02738: zoar: The rom marked as bad dump is not bad.

ID	Category [?]	Severity [?]	Reproducibility	Date Submitted	Last Update
02738	Documentation	Minor	N/A	Dec 12, 2008, 23:50	Dec 27, 2008, 02:50

Tester	Smitdogg	View Status	Public	Platform
Assigned To	aaron	Resolution	Fixed	OS
Status [?]	Resolved			Driver	dataeast/btime.cpp
Version	0.128u5	Fixed in Version	0.129	Build
		Fixed in Git Commit		Github Pull Request #

Summary	02738: zoar: The rom marked as bad dump is not bad.
Description	The rom marked as bad dump is not bad. I have the pcb and I dumped that rom many times and I get the same crc every time.
Steps To Reproduce
Additional Information
Github Commit

Flags	Verified with Original
Regression Version
Affected Sets / Systems	zoar

Attached Files

No.03291 Tafoid Administrator Dec 13, 2008, 00:16	From Source: /* At location 0xD50A is what looks like an undocumented opcode. I tried implementing it given what opcode 0x23 should do, but it still didn't work in demo mode. So this could be another protection or a bad ROM read. I'm NOPing it out for now. */ Given your board isn't the original board used for the set and the roms are the same for that main cpu rom that has been around a long time and the repeated attempts - I'd lean on it being a opcode issue rather than a BAD_DUMP issue.
No.03296 Haze Senior Tester Dec 13, 2008, 11:58	yeah, probably.. there are similar issues with treasure island too. It might be a subtle extra to the protection / encryption.. does this board (if it's an original you have) use one of those big data east CPU blocks too?
No.03300 Smitdogg Senior Tester Dec 13, 2008, 16:23	Yeah, it's on there, says CPU-7 on it. I'm sending pics to Guru later and I will eventually get the pals dumped and I'm trying to get a rig set up where I can make pcb videos, so I'll have more info later.
No.03301 Kale Developer Dec 13, 2008, 20:45 edited on: Dec 13, 2008, 20:47	Auto-citation (Treasure Island): "At location 0xa2b6 there's a strange RLA followed by a BPL that reads from an unmapped area that causes the game to fail in several circumstances.On the Cassette version the RLA (33) is in reality a BIT (24),so I'm guessing that there's something wrong going on in the encryption scheme." For sure it's not a ROM issue (unlikely it's a bitrot),more likely to be an encryption issue. Probably the DECO-7 can do more than just shifting the opcodes bits on write types,and it's likely that it should be properly converted & handled to a cpu-sub-type of M6502 instead of just change the opcodes at run time...
No.03310 Smitdogg Senior Tester Dec 15, 2008, 20:22	I'm sending the board to Guru so he can "attack the epoxy block" and take measurements etc.
No.03314 Haze Senior Tester Dec 15, 2008, 22:10	makes sense I guess, although I think there are bootlegs which reproduce the Burger Time encryption with simply logic (although this might be something more, and I guess it would be interesting to see exactly which CPU is inside the block, as it could be relying on custom behavior of the chips produced by a certain manufacturer..)
No.03409 aaron Developer Dec 27, 2008, 02:50	Fixed the ROM definition. Other issues are still present, obviously.