I'll be using this opening post to keep track of current progress on console verifications and other testing.
Currently testing:
- HDMA edge cases in single speed mode
_____________________________________
Dev Build:
https://ci.appveyor.com/project/zeromus/bizhawk-udexo/build/artifacts
Repo for input timestamp files:
https://github.com/alyosha-tas/GBI_timestamps
Dump script for 2.6.1 and older:
http://tasvideos.org/userfiles/info/68761073876269603
Dump script for 2.6.2 and up:
http://tasvideos.org/userfiles/info/71650785644185599
NOTE: use the setting 'TotalExecutedCycles Return Value' in GB->Settings in BizHawk and set it to GBI for double speed mode games
_____________________________________
Several games use uninitialized RAM in a way that causes desyncs on real hardware, below is a list of some examples I found:
Race Days (Select Dirty Racing) : This game checks a memory address to see if a block needs to be initialized. It does this many times in different locations. The same movie will desync in different ways each playthrough on console. Ex:
02EB: EA 24 C2 LD (#C224h),A A:00 F:90 B:A8 C:FC D:02 E:00 H:A0 L:00 SP:DFFA Cy:61216123 LY:128 ZnhCie
Super R.C. Pro Am : This game checks 4 locations in memory before initializing them, these could be related to linking but I'm not really sure. There seems to be a high probability of these cases passing on a real console in my testing:
FFCD > 0x02 (FFCD unwritten)
FFCC != 0xDE (checks 3 times, FFCC unwritten)
FFCC != 0xDE (checks 3 times, FFCC unwritten)
FFC9 != FFC8 (FFC9 unwritten, FFC8 written 1)
Pokemon Crystal: When loading up the credits sequence it checks player state in the wrong bank, this doesn't seem to have any real impact, but would possibly cause a desync if you wanted to continue a movie past the credits:
4384: FA 5D D9 LD A,(#D95Dh) A:00 F:80 B:C2 C:91 D:0A E:00 H:C2 L:91 SP:C0B9 Cy:323183331 LY:149 ZnhciE
4387: FE 01 CP #01h A:00 F:80 B:C2 C:91 D:0A E:00 H:C2 L:91 SP:C0B9 Cy:323183347 LY:149 ZnhciE
Daiku no Gen-san - Ghost Building Company (Japan): uses uninitialized RAM at the start of level 2, this causes a desync towards the end of the level. It looks like all it does is use these values for starting xscroll possibly for the clouds.
Battletoads: Reads from HRAM at the start of level one to set some values in WRAM. Doesn't effect the TAS but might effect casual play, not sure
09F9: F0 C2 LDH A,(#FFC2h) A:24 F:00 B:00 C:20 D:C9 E:18 H:9A L:C3 SP:CAFC Cy:105661311 LY:0 znhcie
Another milestone is finally reached wiith Scrapyard Dog's sewer level finally being correctly emulated. This level was another main driver for me to create this emulator after seeing Mothrayas' TAS of this game contain a pretty bad graphical glitch in this level.
What this glitch comes down to is DMA timing around what is called 'holey DMA.' Basically DMA can be told to ignore certain address blocks, and when it does so there is no DMA read penalty for addresses in that block.
Except, there is a penalty for the first read in order to identify that block in the first place. Getting this (and pretty much everything else DMA related) right is what fixes scrapyard dog.
EDIT: @MESHUGGAH: I unified the controller definitions between EMU7800 and A7800Hawk (at least for 2 button and joystick controllers) so you should be able to compare movies now I think.
I rewrote the Maria renderer in A7800Hawk to make it pixel accurate. Now the one stray line in Kung Fu Master is fixed.
This also brought with it a dramatic simplification of the code, which makes it much more managable.
As accuracy picks up the core is starting to slow down a bit though. I used to be stable 240 FPS at 4x speed, but now I'm maxed out at 200 or so. I'll try to recover a bit of that slowdown once all the bugs are worked out.