MAMEWorld >> News
Previous thread Previous  View all threads Index   Next thread Next   Threaded Mode Threaded  

Pages: 1

ShimaPong
MAME Fan
Reged: 03/12/05
Posts: 783
Send PM


MAME 0.141
#242325 - 01/01/11 04:52 AM


http://mamedev.org/

0.141
-----


MAMETesters Bugs Fixed
----------------------
- 04161: [Crash/Freeze] nibbler and clones: Mame crashes after pressing
OK (Aaron Giles)
- 00859: [Color/Palette] blandia, blandiap: Tilemap colors issue.
(Pierpaolo Prazzoli)
- 00860: [Color/Palette] blandia, blandiap: Incorrect colors
(Pierpaolo Prazzoli)
- 04120: [Misc.] SDLMAME for Windows does not work unless you use
-window (couriersud)
- 04156: [Crash/Freeze] cabal, cabala, cabalbl2, cabalus, cabalus2,
tharrier, tharrierj: memory_set_bank called for unknown bank
'bank1' (hap)
- 04153: [Misc.] hydrthnd: SHA1 of CHD reported incorrectly.
(David Haywood)
- 03916: [Graphics] arabian: Missing bird at top of the screen.
(Phil Bennett)



Source Changes
--------------
Redumped sprite roms in X Se Dae Quiz [Guru]

Added preliminary (disabled by default) discrete sound emulation to
Dai San Wakusei Meteor [Derrick Renaud, Andy Welburn]

Added DMA slots to Legionnaire HW, fixes Godzilla regression
[Angelo Salese]

Fixed Bullet key back to what we dumped from original PCB.
[Aaron Giles]

Improved sprite GFX ROM order in Godzilla [Angelo Salese]

AY-3-8910 unused bits in registers read back as 0 [hap]

Fixed Seibu logo tilemap colors in Seibu Cup Soccer [Angelo Salese]

s3c24xx.c: Fixing compiling process on Linux and MacOSX [qmc2]

Assigned MB88xx interrupt sources unique vectors, as expected by the
Arabian MCU program [Phil Bennett]

Hooked up Arabian MB8841 MCU and removed obsolete simulation code
[Phil Bennett]

Adjusted ym2203 clock speed and added XTAL for goindol.c based on a
reference video recording. [Tafoid, Rogerio]

Rewrote ics2115 sound core using modern device standards, improving
audio in Oriental Legend, Killing Blade and other PGM titles in the
process. [austere, nimitz, trap15]

Implemented pdrawgfx for sprite routines in Legionnaire HW, and added
wraparound for horizontal area smaller than 320, fixing left-border
sprites in Legionnaire and Heated Barrel [David Haywood]

Implemented a preliminary sort-DMA in Seibu COP routines, fixes
sprite-sprite priorities in Seibu Cup Soccer [Angelo Salese]

z80dart.c: [Wilbert Pol]
* The channel B modified interrupt vector can be read immediately
after setting it.
* Reading from a read register other than 0 also masks out the
register index.

Fixed RAM fill in Seibu DMA COP command 0x116, allowing Godzilla to
boot [Angelo Salese]

Neo-Geo updates [Johnboy]:
* burningfh - Renamed p1 to hp1
* gowcaizr - Added correct C8 ROM
* 8man - Tagged MVS and AES version
* jockeygpa - Added alternate jockeygp set
* mslug5h - Corrected chip labels

Hooked up cocktail mode and proper controls for both players in Maze
Invaders, adding button labels to describe the action of each.
[sxevious, Tafoid]

Fixed LDM opcode register writeback in ARM CPU core [Tim Schuerewegen]

General stability issues with 64-bit builds fixed for cybertnk.c
[David Haywood]

Implemented palette DMA brightness algorythm mode 5 in Seibu COP
games, used by Seibu Cup Soccer, Godzilla, SD Gundam Rainbow Tairiku
Senki and Legionnaire for fading effects [Angelo Salese]

hvyunit.c: Hooked up Mermaid protection MCU and removed obsolete
simulation code. Also updated the driver to use a state class.
[Phil Bennett, Dr. Decapitator, Andrew Gardner]

Implemented preliminary road emulation in Cyber Tank [David Haywood]

Renamed ilpag.c as blitz68k.c and added several games to the driver
[Luca Elia]

sdlmame now also looks for fonts in "-fontpath". In addition, moved
font config stuff into function search_font_config. This is in
preparation to add ttf support to windows sdlmame build. [couriersud]

Fixed an ARM CPU core bug with pre-indexed addressing opcodes, that
was preventing Poizone to boot [Tim Schuerewegen]

Fix most-likely-to-fail SDL_TTF version check. Added a warning if
strikethrough is specified but not supported. [Couriersud]

Reimplemented scatter-gather Sprite DMA in Seibu COP used by SD Gundam
Rainbow Tairiku Senki, also improving the functionality in it.
[Angelo Salese]

Greatly improved inputs and dip switches for ertictac.c [Tafoid]

