps only, or fps handling is broken somehow, the game will run twice the speed.
best regards,
- dink
Yeah i had the same exact problem on the xbox with Green Beret / Rush'n Attack another also affected was Ironhorse, so it's
definately a platform specific issue, i have a workaround however to bring the speed down to the correct level for the above
games via iq_132........
@barbudreadmon i'll push a commit later which should sort this.
-
Not actually a bug (and not worth its own thread, I guess), but something that everyone probably already knows about: the game filter categories are kind of arbitrary (which companies are listed or not, and some overlap, too). Also, games are not always listed under the most appropriate filter (for example, many are lumped together in Misc pre 90s rather than being put under the concerned company).
Are there plans to overhaul this, maybe using game metadata instead?
-
Stifu, thanks for the feedback.
This is a tough one, and there is definitely room for improvement. There are no plans really, but if we notice something out of place, we'll try to fix it.
best regards,
- dink
-
Not actually a bug (and not worth its own thread, I guess), but something that everyone probably already knows about: the game filter categories are kind of arbitrary (which companies are listed or not, and some overlap, too). Also, games are not always listed under the most appropriate filter (for example, many are lumped together in Misc pre 90s rather than being put under the concerned company).
Are there plans to overhaul this, maybe using game metadata instead?
Completely arbitrary. I would not be against moving everything into simply drv and anything that doesn't have game structs into some other directory. The down side is that we'd lose all the history on those files.
-
iq_132 oh please no.. that would be horrible.
-
Thanks for your replies.
I noticed something that seems weird to me. I'm under the impression "gprider" and "gpriders" (clone of gprider) have the exact same 30 files, with the same crc. I noticed that because Romcenter tells me I have "gpriders" even though I only have "gprider".
-
Thanks for your replies.
I noticed something that seems weird to me. I'm under the impression "gprider" and "gpriders" (clone of gprider) have the exact same 30 files, with the same crc. I noticed that because Romcenter tells me I have "gpriders" even though I only have "gprider".
The parent is the dual board, and the clone the single board. The roms match MAME, however, we don't emulate the dual board so they are effectively the same set.
-
I must be cursed or something with arkanoids, i picked a random version (arkgcbl), and ended up with the same "bad hardware" screen i had with arkatour on .42, except i get it later (after intro).
On a sidenote, the gberet issue was fixed (basically, the timings i computed for the libretro api were wrong because nBurnSoundRate/1000 was greater than nBurnFPS/100).
-
Try using a real version of Arkanoid (arkanoid, arkanoidj, arkanoidu, arkatour), those are the only ones I worked on. I'll check out the bootlegs though. the bootlegs with mcu should be ok.
edit: later that day... fixed the issue with no-mcu bootlegs.
Glad ya worked out the gberet issue :)
best regards,
- dink
I must be cursed or something with arkanoids, i picked a random version (arkgcbl), and ended up with the same "bad hardware" screen i had with arkatour on .42, except i get it later (after intro).
On a sidenote, the gberet issue was fixed (basically, the timings i computed for the libretro api were wrong because nBurnSoundRate/1000 was greater than nBurnFPS/100).
-
Hi !
First of all, thank you very much for this fantastic emulator!
I think I found a bug.
When I tried to load a particular game, "super speed race Junior (Japan)", FBA crashes. Rom name is ssrj.zip
It happends as soon as I click on the game's name. Before even trying to double click to start it.
Im using windows 64bits.
Hope it helps.
Cheers,
TMZ
-
Hi tmznix,
Super Speed Race Junior is broken for now, it will be fixed in the next version of FBA. Sorry about that!
best regards,
- dink
-
Hi tmznix,
this is a known issue, it afflicts the binary versions (32 and 64 bit), but not the XP binary version... as dink wrote, it will be fixed in the next FBA release! :)
-
Hi ^^
In first, i want to thank you for all your work :wink:
I don't know if those "bugs" are know issue, lot of time i do post it.
They're on Cps_1.
Captain commando : On the title screen appears, look on I of CAPTAIN.
Chiki Chiki boys : The screen titles appears and disappears the time of one frame (i think).
Similar for the screens of the demo.
Not a bug, just an information ^^
For the B_18 PPU ID, i have see it's not matching mame on ctrl, pal ctrl and layers enable. (i have game converted on B_18 for my personnal use.).
I have looked on it because all my conversion are black screen :)
After created another B_18 (matching mame), my conversions works well, but i have break sf2ue and sfee (only this 2 games work on b_18)
Need more look for me on the code for understand ;) :)
Best Regards from france :)
And sorry for my limited english :redface:
-
I can't see any issue?
