off-topic, but I saw Joe Magiera dumped some Zoar PALs.
ideally the program ROMs need verifying on that game too, as MAME requires a single byte patch to avoid the game just locking up rather than showing attract mode.
it could be an undocumented CPU feature, or it could be a bad dump, it seems some of the chips Data East used are a little prone to bitrot based on other dumps made of other games in that driver (we've seen Disco #1 and Burger Times with bitrot before too when compared with the good ones we have)
I'm 99% sure the ROM based version of Treasure Island is a bad dump too, the patches you need to make it work (which nobody has fully done) don't even make sense however you look at the encryption, they don't even fall on addresses that should be affected by it in the ways we see. Of course that one was dumped probably 15 years ago and I can't even remember who dumped it.
> > Ordyne dumped a Cycraft Club Kart Motion PC > > I wonder if any of the other PC control programs will show up, they're an interesting > curiosity, and don't seem well documented. >
yeah, Tafoid let me know, means we can update the comment in here at least to confirm it isn't a bad dump. Knowing the dump is fine gives more incentive to find the real problem at least, thanks to Joe for checking. This comment has been in MAME for a long time.
/* 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. */ memset(&rom[0xd50a],0xea,8);
m_audio_nmi_enable_type = AUDIO_ENABLE_AY8910; }
The Treasure Island situation is slightly different as we know what the opcodes should be already from the cassette version they can't be NOPs. In the case of Zoar maybe they're using a CPU type where this undocumented instruction does simply act as a NOP.