Fixed playfield color issues in Seibu Cup Soccer [David Haywood]

Fixed *BSD compilation for SDLMAME. [ElBarto]

Fixed sfx in Chouji Meikyuu Legion [Tomasz Slanina]

Implemented preliminary Seibu COP macro command 0xdde5, used by Seibu
Cup Soccer x/y radar positions [Angelo Salese]

Implemented ROM banking support in Raiden 2 HW, improving Raiden 2 and
Zero Team behaviour [O.Galibert]

plygonet.c : Fixed save state & device handler regressions.
[Andrew Gardner]

Golden Poker driver: Added new sets, and some notes about ipc-1
daughterboard encryption [Roberto Fresca]

Preliminary simulation of text layer protection in Chouji Meikyuu
Legion [Tomasz Slanina]

Further ICS2115 work [austere, nimitz, trap15]

mpu4drvr.c: Account for the 45 degree trackball mounting
[James Wallace]

Adds the recent 87C52 MCU decaps to Wonder League Star and Wonder
League '96 [Dr. Decapitator]

superchs.c - Moved Button 6 assignment to a DIP selection which is
more appropriate for "Freeze Screen" behavior and added some missing
port names. [Tafoid]

Updated the RSP and RDP implementations to be largely bit-perfect and
pixel-accurate. [angrylion, Harmony]

Fixed alpha compare in the N64 RDP implementation. [Harmony]

More Golden Poker driver cleanup. Added more references and
reorganized some parent-clone relationship. Added some new sets.
[Roberto Fresca]

Implemented Seibu COP macro 0x905, used for ark movements in
Legionnaire / Zero Team when a knock down occurs [Angelo Salese]

ARM7+ updates: [Tim Schuerewegen]
* Enhanced MMU with page fault support
* Preliminary 26-bit mode support
* Fixed bugs with S-flag operations on R15
* Fixed STRH involving R15
* Support for ARM models 7500 and 920T.

Add directxor parameter to direct memory accessors. Update all callers
that manually XORed addresses to pass the XOR instead. This improves
behavior when direct accessors hit non-RAM regions, or when watchpoints
are used, because we now know the original, un-xored address and can
fall back to standard read/write handlers properly. Also fixes
glitched disassembly when read watchpoints are enabled. [Aaron Giles]

s3c24xx: some improvements and fixes [Tim Schuerewegen]

Removed public decodechar function. [Aaron Giles]

Fixed writes in the MIPS III interpreter. [MooglyGuy]

Some bulk naming cleanups across the whole source base: [Aaron Giles]
1. MDRV_* -> MCFG_*
2. running_device -> device_t
3. mame_rand() -> machine->rand()
4. memory_region() -> machine->region()->base()
5. memory_region_length() -> machine->region()->bytes()
6. region_info -> memory_region



New games added or promoted from NOT_WORKING status
---------------------------------------------------
Heavy Unit [Phil Bennett, Dr. Decapitator, Andrew Gardner]
Poizone [Tim Schuerewegen]
Silver Game [Roberto Fresca, Team Europe & HAL9K]
Jack Potten's Poker (NGold) (3 sets) [Roberto Fresca, Darran]



New clones added
----------------
Multi Champ Deluxe (ver. 1114, 14/11/1999) [caius] (not working)
Gran Tesoro? / Play 2000 (v4.0) (Italy)
[Corrado Tomaselli] (not working)
Raiden (US, set 2) [Smitdogg, The Dumping Union]
Golden Tee Golf (Trackball, v2.0) [Smitdogg, The Dumping Union]
Relief Pitcher (set 2, 26 Apr 1992 / 08 Apr 1992)
[Smitdogg, The Dumping Union]
G.I. Joe (World, EB8, prototype?) [caius]
Space Firebird (2 sets) [Andrew Welburn]



New games marked as GAME_NOT_WORKING
------------------------------------
Bank Robbery (Ver. 2.00)
[ChrisQC, Smitdogg, Luca Elia, The Dumping Union]
Bank Robbery (Ver. 3.32)
[ChrisQC, Smitdogg, Luca Elia, The Dumping Union]
Deuces Wild 2 - American Heritage (Ver. 2.02F)
[ChrisQC, Smitdogg, Luca Elia, The Dumping Union]
Maxi Double Poker (Ver. 1.10)
[ChrisQC, Smitdogg, Luca Elia, The Dumping Union]
Mega Double Poker (Ver. 1.63 Espagnol)
[ChrisQC, Smitdogg, Luca Elia, The Dumping Union]
Mega Double Poker Jackpot (Ver. 1.26)
[ChrisQC, Smitdogg, Luca Elia, The Dumping Union]
Poker 52 (Ver. 1.2) [ChrisQC, Smitdogg, Luca Elia, The Dumping Union]
The Hermit (Ver. 1.14)
[ChrisQC, Smitdogg, Luca Elia, The Dumping Union]
Mambo A Go-Go (GQA40 VER. JAB)
[Kevin Eshbach, Tormod, gatt9, Smitdogg, The Dumping Union, smf]
"Unknown french poker game" [Roberto Fresca, Team Europe & HAL9K]
"Unknown encrypted poker game" [Roberto Fresca, Team Europe & HAL9K]

