RepoMirrors
/
mpv
mirror of https://github.com/mpv-player/mpv
1
0
Fork 0
Code Issues Releases Wiki Activity

100l, *ppc++ was supposed to be replaced by ppc[i] in r25100, but that is not any faster. 

Just removing the += s->samples_overlap - s->num_channels; still provides a
ca. 20% speedup on x86 (AThlon X2 64) with gcc 3.4 (compiler stupidity?)


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@25102 b3059339-0415-0410-9bf9-f77b7e298cf2
This commit is contained in:
reimar 2007-11-18 18:33:45 +00:00
parent 66c4759992
commit 13a205e075
1 changed files with 0 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
libaf/af_scaletempo.c
View File

@ -171,8 +171,6 @@ static int best_overlap_offset_s16(af_scaletempo_t* s)
int32_t corr = 0;
int16_t* ps = search_start;
ppc = s->buf_pre_corr;
ppc += s->samples_overlap - s->num_channels;
ps += s->samples_overlap - s->num_channels;
i = -(s->samples_overlap - s->num_channels);
do {
corr += ( *ppc++ * *ps++ ) >> s->shift_corr;