-
Hi Barry :)
Look on the picture, you can see just up to AI of captain the problem ;)
When the title appear.
-
Bonky0013, thanks. chiki video glitches are fixed :)
re: captcomm
I can't reproduce this problem. is the romset OK? how about CPU speed - is it at 100%? (Game -> Reset CPU speed to 100%)
will you post your config/games/captcomm.ini file?
EDIT I found the problem!
best regards,
- dink
-
Bonky0013, problem with captcomm also fixed :) thanks!
note to others; the problem happens while the Captain Commando text is being displayed with a raster effect.
best regards,
- dink
-
Bonky0013, problem with captcomm also fixed :) thanks!
note to others; the problem happens while the Captain Commando text is being displayed with a raster effect.
best regards,
- dink
Did these bugs happen in older versions as well? The fixes are hacky at best and indicate a problem with our timing which probably needs further investigation. Especially if they worked fine in previous versions.
-
Treble Winner,
I've been wanting to fix the chiki one for ages, but forgot about it actually. I verified that they both happen in past versions. chiki since I've been with the project. Captcomm always did the raster a bit wrong, but didn't write the weird character above until around the time of the new m68k core. You probably didn't notice the problems because they only happen for a frame or 2 during raster effects (scroll register hackery) by the game.
chiki/captcomm actually run at 262 lines (which can be verified in the emulator encyclopedia aka MAME), having it at 259 previously was basically running vbl at the wrong line. secondly, drawing after vblank is bad for any raster effect.
Explanation: after vbl and until line 16, the game prepares some the next scene / scroll registers / etc. The reason for the glitches is simple, the game was drawing during the preparation of the next scene.
best regards,
- dink
-
Hi ^^
Fixed so fast ^^
Thanks :biggrin:
These two problems was very old ;)
-
Treble Winner,
I've been wanting to fix the chiki one for ages, but forgot about it actually. I verified that they both happen in past versions. chiki since I've been with the project. Captcomm always did the raster a bit wrong, but didn't write the weird character above until around the time of the new m68k core. You probably didn't notice the problems because they only happen for a frame or 2 during raster effects (scroll register hackery) by the game.
chiki/captcomm actually run at 262 lines (which can be verified in the emulator encyclopedia aka MAME), having it at 259 previously was basically running vbl at the wrong line. secondly, drawing after vblank is bad for any raster effect.
Explanation: after vbl and until line 16, the game prepares some the next scene / scroll registers / etc. The reason for the glitches is simple, the game was drawing during the preparation of the next scene.
best regards,
- dink
Can confirm:
MCFG_SCREEN_RAW_PARAMS(XTAL(16000000)/2, 512, 64, 448, 262, 16, 240)
That should also put our fps to 59.637404580152671755725190839695 hz ^^
(1600000/2/512/262)
-
Megadrive Streets of Rage level 3.
Corruoption of tiles at the left hand side of screen.
-
Megadrive Streets of Rage level 3.
Corruoption of tiles at the left hand side of screen.
Fixed, thanks for the report & state! :)
best regards,
- dink
-
I'm confused about a little something.
Somewhere between FBA 0.2.97.40 and .42, support for Samurai Shodown II and IV "Special" hacks has been added. (If memory serves, it was actually forgotten in .40 despite being mentioned in the release notes, but whatever). Now, looking at .43, I can't see these hacks anymore among the samsho2 or samsho4 clones.
Were they removed on purpose or not? What's the story? Thanks.
-
We had to remove all hacks by GSC2007 because this guy is absolutely mental, and only wants his hacks to run on his "special" version of mame - which has no source code, only exe's. ... which in itself is very suspicious. Anyways, we had to remove them to avoid an influx of insults and accusations from this guy. Sorry about that.
best regards,
- dink
-
I'm confused about a little something.
Somewhere between FBA 0.2.97.40 and .42, support for Samurai Shodown II and IV "Special" hacks has been added. (If memory serves, it was actually forgotten in .40 despite being mentioned in the release notes, but whatever). Now, looking at .43, I can't see these hacks anymore among the samsho2 or samsho4 clones.
Were they removed on purpose or not? What's the story? Thanks.
If you find interesting hacks in hbmame (obviously not the ones by GSC2007), i can look into porting them to fba as i did for kof2k2omg :).
-
Ah, I remember reading something about that. I didn't realize those hacks were part of it (I don't even know what they add to the original games, I just wanted to check them out). This whole thing is ridiculous. The only reason this guy could create his modified emulator and ROMs is thanks to the efforts of all the people who freely and nearly unconditionally shared their work with everyone. He wants to benefit from it while imposing his petty conditions (does that even respect MAME's licensing conditions?). It's selfish, immature and shortsighted. He's wilfully limiting the reach of his work, and ensuring his pet projects won't outlive him.
