03722: vasara2a: Unhandled OpCode found : 14 at 00000021

ID	Category [?]	Severity [?]	Reproducibility	Date Submitted	Last Update
03722	Crash/Freeze	Critical (emulator)	Random	Feb 9, 2010, 00:55	Nov 13, 2016, 12:23

Tester	iveinsomnia	View Status	Public	Platform
Assigned To		Resolution	Unable to reproduce	OS
Status [?]	Closed			Driver	seta/ssv.cpp
Version	0.136u2	Fixed in Version		Build	Normal
		Fixed in Git Commit		Github Pull Request #

Summary	03722: vasara2a: Unhandled OpCode found : 14 at 00000021
Description	Vasara 2 (set 2) Crash with output: "Unhandled OpCode found : 14 at 00000021"
Steps To Reproduce	Difficulty: Difficult Ship: The green one Cheat: no Configuration: default game & hardware. Level: 1 - in the center of the four shooting tower in my screen (few seconds after the start) kill the four tower around the center and hit the center with the ship. I've try like 12 times and wasn't able to reproduce.
Additional Information	already done this level before without problems.
Github Commit

Flags
Regression Version
Affected Sets / Systems	vasara2a

Attached Files	vasara2a.png (101,018 bytes) Feb 9, 2010, 11:54

No.05669 Haze Senior Tester Feb 9, 2010, 01:22 edited on: Feb 9, 2010, 01:23	one of the Vasara 2 sets is thought to be a bad dump, but it wasn't certain which one. If this only happens with the vasara2a set, then it seems likely that it's one one which is bad, and it should be removed. Having a reproducable inp test case to verify this and compare it with the other set would help a lot tho.. It could just be a game bug which can happen in either set. (the fatalerror may cause recording issues tho)
No.05672 Fujix Administrator Feb 9, 2010, 10:51 edited on: Feb 9, 2010, 10:53	We need details where you experienced the crash to re-test the report, Easy or Difficult, which tower you shot, whether you used cheat or not, etc. And could you post a screenshot where the crash occurred?
No.05674 iveinsomnia Tester Feb 9, 2010, 12:02	Screen added and repro rewrited. (sorry this is my first report in here, and my first "real" mame crash, also sorry for my english)
No.05675 Fujix Administrator Feb 9, 2010, 17:20	Thanks for new info. There are a lot of non-native English speakers here, me neither, you don't have to worry about it at all :-)
No.05691 Fujix Administrator Feb 10, 2010, 13:42	I tried it more than 10 times but no repro. I set this report "Acknowledged" status.
No.05692 Haze Senior Tester Feb 10, 2010, 13:52	All I remember is that Brian Troha who dumped one of the sets said at one point there was a software glitch in his new programmer, which caused everything he'd dumped to have 2 bytes wrong, early on in the ROM. At the time I couldn't remember which set was the one he dumped tho. A binary compare of them, and analysis of the code (if it's a code region) might help you, although I can't remember if this was already done, and it was found to be data, hence it hard to draw a conclusion. You might want to raise it with him?
No.05706 Layne Tester Feb 11, 2010, 19:09	Talking with Brian about this bug, he said: "Haze is not completly correct here. I dumped the original Vasara 2 set, the second set came from an alt source. The issue he refers to did happen with Eagle Shot Golf and the bad dump was removed. The program cause the actual rom to have a 0x00 at bytes 0x200 & 0x201... not the case with Vasara 2."
No.05707 Haze Senior Tester Feb 11, 2010, 20:17 edited on: Feb 11, 2010, 20:18	Hmm.. Then why do I have an old note saved saying that the same problem applies to the Vasara2 set Brian dumped, and that I should follow up his email.. I wonder where that came from? Maybe I just typed the wrong set name when I made it?
No.05708 Layne Tester Feb 11, 2010, 22:05	Brian added another comment: "Load up a hex editor and check each Vasara 2 program and see if both bytes at 0x200 & 0x201 are 0x00 in each rom... I don't have a 64bit hex editor handy to check myself. Anyways the issue was only with TI 27C040 roms, as far as I know, no other roms were affected. If I read the same TI rom, but used the AMD manufacturer's settings, it read right."
No.13357 Fujix Administrator Nov 13, 2016, 12:23	Cleaning up old reports. No repro in 0.179. Closed the report.