Edited by twisty (01/01/11 07:36 AM)



"Any company has no power to stop people emulating"
MAME is the emulator of no giving in the pressure from any company even if they don't allow



abelenki
Super Drunk
Reged: 10/20/03
Posts: 962
Loc: Perm, Russia
Send PM


Re: MAME 0.141 new [Re: ShimaPong]
#242327 - 01/01/11 04:59 AM


> http://mamedev.org/

eh? there is no MAME 0.141.

p.s.: can someone ban this idiot? he's getting really annoying...

Edited by Anton Belenki (01/01/11 05:04 AM)



MAME Stuff | MAME favorites | Bubble Bobble Series chart / wiki | PC configuration



R. Belmont
Cuckoo for IGAvania
Reged: 09/21/03
Posts: 9711
Loc: ECV-197 The Orville
Send PM


Re: MAME 0.141 new [Re: ShimaPong]
#242328 - 01/01/11 05:07 AM


Oh, good, we're back to *this* bullshit again where partial whatsnew.txt files get leaked.

For the record:


0.141
-----


MAMETesters Bugs Fixed
----------------------
- 04161: [Crash/Freeze] nibbler and clones: Mame crashes after pressing
OK (Aaron Giles)
- 00859: [Color/Palette] blandia, blandiap: Tilemap colors issue.
(Pierpaolo Prazzoli)
- 00860: [Color/Palette] blandia, blandiap: Incorrect colors
(Pierpaolo Prazzoli)
- 04120: [Misc.] SDLMAME for Windows does not work unless you use
-window (couriersud)
- 04156: [Crash/Freeze] cabal, cabala, cabalbl2, cabalus, cabalus2,
tharrier, tharrierj: memory_set_bank called for unknown bank
'bank1' (hap)
- 04153: [Misc.] hydrthnd: SHA1 of CHD reported incorrectly.
(David Haywood)
- 03916: [Graphics] arabian: Missing bird at top of the screen.
(Phil Bennett)



Source Changes
--------------
Redumped sprite roms in X Se Dae Quiz [Guru]

Added preliminary (disabled by default) discrete sound emulation to
Dai San Wakusei Meteor [Derrick Renaud, Andy Welburn]

Added DMA slots to Legionnaire HW, fixes Godzilla regression
[Angelo Salese]

Fixed Bullet key back to what we dumped from original PCB.
[Aaron Giles]

Improved sprite GFX ROM order in Godzilla [Angelo Salese]

AY-3-8910 unused bits in registers read back as 0 [hap]

Fixed Seibu logo tilemap colors in Seibu Cup Soccer [Angelo Salese]

s3c24xx.c: Fixing compiling process on Linux and MacOSX [qmc2]

Assigned MB88xx interrupt sources unique vectors, as expected by the
Arabian MCU program [Phil Bennett]

Hooked up Arabian MB8841 MCU and removed obsolete simulation code
[Phil Bennett]

Adjusted ym2203 clock speed and added XTAL for goindol.c based on a
reference video recording. [Tafoid, Rogerio]

Rewrote ics2115 sound core using modern device standards, improving
audio in Oriental Legend, Killing Blade and other PGM titles in the
process. [austere, nimitz, trap15]

Implemented pdrawgfx for sprite routines in Legionnaire HW, and added
wraparound for horizontal area smaller than 320, fixing left-border
sprites in Legionnaire and Heated Barrel [David Haywood]

Implemented a preliminary sort-DMA in Seibu COP routines, fixes
sprite-sprite priorities in Seibu Cup Soccer [Angelo Salese]

z80dart.c: [Wilbert Pol]
* The channel B modified interrupt vector can be read immediately
after setting it.
* Reading from a read register other than 0 also masks out the
register index.

Fixed RAM fill in Seibu DMA COP command 0x116, allowing Godzilla to
boot [Angelo Salese]

Neo-Geo updates [Johnboy]:
* burningfh - Renamed p1 to hp1
* gowcaizr - Added correct C8 ROM
* 8man - Tagged MVS and AES version
* jockeygpa - Added alternate jockeygp set
* mslug5h - Corrected chip labels

Hooked up cocktail mode and proper controls for both players in Maze
Invaders, adding button labels to describe the action of each.
[sxevious, Tafoid]

Fixed LDM opcode register writeback in ARM CPU core [Tim Schuerewegen]

General stability issues with 64-bit builds fixed for cybertnk.c
[David Haywood]

Implemented palette DMA brightness algorythm mode 5 in Seibu COP
games, used by Seibu Cup Soccer, Godzilla, SD Gundam Rainbow Tairiku
Senki and Legionnaire for fading effects [Angelo Salese]

hvyunit.c: Hooked up Mermaid protection MCU and removed obsolete
simulation code. Also updated the driver to use a state class.
[Phil Bennett, Dr. Decapitator, Andrew Gardner]