Anyway, sorry for the rant. Thanks for the offer to add cool hacks, barbudreadmon. If I ever stumble on any, I'll let you know.
-
Ah, I remember reading something about that. I didn't realize those hacks were part of it (I don't even know what they add to the original games, I just wanted to check them out). This whole thing is ridiculous. The only reason this guy could create his modified emulator and ROMs is thanks to the efforts of all the people who freely and nearly unconditionally shared their work with everyone. He wants to benefit from it while imposing his petty conditions (does that even respect MAME's licensing conditions?). It's selfish, immature and shortsighted. He's wilfully limiting the reach of his work, and ensuring his pet projects won't outlive him.
Anyway, sorry for the rant. Thanks for the offer to add cool hacks, barbudreadmon. If I ever stumble on any, I'll let you know.
I still think we should put them back. His "work" is based on SNK's IP and we don't remove their stuff.
-
They're back :)
-
They're back :)
My life is truely now complete :D
-
Thanks for that, dink. I hope they're worth it! I still haven't had a chance to try them.
Anyway... I found something that looks like a bug, but I don't know whether it's a game bug or emulation bug.
It's in the intro of Crude Buster, in the scene where you see the big punk shake hand with the government official. It looks buggy (wrong colors, apparent tiles) at first, then it looks fine after a second. The bug can be seen here: https://youtu.be/8-49fSaAnQc?t=40s
-
Stifu,
Regarding Crude Buster:
The pcb does the exact same thing - I wonder if they were going for a lights off / lights on effect?
best regards,
- dink
-
I see, thanks for clearing that up. I don't think it's intentional. Weird colors aside, the apparent tile squares definitely look buggy. Also note the Mega Drive / Genesis version doesn't have those issues.
-
You're welcome. Yes, it does look odd w/ the missing tile(s) and weird colors. Maybe the original testers of the game didn't notice it because they didn't have the modern features we have (pause/ffwd etc). Who knows :) I guess we'll never know without asking the original designers of the game.
best regards,
- dink
I see, thanks for clearing that up. I don't think it's intentional. Weird colors aside, the apparent tile squares definitely look buggy. Also note the Mega Drive / Genesis version doesn't have those issues.
-
About Crude Buster again, I assume the buggy tiles around the second punk in the ending scene are also a game bug rather than an emulation bug?
https://youtu.be/E6sY_sMl2aQ?t=32m11s
This bug is even in "Data East Arcade Classics" for the the Wii, but sometimes such compilations have emulation bugs, so better safe than sorry...
Also noticed I have bleeding colors on the sides of the screen when I'm in full screen mode with FBA (see screenshot). When sprites are located on the rightmost part of the screen, I get bleeding pixels on the left side, and vice versa. I guess that's just some imperfect video setting or something, rather than an actual FBA bug.
I can provide save states for both of these things, if needed.
-
Stifu, thanks, please post the states :)
-
Here you go. Thanks for having a look.
- cbuster_bullboss: color bleeding in full screen mode. It actually happens all the time, not just here, but that particular scene makes the problem more obvious due to the colors used. Go in full screen mode, and move your character either to the far left or far right side.
- cbuster_ending: second punk buggy tile issue (took the save just a few seconds before the bug occurs).
-
I can't reproduce either issue - please also post your ini file so I can investigate. I think you might be using a scaler or fx that has the bug in it (a couple are buggy and actually cause the bleeding issue which depends on the game's screen size).
Also try with all effects disabled and let me know how it looks?
oops, I mean with the ending scene, I don't see what the bad tiles are in the video or in the state. Are you talking about the money that is sorta "floating" out of his bag? :)
best regards,
- dink
-
Here's the ini. As I expected, the full-screen color bleeding goes away if I disable bi-linear filtering.
About the buggy tiles... Never mind, I'm stupid. They're just badly-drawn dollar bills. :\
-
Thanks, I'll try to figure out whats going wrong with bilinear filtering.
It's alright!
Here's the ini. As I expected, the full-screen color bleeding goes away if I disable bi-linear filtering.
About the buggy tiles... Never mind, I'm stupid. They're just badly-drawn dollar bills. :\
-
Hi,
it seems there is a color issue in dkong, here is the original pcb :
(https://retropie.org.uk/forum/assets/uploads/files/1528726484805-img_8906-resized.jpg)
current mame is close to those colors, i guess it is related to this change :
0.141u2: Couriersud fixed a bug in palette_normalize_range which caused color distortion. The luminance normalization now converts r,g,b to y,u,v and normalizes y prior to converting back to r,g,b. This affects e.g. Radar Scope, Donkey Kong and Mario Bros.
fba is a lot more redish.
