Emulator Cheats

This forum is for discussion about the cheat engine in MAME, not the PC program with a similar name. To answer your question, though, the "set direction flag" instruction only modifies registers, it can't modify memory. Something is wrong with your analysis of the code, or CE is breakpoint...

Sorry about this - I noticed he had been working on this for a while, and assumed it was happening with the knowledge and support of you guys, otherwise I would have spoken up on the list. I didn't look closely at the change enough. Some changes he has sent in are good, mainly converting function/va...

I haven't had a chance to test this yet, but at around line 9179, change: static void InvalidateEntireRegion(SearchInfo * search, SearchRegion * region) { memset(region->status, 0, region->length); search->numResults -= region->numResults; region->numResults = 0; } to static void InvalidateEntireReg...

Gah, sorry about that again. Edited original post.

Heh, sorry about that. Was posting from work and couldn't check the code. There's a typo in ReadData that makes it applies random byte-swapping when reading an indirect value. To fix, replace: case kLocation_IndirectIndexed: { UINT32 address; INT32 offset = action->extendData; UINT8 cpu = (parameter...

Both the relative offset and the bitmask use the extend data field (the last numeric field in the cheat), so only one can work at a time.

Aaron said he's fixed these issues in u5, let me know if there's anything still unresolved.

After getting familiar with the new rendering code, it looks like it's still very much under development. I'll take a look again once things have calmed down; many things are changing with each 'u' patch. Thanks for the bug reports, though, I will make sure these are fixed once the render core is mo...

Replace the beginning of LoadCheatDatabase (line 8672) with: static void LoadCheatDatabase(void) { char buf[4096]; const char * inTraverse; char * outTraverse; char * mainTraverse; int first = 1; char data; inTraverse = options_get_string("cheat_file", 0); if(!inTraverse) inTraverse = &quo...

Confirmed, fix coming in a few minutes.

It's a very powerful disassembler with good analysis tools, but out of the box it's only a debugger for x86. I'm not sure how extendable the debugger is; if it's general-case enough someone could probably write a debugger client for MAME that could talk with IDA via a plugin or the remote debugger i...

IDA pro (the legit Professional version at least) has support for tons of different CPUs; it can be very useful for this sort of thing as a companion to the MAME debugger.

Yeah, I haven't had a chance to take a look at the code yet, but it's almost guaranteed to be caused by the video rewrite. I'll take a look tomorrow.

Not currently.

Cool, I'll add the change. Sorry I didn't think about that when I fixed the first issue.

Hmm. I was actually thinking about making each cheat in a linked cheat an instruction in the program; it'd be by far easiest that way as most of the code for that exists already. Will think of some alternative. Again, the major weakness is branching, unless you are thinking of, say, "jump over ...

I could see something like this being caused when Aaron moved the cheat application timer. It doesn't look like it currently is called at the beginning of the first frame, and the CPU may pass the code you're patching before the game reaches the end of the frame and the cheat is applied. Try changin...

Thanks for the list, but all of these problems are caused by use of the new decrypted CPU memory system (memory_set_decrypted_region and related functions). Right now there is no API set up for writing to these regions, so there isn't any way for the cheat engine to modify the data there. I plan to ...

Ah. That's a good suggestion, I'll add it to the list. System 16 games have a completely dynamic memory map that can for example move the RAM around during runtime, so it'd be impossible for the cheat engine to autodetect things like that without per-driver code.

Sorry for the delay on these posts, couldn't get to anything hosted at ZTNet over the weekend. It's an interesting idea, but not really that much less code than the original implementation. I'd much rather just set up something with conditional branching than do a partial implementation or something...