Implemented preliminary road emulation in Cyber Tank [David Haywood]

Renamed ilpag.c as blitz68k.c and added several games to the driver
[Luca Elia]

sdlmame now also looks for fonts in "-fontpath". In addition, moved
font config stuff into function search_font_config. This is in
preparation to add ttf support to windows sdlmame build. [couriersud]

Fixed an ARM CPU core bug with pre-indexed addressing opcodes, that
was preventing Poizone to boot [Tim Schuerewegen]

Fix most-likely-to-fail SDL_TTF version check. Added a warning if
strikethrough is specified but not supported. [Couriersud]

Reimplemented scatter-gather Sprite DMA in Seibu COP used by SD Gundam
Rainbow Tairiku Senki, also improving the functionality in it.
[Angelo Salese]

Greatly improved inputs and dip switches for ertictac.c [Tafoid]

Fixed playfield color issues in Seibu Cup Soccer [David Haywood]

Fixed *BSD compilation for SDLMAME. [ElBarto]

Fixed sfx in Chouji Meikyuu Legion [Tomasz Slanina]

Implemented preliminary Seibu COP macro command 0xdde5, used by Seibu
Cup Soccer x/y radar positions [Angelo Salese]

Implemented ROM banking support in Raiden 2 HW, improving Raiden 2 and
Zero Team behaviour [O.Galibert]

plygonet.c : Fixed save state & device handler regressions.
[Andrew Gardner]

Golden Poker driver: Added new sets, and some notes about ipc-1
daughterboard encryption [Roberto Fresca]

Preliminary simulation of text layer protection in Chouji Meikyuu
Legion [Tomasz Slanina]

Further ICS2115 work [austere, nimitz, trap15]

mpu4drvr.c: Account for the 45 degree trackball mounting
[James Wallace]

Adds the recent 87C52 MCU decaps to Wonder League Star and Wonder
League '96 [Dr. Decapitator]

superchs.c - Moved Button 6 assignment to a DIP selection which is
more appropriate for "Freeze Screen" behavior and added some missing
port names. [Tafoid]

Updated the RSP and RDP implementations to be largely bit-perfect and
pixel-accurate. [angrylion, Harmony]

Fixed alpha compare in the N64 RDP implementation. [Harmony]

More Golden Poker driver cleanup. Added more references and
reorganized some parent-clone relationship. Added some new sets.
[Roberto Fresca]

Implemented Seibu COP macro 0x905, used for ark movements in
Legionnaire / Zero Team when a knock down occurs [Angelo Salese]

ARM7+ updates: [Tim Schuerewegen]
* Enhanced MMU with page fault support
* Preliminary 26-bit mode support
* Fixed bugs with S-flag operations on R15
* Fixed STRH involving R15
* Support for ARM models 7500 and 920T.

Add directxor parameter to direct memory accessors. Update all callers
that manually XORed addresses to pass the XOR instead. This improves
behavior when direct accessors hit non-RAM regions, or when watchpoints
are used, because we now know the original, un-xored address and can
fall back to standard read/write handlers properly. Also fixes
glitched disassembly when read watchpoints are enabled. [Aaron Giles]

s3c24xx: some improvements and fixes [Tim Schuerewegen]

Removed public decodechar function. [Aaron Giles]

Fixed writes in the MIPS III interpreter. [MooglyGuy]

Some bulk naming cleanups across the whole source base: [Aaron Giles]
1. MDRV_* -> MCFG_*
2. running_device -> device_t
3. mame_rand() -> machine->rand()
4. memory_region() -> machine->region()->base()
5. memory_region_length() -> machine->region()->bytes()
6. region_info -> memory_region



New games added or promoted from NOT_WORKING status
---------------------------------------------------
Heavy Unit [Phil Bennett, Dr. Decapitator, Andrew Gardner]
Poizone [Tim Schuerewegen]
Silver Game [Roberto Fresca, Team Europe & HAL9K]
Jack Potten's Poker (NGold) (3 sets) [Roberto Fresca, Darran]



New clones added
----------------
Multi Champ Deluxe (ver. 1114, 14/11/1999) [caius] (not working)
Gran Tesoro? / Play 2000 (v4.0) (Italy)
[Corrado Tomaselli] (not working)
Raiden (US, set 2) [Smitdogg, The Dumping Union]
Golden Tee Golf (Trackball, v2.0) [Smitdogg, The Dumping Union]
Relief Pitcher (set 2, 26 Apr 1992 / 08 Apr 1992)
[Smitdogg, The Dumping Union]
G.I. Joe (World, EB8, prototype?) [caius]
Space Firebird (2 sets) [Andrew Welburn]



