05041: kof2001: Crash in the middle of the ending staff credit.

ID	Category [?]	Severity [?]	Reproducibility	Date Submitted	Last Update
05041	Gameplay	Critical (emulation)	Always	Oct 21, 2012, 12:48	13 days ago

Tester	EVA7GDB	View Status	Public	Platform	MAME (Unofficial Binary)
Assigned To		Resolution	Open	OS	Windows XP
Status [?]	Confirmed			Driver	neogeo/neogeo.cpp
Version	0.147	Fixed in Version		Build	Normal
		Fixed in Git Commit		Github Pull Request #

Summary	05041: kof2001: Crash in the middle of the ending staff credit.
Description	The game crashes in the middle of the ending staff credit with any team combination, it resets to the title screen. The bug can be seen in any of the parent and clone sets. -- Originally it had occurred as a symptom since I migrated to NEODRV.C, but had not posted from the recognition that it would be fixed sometime, I think not heal forever, and it's not already noticed anyone, it is allowed to post whats.
Steps To Reproduce
Additional Information
Github Commit

Flags
Regression Version
Affected Sets / Systems	kof2001

Attached Files	7.zip (24,742 bytes) Oct 21, 2012, 20:52 Uploaded by Tafoid kof2001 (-state 7) about 8 seconds before game resets.

No.09048 NekoEd Senior Tester Oct 21, 2012, 16:30	If you could please provide a savestate at the beginning of the credits or an INP file to play through the game (both would be even better) that would help us to verify this issue.
No.09049 Tafoid Administrator Oct 21, 2012, 20:48 edited on: Oct 21, 2012, 20:54	I confirmed using a current SVN of 0.147u1, using kof2001 parent, but I'm sure it'll load in 0.147 without issues (-state 7) save state attached to report. The "BACK DESIGNER" is titled but never shown as the game resets.
No.09050 NekoEd Senior Tester Oct 21, 2012, 21:11	It loads and works as intended in 0.147u1 - Watchdog reset at the "BACK DESIGNER" credit.
No.09053 Reyn Tester Oct 27, 2012, 07:58 edited on: Oct 27, 2012, 07:58	Run a test on real HW MV2F with original cart (set kof2001 in current MAME): There is NO reset at "BACK DESIGNER" credit in staff roll. It is an emulation bug. For proof see uploaded vid here: http://www.sendspace.com/file/a3tfdk
No.09055 Haze Senior Tester Oct 27, 2012, 09:43	using what bios? (it could be important)
No.09056 Reyn Tester Oct 27, 2012, 10:32 edited on: Oct 27, 2012, 10:45	@Haze: Universe Bios 3.0 (i have a registred version with S/N) If you want to test, Razoola made version 3.0 public some time ago, could be added to MAME? http://unibios.free.fr/download.html
No.09058 EVA7GDB Tester Oct 28, 2012, 09:50	Please verify various everyone, in this matter, Thank you. Also, because my English is insufficient, a little ease not know, also, may not be fit well with the flow here, s I is due For me, a little older version of MAME32PLUS, that of the time of switching NEODRV.C, MAME you are able to check the symptoms, 0.147 of MAME32FXIn, I have confirmed. We will use a 32bit version of MAME also either, not in UNIVERSE, using a common type of BIOS, BIOS of NEOGEO are used for emulation is the state of the region JAPANESE.
No.09073 Fujix Administrator Nov 3, 2012, 11:52	did some cleanups of the description text. EVA7GDB, we use only official MAME build to test. Any result using derivatives such like MAMEPlus or FX can't prove anything here.
No.09075 NekoEd Senior Tester Nov 3, 2012, 21:01	Do we accept reports/notes using UNIVERSE BIOS? I don't recall, but I think we disallow that. Twisty and I both confirmed it with the default BIOS from the savestate Twisty provided. (Though I believe Twisty actually played through it, I just checked the savestate.)
No.09076 Tafoid Administrator Nov 3, 2012, 21:58	Twisty is not Tafoid.. but yeah - it's confirmed :)
No.10371 hap Developer Mar 18, 2014, 23:06	please try again with svn #28680 or newer
No.10374 Reyn Tester Mar 20, 2014, 06:51 edited on: Mar 20, 2014, 06:54	Tested with set kof2001 (svn used where you added simple waitstates). Still here (used Universe Bios 3.1). The recent fixes by you and awj also have fixed an issue with this: http://www.pouet.net/prod.php?which=59253 (After 3+ minutes the sound went completely silent - Last tested with pre 0.150 - now working fine; tested 15+ minutes). THANK YOU!
No.10375 AWJ Developer Mar 20, 2014, 15:53	If it's still resetting, try r28756.
No.10376 Reyn Tester Mar 20, 2014, 16:38	Applied your change from r28756 and recompiled. No change still resetting. Will do clean recompile and test again to be sure
No.10377 Reyn Tester Mar 20, 2014, 17:00 edited on: Mar 20, 2014, 17:01	Clean recompile - same result: After 'Back Designer' system reset Tested with both kof2001 and kof2001h set.
No.13946 Fujix Administrator Jun 25, 2017, 19:19	Reconfirmed in 0.186.
No.22413 peterferrie Developer 13 days ago	interrupt line 2 gets asserted but never clears, leading to infinite interrupt dispatch.