I'm not 100% sure but i guess it is related to one (or both) of those changes in mame :
diff -Naur mame-mame0141u1/src/emu/video/resnet.c mame-mame0141u2/src/emu/video/resnet.c
--- mame-mame0141u1/src/emu/video/resnet.c 2011-01-24 06:09:05.000000000 +0100
+++ mame-mame0141u2/src/emu/video/resnet.c 2011-02-09 16:01:01.000000000 +0100
@@ -564,7 +564,7 @@
cut = 0.0;
break;
case RES_NET_AMP_DARLINGTON:
- minout = 0.9;
+ minout = 0.7;
cut = 0.0;
break;
case RES_NET_AMP_EMITTER:
@@ -685,13 +685,15 @@
v = vcc - v;
v = MAX(0, v-0.7);
v = MIN(v, vcc - 2 * 0.7);
+ v = v / (vcc-1.4);
+ v = v * vcc;
break;
case RES_NET_MONITOR_ELECTROHOME_G07:
/* Nothing */
break;
}
- return (int) (v *255 / vcc + 0.4);
+ return (int) (v * 255 / vcc + 0.4);
}
rgb_t *compute_res_net_all(running_machine *machine, const UINT8 *prom, const res_net_decode_info *rdi, const res_net_info *di)
diff -Naur mame-mame0141u1/src/lib/util/palette.c mame-mame0141u2/src/lib/util/palette.c
--- mame-mame0141u1/src/lib/util/palette.c 2011-01-24 06:09:05.000000000 +0100
+++ mame-mame0141u2/src/lib/util/palette.c 2011-02-09 16:01:01.000000000 +0100
@@ -682,8 +682,8 @@
void palette_normalize_range(palette_t *palette, UINT32 start, UINT32 end, int lum_min, int lum_max)
{
- UINT32 ymin = 1000 * 255, ymax = 0;
- UINT32 tmin, tmax;
+ INT32 ymin = 1000 * 255, ymax = 0;
+ INT32 tmin, tmax;
UINT32 index;
/* clamp within range */
@@ -707,11 +707,13 @@
for (index = start; index <= end; index++)
{
rgb_t rgb = palette->entry_color[index];
- UINT32 y = 299 * RGB_RED(rgb) + 587 * RGB_GREEN(rgb) + 114 * RGB_BLUE(rgb);
- UINT32 target = tmin + ((y - ymin) * (tmax - tmin + 1)) / (ymax - ymin);
- UINT8 r = (y == 0) ? 0 : rgb_clamp(RGB_RED(rgb) * 1000 * target / y);
- UINT8 g = (y == 0) ? 0 : rgb_clamp(RGB_GREEN(rgb) * 1000 * target / y);
- UINT8 b = (y == 0) ? 0 : rgb_clamp(RGB_BLUE(rgb) * 1000 * target / y);
+ INT32 y = 299 * RGB_RED(rgb) + 587 * RGB_GREEN(rgb) + 114 * RGB_BLUE(rgb);
+ INT32 u = ((INT32)RGB_BLUE(rgb)-y /1000)*492 / 1000;
+ INT32 v = ((INT32)RGB_RED(rgb)-y / 1000)*877 / 1000;
+ INT32 target = tmin + ((y - ymin) * (tmax - tmin + 1)) / (ymax - ymin);
+ UINT8 r = rgb_clamp(target + 1140 * v / 1000);
+ UINT8 g = rgb_clamp(target - 395 * u / 1000 - 581 * v / 1000);
+ UINT8 b = rgb_clamp(target + 2032 * u / 1000);
palette_entry_set_color(palette, index, MAKE_RGB(r, g, b));
}
}
Do you think this change is somehow appliable to fba ?
Best regards :).
-
It seems there's an input issue in Sangokushi III Gaiden: Kakou-On's Revenge DX (a hack of wofch, and its ROM name is wofchdx). This game is unplayable since START button does not work. Is it a known issue?
-
Hi,
wofch and wofchdx will play fine: but its a weird game, you have to wait until the game starts demo play before you can start. I think because its a CPS Changer game, so its on FreePlay mode.
best regards,
- dink
-
Hi,
I found 3 new bugs (tested all of them in fba standalone) :
1. P2 can't join in monster maulers and its clone
2. Most games in Taito "The NewZealand Story" driver are broken (fba crash). Actually i noticed something really weird : "kabukiz" is not present in the fba arcade dat file while the game is present and marked as working in the driver.