New games marked as GAME_NOT_WORKING
------------------------------------
Bank Robbery (Ver. 2.00)
[ChrisQC, Smitdogg, Luca Elia, The Dumping Union]
Bank Robbery (Ver. 3.32)
[ChrisQC, Smitdogg, Luca Elia, The Dumping Union]
Deuces Wild 2 - American Heritage (Ver. 2.02F)
[ChrisQC, Smitdogg, Luca Elia, The Dumping Union]
Maxi Double Poker (Ver. 1.10)
[ChrisQC, Smitdogg, Luca Elia, The Dumping Union]
Mega Double Poker (Ver. 1.63 Espagnol)
[ChrisQC, Smitdogg, Luca Elia, The Dumping Union]
Mega Double Poker Jackpot (Ver. 1.26)
[ChrisQC, Smitdogg, Luca Elia, The Dumping Union]
Poker 52 (Ver. 1.2) [ChrisQC, Smitdogg, Luca Elia, The Dumping Union]
The Hermit (Ver. 1.14)
[ChrisQC, Smitdogg, Luca Elia, The Dumping Union]
Mambo A Go-Go (GQA40 VER. JAB)
[Kevin Eshbach, Tormod, gatt9, Smitdogg, The Dumping Union, smf]
"Unknown french poker game" [Roberto Fresca, Team Europe & HAL9K]
"Unknown encrypted poker game" [Roberto Fresca, Team Europe & HAL9K]



tamalmalamarrado
MAME Fan
Reged: 06/10/07
Posts: 48
Send PM


Re: MAME 0.141 new [Re: R. Belmont]
#242329 - 01/01/11 05:28 AM


Maybe he got it from wikileaks, .

BTW, nice changelog (y)(y)(y)



redk9258
Regular
Reged: 09/21/03
Posts: 3968
Loc: Troy, Illinois USA
Send PM


Re: MAME 0.141 new [Re: abelenki]
#242330 - 01/01/11 05:31 AM


> p.s.: can someone ban this idiot? he's getting really annoying...

I second the motion!

How about a mod just nuke the whole thread!

Edited by redk9258 (01/01/11 05:34 AM)



CptGuapo
Beat'em-ups Lover
Reged: 03/18/08
Posts: 342
Loc: Off to Never Never Land
Send PM


Re: MAME 0.141 new [Re: ShimaPong]
#242333 - 01/01/11 06:01 AM


That's why I was thinking my RSS feed doesn't be functional...There's nothing on MAMEDev.org...

Hmmm...Shima, fella...you're in trouble. Again.



"Mythology is what grownups believe, folklore is what they tell children and religion is both."



redk9258
Regular
Reged: 09/21/03
Posts: 3968
Loc: Troy, Illinois USA
Send PM


Re: MAME 0.141 new [Re: CptGuapo]
#242334 - 01/01/11 06:54 AM


It's live now......



gregf
Ramtek's Trivia promoter
Reged: 09/21/03
Posts: 8589
Loc: southern CA, US
Send PM


Re: MAME 0.141 new [Re: R. Belmont]
#242341 - 01/01/11 08:54 AM




======
0.141
-----

Source Changes
--------------


Added preliminary (disabled by default) discrete sound emulation to
Dai San Wakusei Meteor [Derrick Renaud, Andy Welburn]
=======

Interesting.


As for the various Seibu games, good to see lots of improvements.
I am guessing the Legionnaire HW games will be working much sooner compared to Raiden II.



R. Belmont
Cuckoo for IGAvania
Reged: 09/21/03
Posts: 9711
Loc: ECV-197 The Orville
Send PM


Re: MAME 0.141 new [Re: gregf]
#242342 - 01/01/11 09:01 AM


> As for the various Seibu games, good to see lots of improvements.
> I am guessing the Legionnaire HW games will be working much sooner compared to
> Raiden II.

Probably, although Raiden II in 0.141 can be coined up and actually plays with sound and enemies and powerups and stuff. The graphics are fairly garbly and there's invisible stuff that kills you frequently though, and it can be made to glitch completely out and crash MAME. Still, it's noticable progress from where things were even recently.



krick
Get Fuzzy
Reged: 02/09/04
Posts: 4235
Send PM


Re: MAME 0.141 new [Re: R. Belmont]
#242343 - 01/01/11 09:30 AM


>
> Fixed Bullet key back to what we dumped from original PCB.
> [Aaron Giles]
>

What's the full story on this one?

I remember reading Aarons blog a while back about this game...
http://aarongiles.com/?p=207
But I don't recall anyone posting a status update about dumping the key or anything. Maybe I was just not paying attention.


