Max/SameBoy
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Notes about the DMG wave-ram glitch

Browse Source
This commit is contained in:
Lior Halphon 2018-05-15 23:02:07 +03:00
parent 7df571d42f
commit 562b43a7c5
1 changed files with 7 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
Core/apu.c
View File

@ -691,8 +691,13 @@ void GB_apu_write(GB_gameboy_t *gb, uint8_t reg, uint8_t value)
gb->apu.wave_channel.enable) { gb->apu.wave_channel.enable) {
unsigned offset = ((gb->apu.wave_channel.current_sample_index + 1) >> 1) & 0xF; unsigned offset = ((gb->apu.wave_channel.current_sample_index + 1) >> 1) & 0xF;
/* On SGB2 (and probably SGB1 and MGB as well) this behavior is not accurate, /* This glitch varies between models and even specific instances:
however these systems are not currently emulated. */ DMG-B: Most of them behave as emulated. A few behave differently.
SGB: As far as I know, all tested instances behave as emulated.
MGB, SGB2: Most instances behave non-deterministically, a few behave as emulated.
Additionally, I believe DMGs, including those we behave differently than emulated,
are all deterministic. */
if (offset < 4) { if (offset < 4) {
gb->io_registers[GB_IO_WAV_START] = gb->io_registers[GB_IO_WAV_START + offset]; gb->io_registers[GB_IO_WAV_START] = gb->io_registers[GB_IO_WAV_START + offset];
gb->apu.wave_channel.wave_form[0] = gb->apu.wave_channel.wave_form[offset / 2]; gb->apu.wave_channel.wave_form[0] = gb->apu.wave_channel.wave_form[offset / 2];