mirror of
https://github.com/ZDoom/qzdoom.git
synced 2024-11-27 22:42:57 +00:00
- Fixed: Some OPL state needs to be restored before calculating rhythm. Also,
since only the rhythm section uses the RNG, it doesn't need to be advanced for the normal voice processing. SVN r1371 (trunk)
This commit is contained in:
parent
f269af052c
commit
8f26956310
2 changed files with 13 additions and 17 deletions
|
@ -1,3 +1,8 @@
|
|||
January 26, 2009
|
||||
- Fixed: Some OPL state needs to be restored before calculating rhythm. Also,
|
||||
since only the rhythm section uses the RNG, it doesn't need to be advanced
|
||||
for the normal voice processing.
|
||||
|
||||
January 25, 2009 (Changes by Graf Zahl)
|
||||
- fixed: AFastProjectile::Tick must call Effect only 8 times per tic, regardless
|
||||
of the amount of steps taken.
|
||||
|
|
|
@ -1,10 +1,11 @@
|
|||
/*
|
||||
|
||||
This file is based on fmopl.c from MAME 0.95. The non-YM3816 parts have been
|
||||
ripped out in the interest of trying to make this a bit faster, since Doom
|
||||
music doesn't need them. I also made it render the sound a voice at a time
|
||||
instead of a sample at a time, so unused voices don't waste time being
|
||||
calculated.
|
||||
ripped out in the interest of making this simpler, since Doom music doesn't
|
||||
need them. I also made it render the sound a voice at a time instead of a
|
||||
sample at a time, so unused voices don't waste time being calculated. If all
|
||||
voices are playing, it's not much difference, but it does offer a big
|
||||
improvement when only a few voices are playing.
|
||||
|
||||
Here is the appropriate section from mame.txt:
|
||||
|
||||
|
@ -1919,41 +1920,32 @@ void YM3812UpdateOne(void *chip, float *buffer, int length)
|
|||
UINT32 eg_timer_bak = OPL->eg_timer;
|
||||
UINT32 eg_cnt_bak = OPL->eg_cnt;
|
||||
|
||||
UINT32 noise_p_bak = OPL->noise_p;
|
||||
UINT32 noise_rng_bak = OPL->noise_rng;
|
||||
|
||||
UINT32 lfo_am_cnt_out = lfo_am_cnt_bak;
|
||||
UINT32 eg_timer_out = eg_timer_bak;
|
||||
UINT32 eg_cnt_out = eg_cnt_bak;
|
||||
|
||||
UINT32 noise_p_out = noise_p_bak;
|
||||
UINT32 noise_rng_out = noise_rng_bak;
|
||||
|
||||
for (i = 0; i <= (rhythm ? 5 : 8); ++i)
|
||||
{
|
||||
OPL->lfo_am_cnt = lfo_am_cnt_bak;
|
||||
OPL->eg_timer = eg_timer_bak;
|
||||
OPL->eg_cnt = eg_cnt_bak;
|
||||
OPL->noise_p = noise_p_bak;
|
||||
OPL->noise_rng = noise_rng_bak;
|
||||
if (CalcVoice (OPL, i, buffer, length))
|
||||
{
|
||||
lfo_am_cnt_out = OPL->lfo_am_cnt;
|
||||
eg_timer_out = OPL->eg_timer;
|
||||
eg_cnt_out = OPL->eg_cnt;
|
||||
noise_p_out = OPL->noise_p;
|
||||
noise_rng_out = OPL->noise_rng;
|
||||
}
|
||||
}
|
||||
|
||||
OPL->lfo_am_cnt = lfo_am_cnt_out;
|
||||
OPL->eg_timer = eg_timer_out;
|
||||
OPL->eg_cnt = eg_cnt_out;
|
||||
OPL->noise_p = noise_p_out;
|
||||
OPL->noise_rng = noise_rng_out;
|
||||
|
||||
if (rhythm) /* Rhythm part */
|
||||
{
|
||||
OPL->lfo_am_cnt = lfo_am_cnt_bak;
|
||||
OPL->eg_timer = eg_timer_bak;
|
||||
OPL->eg_cnt = eg_cnt_bak;
|
||||
CalcRhythm (OPL, buffer, length);
|
||||
}
|
||||
}
|
||||
|
@ -1979,7 +1971,6 @@ static bool CalcVoice (FM_OPL *OPL, int voice, float *buffer, int length)
|
|||
OPL_CALC_CH(CH, buffer + i);
|
||||
|
||||
advance(OPL, voice, voice);
|
||||
advance_noise(OPL);
|
||||
}
|
||||
return true;
|
||||
}
|
||||
|
|
Loading…
Reference in a new issue