3. In megadrive's general chaos, it seems multitap is detected, but either i don't understand how to move P2/P3/P4, or their controls are broken
-
Hi,
I found 3 new bugs (tested all of them in fba standalone) :
1. P2 can't join in monster maulers and its clone
2. Most games in Taito "The NewZealand Story" driver are broken (fba crash). Actually i noticed something really weird : "kabukiz" is not present in the fba arcade dat file while the game is present and marked as working in the driver.
3. In megadrive's general chaos, it seems multitap is detected, but either i don't understand how to move P2/P3/P4, or their controls are broken
Thanks for the reports! I fixed up tnzs a bit, but need to do some quick testing tonight when I get home.
General Chaos, select 4 players from the title screen, then theres a tricky part about squad selection, I forget how to get past that other than random button presses :) But once the game finally starts I can control all 4 players.
best regards,
- dink
-
Actually i noticed something really weird : "kabukiz" is not present in the fba arcade dat file while the game is present and marked as working in the driver.
kabukiz is only in debug builds (both for the game list and the dat).
-
kabukiz is only in debug builds (both for the game list and the dat).
Thanks for the information, i didn't think about checking gamelist.txt, is there any way to know this when looking at
struct BurnDriverD BurnDrvKabukiz = {
"kabukiz", NULL, NULL, NULL, "1988",
"Kabuki-Z (World)\0", NULL, "Taito Corporation Japan", "Miscellaneous",
NULL, NULL, NULL, NULL,
BDF_GAME_WORKING | BDF_HISCORE_SUPPORTED, 2, HARDWARE_TAITO_MISC, GBF_SCRFIGHT, 0,
NULL, kabukizRomInfo, kabukizRomName, NULL, NULL, CommonInputInfo, KabukizDIPInfo,
Type2Init, DrvExit, DrvFrame, DrvDraw, DrvScan, &DrvRecalc, 0x200,
256, 224, 4, 3
};
?
-
BurnDriver - Normal
BurnDriverD - Debug only
BurnDriverX - Don't show ever
-
Wow, I'm a dummy! I fixed / made Kabuki-Z playable waaay back in .37. And it's been BurndriverD'd this whole time. DERP!
I remember being super happy about finally getting that game working back then, too.. hehe.. oh well - sorry guys!
best regards,
- dink
-
We should run a burndriver check before releases. This isn't the first time that this has happened
-
We should run a burndriver check before releases. This isn't the first time that this has happened
As long as I get credit for the last time it happened! :P
-
Regarding VS Block Breaker issue with the hamster-paddle tile covering up part of the titlescreen:
This was finally fixed tonight. It was quite perplexing, and I'm happy to finally fix it. On a side note,
another bug got fixed that "covered up" the flashing red or yellow that happens when the hamster
gets struck with the paddle in the intro. yay!
best regards,
- dink
This one almost drove me crazy as i wasn't able to reproduce it and was starting to wonder if i dreamed it, but i was finally able to reproduce it.
-
Hi,
Contra color issues
-
Hello, first of all thanks for your work; has found a couple of errors: in the Wizard Fire (Over Sea v2.1) in the second and last level (attached screenshot) the clouds prevent to see the stage (In MAME there are transparencies)
In Shadow Dancer (set 3, US) in phase 2-4 the Final Burn Alpha has an error and closes
I copy what the log tells me
Thanks for everything. A greeting!
Exception 0xC0000005 (EXCEPTION_ACCESS_VIOLATION) thrown.
EIP: 0x00C8
(attempting to write address 0x77CCF456)
EAX: 0x00000000, EBX: 0x00000000, ECX: 0x00000000, EDX: 0x77CCF456
ESI: 0xFFFF8000, EDI: 0x004EC8F0, ESP: 0x001BDA30, EBP: 0x7A120000
-
Neildark, hi, thanks for the bug reports!
Do you have savestates for Wizard Fire cloud level and Shadow Dancer from just before the crash? :)
best regards,
- dink
-
Hi, thanks for the report, Contra is fixed
best regards,
- dink
Hi,
Contra color issues
-
Hi guys,
I made 3 savestates related to the translucent issues in Wizard Fire! ;)
1) In the second stage the fog should be translucent, however it's totally opaque.
2) In the arena with the golems (at the end of stage 4) the energy column should be traslucent, however it's totally opaque.
3) In the fifth stage the clouds should be translucent, however they are totally opaque.
-
I also made a savestate for Shadow Dancer (at the end of level 2-3)...
after few seconds that level 2-4 is started the emulator crashes (Guru Meditation).
-
Hello again guys, I have also detected a bug in G.I. Joe (World, EAB, set 1) in the Cavern Base area (Final Second Stage), in MAME water effects are seen in the whole final part, in Final Burn alpha it does not exist. I attached images in MAME and FBA. Greetings, and thanks for your work.