On an unrelated side-note, what's the deal with "bagofglass.com"? It looks like Aaron's entire site is mirrored there. Example...
http://bagofglass.com/?p=207



GroovyMAME support forum on BYOAC



ShimaPong
MAME Fan
Reged: 03/12/05
Posts: 783
Send PM


Mambo A Go-Go new [Re: ShimaPong]
#242356 - 01/01/11 03:37 PM Attachment: MamboAGoGo.jpg 186 KB (1 downloads)


Mambo A Go-Go (Konami 06.2001)

[ TIPS ]
- Extra Stage
<<See scanned image>>

- Special Mode
In the music selection screen, input the following code (NOTE : L = Left Conga, C = Center Conga, R = Right Conga)
-- HIDDEN : Hold L then tap R, R, R
-- SUDDEN : Hold L then tap R, R, R, R, R, R
-- BLINK : Hold L then tap R, R, R, R, R, R, R, R, R
-- MIRROR : Tap L, R, L, R, L, R
-- RANDOM : Tap L, C, R, L, C, R
-- HI-SPEED2 : Hold L then tap C, C, C, C
-- HI-SPEED4 : Hold L then tap C, C, C, C, C, C, C, C

If you want to cancel special mode, tap R, C, L


- Secret Mode
In the mode selection screen, move cursor to Hard then tap R many times.
You will play with Super Hard. The cancel is tap L many time.

[ATTACHED IMAGE - CLICK FOR FULL SIZE]

Attachment



"Any company has no power to stop people emulating"
MAME is the emulator of no giving in the pressure from any company even if they don't allow



Cyberzinho Punk
Reged: 12/31/09
Posts: 182
Loc: Săo José dos Campos, SP, Brazil
Send PM


Re: MAME 0.141 new [Re: ShimaPong]
#242358 - 01/01/11 03:56 PM


Cool!!!!!!



Sorry, my English is bad!!!
Slackware Linux 14.2 beta 2/Fluxbox 1.3.7
Linux user #438128
MAME for Slackware



hap
Reged: 12/01/08
Posts: 296
Send PM


Re: MAME 0.141 new [Re: redk9258]
#242361 - 01/01/11 04:53 PM


So he's got a friend with access to the mailinglist and/or svn, who cares Happy new year and new MAME.



Shoegazr
Rockstar
Reged: 01/21/06
Posts: 658
Send PM


Re: MAME 0.141 new [Re: ShimaPong]
#242364 - 01/01/11 05:31 PM


Great work to everyone involved with this release. MAME .141 is replete with stratospheric improvements on many fronts.

As a side note, I happened to check out Strider 2 with .141 and noticed it's still crashing. That said, I'm wondering if the driver should be marked as "Not Working" for now. It crashes more frequently than the MAMEtesters report would have you believe, and is really not playable.



Alexis B.
Historian
Reged: 09/20/03
Posts: 415
Loc: Cannes, FRANCE
Send PM


Re: Mambo A Go-Go new [Re: ShimaPong]
#242365 - 01/01/11 05:44 PM


Tips added to arcade-history, thanks.



See my collection: here



R. Belmont
Cuckoo for IGAvania
Reged: 09/21/03
Posts: 9711
Loc: ECV-197 The Orville
Send PM


Re: MAME 0.141 new [Re: krick]
#242380 - 01/01/11 07:56 PM


> What's the full story on this one?
>
> I remember reading Aarons blog a while back about this game...
> http://aarongiles.com/?p=207
> But I don't recall anyone posting a status update about dumping the key or anything.
> Maybe I was just not paying attention.

It goes something like this:

1) Guru posts about getting "Dai 3 Wakusei" that some dude paid $1000 for, and that Dox has started working on a driver for it
2) Team Japump! tells MAMEdev and the DU that they've been hoarding a dump and driver of that game for 2+ years, and by the way they also dumped the key for Bullet

So I guess the moral of the story is that Guru should just randomly claim he's received rare Japanese PCBs occasionally to see what shakes loose ;-)



AaronGiles
Galaxiwarrior
Reged: 09/21/03
Posts: 1343
Send PM


Re: MAME 0.141 new [Re: R. Belmont]
#242389 - 01/01/11 09:37 PM


> It goes something like this:
>
> 1) Guru posts about getting "Dai 3 Wakusei" that some dude paid $1000 for, and that
> Dox has started working on a driver for it
> 2) Team Japump! tells MAMEdev and the DU that they've been hoarding a dump and driver
> of that game for 2+ years, and by the way they also dumped the key for Bullet

Which they also hoarded for over a year, and for which I spent even more than "Dai 3 Wakusei" to purchase. F*ckers.



AaronGiles
Galaxiwarrior
Reged: 09/21/03
Posts: 1343
Send PM


Re: MAME 0.141 new [Re: krick]
#242393 - 01/01/11 09:43 PM


> > Fixed Bullet key back to what we dumped from original PCB.
> > [Aaron Giles]
>
> What's the full story on this one?

It was purchased for an enormous amount of money from hoarders in Japan. Charles MacDonald dumped the FD1094.

Chris Hardy examined the dumped key in detail and detected what appeared to be a single bit error. (We have seen them before in FD1094 dumps, and the proposed fix did correct what appeared to be incorrect disassembly.)

H*ze pointed out that the key as originally dumped fixed the second level crash problem, while the version with the single bit "fix" caused it to happen again, so we reverted to the version that was dumped.

The fact that the proposed fix was completely plausible also helps to explain why my hand-constructed key had the same problem, because I too had assumed that the bit in question was set such that the disassembly would come out right.