-
Hi Neildark, thanks for the bug report! :smilie:
To help the check you should make a savestate (just before the issue happened) and attach it in a post... :smilie:
"Game > Save states... > Save state..." or "Ctrl + F10"
-
Yes of course. Here I leave the savestate, it is just before entering the area of the enemy, which, in theory, is surrounded by water. A greeting.
-
Probably Konami's G.I. Joe has various issues with missing ROZ layers... another example is the boss at the end of Mission 1 (the floor is totally absent/invisible).
-
afaik GI Joe has been broken in MAME for many years, with line effects not being implemented because the way they were handled before was one of the grossest hacks ever.
so comparisons with MAME aren't really going to be valid here.
-
Shadow Dancer crash is fixed, thanks for the bug reports, Neildark!
Also thanks for the savestates, Gab75 :)
-
Wow, ?how fast! Good work, I'm impressed. If you allow me two questions, ?when do you have to think about releasing a new version of the Final Burn Alpha? MAME is already capable of emulating PGM2 games, some interesting ones like The King of Fighters '98 Ultimate Match Hero, ?do you have plans to add the emulation later or is it very complicated? Thanks for everything ;)
-
Neildark what version of MAME did you use to play GI Joe? I tried the latest MAME and it has the same missing water as FBAlpha. Very old version of MAME had a linemapping mode which is probably drawing the water. I will see if we can impliment this effect into FBAlpha.
thanks,
- dink
-
Here's some wizardy fire fixes for view:
p.s. I'm not sure when the next version will be out.
-
As usual, excellent work dink! :)
-
You're right Dink, I've tried it in the latest version of MAME and keeps giving errors, which is a bit strange :confused:. The captures I got from videos on YouTube, but from what I have asked are from very old versions (MAME version 0.119). I thought it would be a case of bad emulation, but it seems that in MAME they can not solve it completely.
Thank you for your work.
-
Hi guys, I also wanted to report some problems with the kof2002 hacks, I know it does not matter much, but if you want to take a look, I checked and the romset is correct:
Greetings to all.
Kof2002 (hack omg) Error: FB Alpha Warning: ?There Was a problem Starting Kof2002 (hack omg)?
Kof2002 (Omega v0.8 Error: Guru meditation
Kof2002 (Omega v0.9 beta) Error: FB Alpha Warning: ?There Was a problem Starting Kof2002 (Omega v0.9 beta)?
Kof2002 (Omega v0.9) Error: FB Alpha Warning: ?There Was a problem Starting Kof2002 (Omega v0.9)?
The King of Fighters 10th Anniversary (bootleg) Error: Guru meditation
-
Thanks, that ione was actually fixed not too long ago :)
check your pm's
-dink
Hi guys, I also wanted to report some problems with the kof2002 hacks, I know it does not matter much, but if you want to take a look, I checked and the romset is correct:
Greetings to all.
Kof2002 (hack omg) Error: FB Alpha Warning: ?There Was a problem Starting Kof2002 (hack omg)?
Kof2002 (Omega v0.8 Error: Guru meditation
Kof2002 (Omega v0.9 beta) Error: FB Alpha Warning: ?There Was a problem Starting Kof2002 (Omega v0.9 beta)?
Kof2002 (Omega v0.9) Error: FB Alpha Warning: ?There Was a problem Starting Kof2002 (Omega v0.9)?
The King of Fighters 10th Anniversary (bootleg) Error: Guru meditation
-
A big thanks you to dink... other fixes to translucent effects in Wizard Fire! The maps at the beginning of each level and the spider demon (the middle boss of level 5)... :wink:
-
Bug Sound Cabal (World, Joystick)
Hello good, today I was playing the game in the FBA and used to do it in MAME I have detected that the sound is not the same: tank explosions, aircraft and houses collapse are much higher than the other FX and it is like reverberating, it is a bit annoying. If you can take a look at comparing the first level in both emulators, you can see it right away. A greeting!
-
Hello, I have already played a full game and the sound is already working correctly (there is a bit of reverb when there are many explosions but nothing important) but the final credits are not seen. I leave you a couple of images and a final savestate so you can check it out. A greeting.
Cabal (World, Joystick)
-
Thanks for the bug report, Neildark!
-
Hi everybody ! This is my first post on the forum ! :cool:
I've also the same problem with SHADOW DANCER on FB Alpha, using the latest version 0.2.97.43,
the set3 US doesn't work...there's the error message :
shdancer was found (C:\MamePlus\roms\shdancer).
? essential program ROM epr-12987.a4 was not found.
The ROMset exhibits the following problems:
? essential data is missing, the game will not run!
? program data is missing
And I've also the same problem with the game crashing at Stage 2-4.