Derrick Renaud
Discrete Coder
Reged: 10/18/03
Posts: 438
Loc: The Local Pub
Send PM


Re: MAME 0.141 new [Re: AaronGiles]
#242397 - 01/01/11 11:26 PM


> > It goes something like this:
> >
> > 1) Guru posts about getting "Dai 3 Wakusei" that some dude paid $1000 for, and that
> > Dox has started working on a driver for it
> > 2) Team Japump! tells MAMEdev and the DU that they've been hoarding a dump and
> driver
> > of that game for 2+ years, and by the way they also dumped the key for Bullet
>
> Which they also hoarded for over a year, and for which I spent even more than "Dai 3
> Wakusei" to purchase. F*ckers.

Bummer, I thought it was well known that we had access to Dai San Wakusei for at least a year. I was sworn to secrecy.



Do not p-mail me for help compiling my updates, ask on the board.
Do not request sound for your favorite game. I work on whatever, when I get around to it.
If you have schematics for discrete sound games not easily found on the net, I would be interested.



R. Belmont
Cuckoo for IGAvania
Reged: 09/21/03
Posts: 9711
Loc: ECV-197 The Orville
Send PM


Re: MAME 0.141 new [Re: Derrick Renaud]
#242404 - 01/02/11 02:07 AM


> > Which they also hoarded for over a year, and for which I spent even more than "Dai
> 3
> > Wakusei" to purchase. F*ckers.
>
> Bummer, I thought it was well known that we had access to Dai San Wakusei for at
> least a year. I was sworn to secrecy.

That would've been your first clue that something's horribly wrong. Nothing legit with MAME ever involves being sworn to secrecy.



sev
ScummVM Team co-leader
Reged: 10/02/05
Posts: 9
Send PM


Re: MAME 0.141 new [Re: AaronGiles]
#242445 - 01/02/11 12:44 PM


> It was purchased for an enormous amount of money from hoarders in Japan. Charles
> MacDonald dumped the FD1094.
>
> Chris Hardy examined the dumped key in detail and detected what appeared to be a
> single bit error.
So, it turned to be the most expensive bit in MAME so far


Eugene



ShimaPong
MAME Fan
Reged: 03/12/05
Posts: 783
Send PM


Re: MAME 0.141 new [Re: ShimaPong]
#242459 - 01/02/11 04:09 PM Attachment: hvyunitj.png 367 KB (1 downloads)


=== cheat3.c ===
- Added cheat_get_xml_attribute_int() to return a value from attribute. Only difference from xml_get_attribute_int() is default format is 'hex', not 'dec' though.

It is very important for me that which value is returned, 'hex' or 'dec'.
Sometimes I fail with unexpected error due to 'dec' value after load a code. In many cases, default format for a value is 'hex' (game program code, debugger expression etc).


- Refined cheat_function (structure) to function_entry (class). But it causes "invalid parameter" via execute_function() in express.c so that I can confirm read a data from file but can't test it...


Also I try searching a memory for Heavy Unit on debugger search system but it's very hard. Credit and Life are stored in Mermeid directly and I can't see the contents of memory for this device.
It uses specified handler to access a value and not mapped standard device so that the debug search and expression can't read/write directly.
In addition, different routine is done between Master and Slave CPU (especially, collision check) and some routines are done even in banked region (it's hard to trace due to different code in the same address.
Anyway, I finish searching/creating some code though it's enough to test the game. The emulation seems to be very good. I can see the ending without any crash/freeze/reset problem.

[ATTACHED IMAGE - CLICK FOR FULL SIZE]

Attachment



ShimaPong
MAME Fan
Reged: 03/12/05
Posts: 783
Send PM


Re: MAME 0.141 new [Re: ShimaPong]
#242524 - 01/03/11 05:05 PM Attachment: denjinmk.png 76 KB (1 downloads)


> But it causes "invalid parameter" via execute_function() in express.c

After tracing express.c via GDB, I find the point of problem in case of function. Now passed "invalid parameter" error but it still has problem.
Expression function doesn't deal with multi-parameters.

Two sample codes via user defined function should write "1234 5678 FEDC BA98" (big endiannee) between $123000 and $123007.
But unexpected value is set in fact.

- Where is first parameter?
- Why does param0 recieve second parameter?
- What parameter is recieved by param1 instead of second parameter?

I don't know the reason is due to my change to prevent from the expression error right now so that I still need to investigate more.



And the person who buy and send dai3wksi PCB last year reports interesting message from Guru.
According to the message, it will fact that Japump has already had this PCB and written the driver before.
It has happend before and will happen in the future that some rare PCB is not opened due to the political motivation by Japanese collector.

In response, one anonymous says,
"It is fact but sound is discreate and no color map ROM so that send it to overseas Dev to check.
If Guru's dump was accepted, the game will not have a color because his PCB is not original.
But he hear that Guru doesn't send because the content of ROM is the same and also he guesses that Guru already have known Japump has already it.
Because Guru has asked the person who can read a trace of discreate on PCB when send a PCB to him.

And the same time, the PCB of Chackn' Pop is send to Guru with express delivery charges to decap quickly.
But Guru don't send to Dr.Decap while a month because Guru send PCBs he had in a lump.
But the decapping has not been done still because has been unable to release due to the same person who provides dai3wksi and chaknpop"

[ATTACHED IMAGE - CLICK FOR FULL SIZE]

Attachment



SmitdoggAdministrator
Reged: 09/18/03
Posts: 16877
Send PM


Re: MAME 0.141 new [Re: ShimaPong]
#242525 - 01/03/11 05:27 PM


I'm not sure what Guru would have known about Japump's 3rd Planet board but basically everyone who has been on the DU for a while knew Japump had it because they told us they did. But this guy who recently bought it didn't ask us if it was needed, he just asked Guru. Guru isn't on the DU list and his memory can get pretty shoddy sometimes, if he was ever actually told about it. About the Chack'n Pop MCU, it was sent with payment for quick service and it got the quick service. The problem is that it was damaged during decapping and now requires special parts and labor to fix to a point of being able to be read out. It was damaged many months ago. This is a risk that goes with sending any chips to be decapped as seen with others like the first Bonze Adventure, and shouldn't be a surprise. The doc didn't even know it was special afaik, he just saw a 68705 with a number on it and a message to do it next. And anyway, hoarding dumps because of something one person (the doc) did is retarded at best.



ShimaPong
MAME Fan
Reged: 03/12/05
Posts: 783
Send PM


Re: MAME 0.141 new [Re: ShimaPong]
#242631 - 01/04/11 03:38 PM Attachment: denjinmk.png 81 KB (1 downloads)


> - Where is first parameter?
> - Why does param0 recieve second parameter?
> - What parameter is recieved by param1 instead of second parameter?


I find problem on debug/express.c:1646


Code:


case TVL_COMMA:
if (!token->is_function_separator())
{
pop_token_rval(t2); pop_token_rval(t1);
push_token(t2);
}
break;



"pop(t2)->pop(t1)->push(t2)" causes two problems.

- younger stack is overwritten by older stack
- stack pointer is different after this sequence.

I guess that it is "swapping younger and older stack" when ',' is found in the expression function and expected sequence is "pop(t2)->pop(t1)->push(t2)->push(t1)".
In this way, younger and older stacks are swapped and stack pointer is the same between before and ago.

So I put "push_token(t1)" then try again...

...???
All parameters will be recieved in function entry but still unexptected value.
It seems that the order of parameters are completely wrong. Left (and first) parameter value is accepted as param3 instead of param0 and second = param0, third = param1 and forth = param2.

[ATTACHED IMAGE - CLICK FOR FULL SIZE]

Attachment



ShimaPong
MAME Fan
Reged: 03/12/05
Posts: 783
Send PM


Re: MAME 0.141 new [Re: ShimaPong]
#242739 - 01/05/11 03:25 PM


Damn! I have overlooked important sign completely!!

"!token->is_function_separator()" means "if this comma is NOT function separator" so that it SHOULD ignore if it is function separator.
Otherwords, the function separator is NOT recognized by expression engine.

Why does it happens? See normalize_operator():1219

Code:


// Determine if , refers to a function parameter
case TVL_COMMA:
for (int lookback = 0; lookback < MAX_STACK_DEPTH; lookback++)
{
parse_token *peek = peek_token(lookback);
if (peek == NULL)
break;

// if we hit an execute function operator, or else a left parenthesis that is
// already tagged, then tag us as well
if (peek->is_operator(TVL_EXECUTEFUNC) || (peek->is_operator(TVL_LPAREN) && peek->is_function_separator()))
{
thistoken.set_function_separator();
break;
}
}
break;


This function is called in parsing phase and checks some operators.
peek_token() picks up requested parse_token from internal token_stack array by index.
But this array is used in executing phase and all are empty in parsing phase.
So "peek == NULL" is always TRUE and never set the flag of function separator.

Troublingly, parse_token uses simple_list template class and it only have the pointer of next and no previous.
And you can only see next and previous parse_tokens in this function.
In sample code, first comma can see 0x1234 and 0x5678.
But you will say, "loop over token list made by parse_string_into_tokens()". It's half correct but half wrong.
Because infix_to_postfix() which calls normalize_operator() detaches and reassembles main token list (it's not copy list!)
So it will not be ensured that the token list in this function is not original parsed by parse_string_into_tokens().

So you will need to set complicated routine to detect.


Pages: 1

MAMEWorld >> News
Previous thread Previous  View all threads Index   Next thread Next   Threaded Mode Threaded  

Extra information Permissions
Moderator:  John IV, Robbbert, Tafoid 
0 registered and 21 anonymous users are browsing this forum.
You cannot start new topics
You cannot reply to topics
HTML is enabled
UBBCode is enabled
Thread views: 5761