You said you fixed the problem, ok. But how can I play with the fixed rom please ?
Thanks for your help !
See you soon !
-
Hi lordpaddle,
You'll need an updated romset for shdancer.zip to avoid the not found error.
The fix will be available in the next FBA release, which might not be too far away, I'm not 100% sure though.
The version before the mapper was added (.39) will work fine in the meantime if you'd like to play Shadow Dancer.
best regards,
- dink
-
Hi Dink !
Thanks for your answer, I'll check this !
Bye,
Lord P.
-
Hi everyone !
Road Fighter has a graphical issue (see https://github.com/libretro/fbalpha/issues/237 , the "viewport" is somehow offsetted to the right), the same is happening on standalone.
-
Hi everyone !
Road Fighter has a graphical issue (see https://github.com/libretro/fbalpha/issues/237 , the "viewport" is somehow offsetted to the right), the same is happening on standalone.
Thanks!
Find: GenericTilemapSetOffsets(0, 0, -16);
change to:
if (select == 2 || select == 3) // roadf*
GenericTilemapSetOffsets(0, 0, +16);
else
GenericTilemapSetOffsets(0, 0, -16);
best regards,
- dink
-
Hi everyone,
twinhawk is also having some issues, see the screenshot.
-
Oops, a regression from my c-chip stuff. sorry about that!
edit drv/taito/d_taitox.cpp, find TaitoX68KReadByte, add this in the switch statement:
case 0x900001: {
return TaitoInput[0];
}
case 0x900003: {
return TaitoInput[1];
}
case 0x900005: {
return TaitoInput[2];
}
best regards,
- dink
-
Thanks :)
-
In super sidekicks 2, the floor are and it looks fuzzy.
This happen with mvs configuration and the last bios
-
Hi, this game is really like this :( It's not a bug in emulation.
edit: oops, it seems I overlooked something, it is a bit worse in fba than mame. I'll see what can be done.
best regards,
- dink
In super sidekicks 2, the floor are and it looks fuzzy.
This happen with mvs configuration and the last bios
-
This, in neo_run.cpp may be worth a look first;
// if (!strcmp(BurnDrvGetTextA(DRV_NAME), "ssideki2")) {
// bRenderLineByLine = true;
// }
-
Treble Winner, Unfortunately that makes no difference. I found another way though.
-
Tiles priority issue at the end of Gradius II - GOFER no Yabou (Japan New ver.).
-
Tatsuya79, thanks for the report!
Can you please send me a savestate for this game ending? If you have one, right before this screen appears is preferred.
best regards,
- dink
-
@Tatsuya79: as dink wrote, a savestate (just before the game ending screen appears) will be very useful... :)
If I'm not wrong, Gradius 2 Japanese release hasn't the continue option and finish the game isn't an easy task! (I suck with the shoot'em ups! :P )
-
The one I got is from retroarch, which isn't compatible with fba stand-alone it seems.
Attaching it anyway.
-
Btw, savestates in the libretro port are only the result of the BurnAreaScan call (no deflate applied), reason for this being performance (we are using savestates for netplay, the no input lag thingy called "read ahead", and the rewind feature which i guess helped a lot to finish this game ?). Knowing this, i hope it's somehow possible to convert them ?
-
A different one that should be easier to check:
Wardner is running at 60hz while it is at 54.877858Hz in MAME.
Also games on the Burger Time Data East machine are at 60 instead of 57.444853Hz.
Tested with Burger Time and Burnin' Rubber.
Same for Break Thru.
Same for games on Deco casette System it seems (tested Lock'n'Chase and Super Astro Fighter).
-
Wardner is running at 60hz while it is at 54.877858Hz in MAME.
Yes, most likely the value at https://github.com/libretro/fbalpha/blob/4b5c569447dd245995e7d071787d5083dd7322c1/src/burn/drv/toaplan/d_wardner.cpp#L997 is wrong, is that supposed to be
INT32 nCyclesTotal[3] = { 5487785.8 / 60, 3500000 / 60, 14000000 / 60 };
?
If so, i can probably take care of those kind of issues and send the diff to dink later.
Edit : i did some tests and i guess it is trickier than this...
-
The office of internal FBA affairs (aka Gab75) is taking care of replaying gradius II to get me a state. :)
Regarding the hz: leaving some games at 60hz was intentional on my part. Also, iq_132 leaves some games at 60hz also, for either the same or different reason than me. Is this a big deal? if it is, I'll change them, but probably a good 80% of the drivers run at a solid 60hz which means a lot of dull work...
best regards,
- dink
-
Well, it's about having the games at their right speed.
I can understand you'd want it to be in sync with standard screens but as variable refresh gets more popular...
There's no throttling options to change the game speed to the monitor refresh in FBA?
I see "force 60hz refresh rate" in the video menu.
-
Can't argue with that - I sent barbudreadmon a refreshrate patch with the changes for those games, :)
best regards,
- dink
-
the reality is almost no arcade games run at 60hz, just in most cases nobody has bothered to accurately measure / calculate the variables which determine what the refresh rates should be.
-
The office of internal FBA affairs (aka Gab75) is taking care of replaying gradius II to get me a state. :)
Regarding the hz: leaving some games at 60hz was intentional on my part. Also, iq_132 leaves some games at 60hz also, for either the same or different reason than me. Is this a big deal? if it is, I'll change them, but probably a good 80% of the drivers run at a solid 60hz which means a lot of dull work...
best regards,
- dink
Many years ago, before you started fixing up sound cores, a lot of them crashed at anything but 60hz. I've gotten into the habit of just leaving them at that
-
Tatsuya79, thanks for the gradius2 bug report!
Konami twin16 hw got a huge re-work tonight:
gradius2 several priority issues fixed
devilw and miaj now has proper shadows
missing text in devilw fixed
miaj + cuebrickj render at the right size.
lots of small issues, as well :)
best regards,
- dink
-
Thank you the world is a better place now!
You don't want a giant sun going through your planet.
-
Tatsuya79, thanks for the excellent bug reports :)
best regards,
- dink
-
There's a problem with Darius bgm, drums in particular can go really loud and sound bad.
The 1st stage doesn't exhibit that, you have to reach zone B (up after 1st boss) to hear what it is about.
It's really awful at the sea stage (I think it was zone K when going always up after bosses).
It sounds better in Mame, hard to find a real arcade video with great audio.
I was playing Darius (Extra) (Japan).
edit: I did a video capture in the sea zone, 1st boss down, 2nd boss up way.
https://youtu.be/1joQzuT_Ehg
-
Thanks :) I'll have a fix soon. It looks like we were only playing the drums out of one channel - oops.
-
Hi,
It seems there are graphical glitches in king of dragons (in multiplayer ?) : https://www.twitch.tv/videos/321325247 (see bottom of screen at 00:33 and 02:07).
While the video is most likely from netplay in the libretro port, i don't think it is related.
-
Thanks :)
Tonight I'm back doing bug fixes, the ccastles driver tied me up for a couple days.
best regards,
- dink
-
barbudreadmon, check pm - I
think we took care of this issue with the recent timing changes :)
edit: verified w/old and new exe: yes, the timing changes take care of this issue.
best regards,
- dink
-
Tatsuya79,
Darius should be fine now (as soon as the second fix makes it into the lr build). There were 2 problems:
it was only outputting sound on one speaker, and due to some timing issue, the digital samples were being slightly distorted. Unfortunately the second one took me quite a while to track down. Thanks for the report!
best regards,
- dink
-
Here's 2 sound samples showing before / after digital audio bug in darius. The clip is from the boss death. Notice the clicks and distortion in "dariusbug.mp3", and the nice "dariusfixed.mp3". Due to the low bitrate of the samples in the game, dariusfixed.mp3 is as good as it gets (period).
best regards,
- dink
-
Nice work!
Sounds a lot more like a giant space fish agonizing scream.
-
thanks :)
-
Boulder Dash _ Boulder Dash Part 2 (Japan) (bouldashj)
Stuck notes or volume going down untill there's no sound any more randomly.
Seems to happen after playing 5/10mn.
Happened to me twice in world 2 (2nd time after restarting the game).
-
Tatsuya79,
Thanks for the report - I've updated some things in the sound section which hopefully takes care of the problem. Unfortunately I suck at this game too much to be able to fully test it :biggrin:
-
Finished world 2 without a sound issue so that's probably fixed.
Thank you!
-
Great news, thanks for testing :)
-
Drift Out '94 - The Hard Order (Japan) (SSV) is behaving strangely, not smooth at all like the internal timing is wrong.
Tested in current RA core and fba stand-alone 0.2.97.43 win7 x64.
-
Tatsuya79, terribly sorry - somehow I overlooked your post by accident :(
I see what you mean, or at least I notice that the game is weirdly-jittery. A fix should be available shortly :)
best regards,
- dink
-
Hello, I noticed on the SPIN MASTER game on neogeo-mvs blinking bug at the bottom of the image. It's easily remarkable if you let the demo start, you can see it very well where there is water.
On the original version there is no this bug
The test was performed on the latest version of FBneo on retroarch. thanks
-
sniper856,
I checked in a fix for this problem, thanks for the bug report :)
best regards,
- dink