virtualenv/FFmpeg
1
0
Fork 0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2025-01-08 13:22:53 +02:00
Code Issues Releases Activity

Use the new idct functions (except chroma as it was slower in benchmarks)

Browse Source 
cathedral +0.5% speed
aladin +0.6% speed [note aladin has been cat-ed 10 times to reduce the influence
of init time]
Speedup also verified via START/STOP_TIMER (difference was very significant
for the changed parts)

Originally committed as revision 16207 to svn://svn.ffmpeg.org/ffmpeg/trunk
This commit is contained in:
Michael Niedermayer 2008-12-18 02:53:18 +00:00
parent ac22385931
commit 2fd1f0e026
1 changed files with 14 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
libavcodec/h264.c
View File

@ -2515,27 +2515,34 @@ static av_always_inline void hl_decode_mb_internal(H264Context *h, int simple){
if(!IS_INTRA4x4(mb_type)){
if(is_h264){
if(IS_INTRA16x16(mb_type)){
if(transform_bypass && h->sps.profile_idc==244 && (h->intra16x16_pred_mode==VERT_PRED8x8 || h->intra16x16_pred_mode==HOR_PRED8x8)){
if(transform_bypass){
if(h->sps.profile_idc==244 && (h->intra16x16_pred_mode==VERT_PRED8x8 || h->intra16x16_pred_mode==HOR_PRED8x8)){
h->hpc.pred16x16_add[h->intra16x16_pred_mode](dest_y, block_offset, h->mb, linesize);
}else{
for(i=0; i<16; i++){
if(h->non_zero_count_cache[ scan8[i] ])
if(h->non_zero_count_cache[ scan8[i] ] || h->mb[i*16])
idct_add (dest_y + block_offset[i], h->mb + i*16, linesize);
else if(h->mb[i*16])
idct_dc_add(dest_y + block_offset[i], h->mb + i*16, linesize);
}
}
}else{
s->dsp.h264_idct_add16intra(dest_y, block_offset, h->mb, linesize, h->non_zero_count_cache);
}
}else if(h->cbp&15){
if(transform_bypass){
const int di = IS_8x8DCT(mb_type) ? 4 : 1;
for(i=0; i<16; i+=di){
int nnz = h->non_zero_count_cache[ scan8[i] ];
if(nnz){
if(nnz==1 && h->mb[i*16])
idct_dc_add(dest_y + block_offset[i], h->mb + i*16, linesize);
else
idct_add(dest_y + block_offset[i], h->mb + i*16, linesize);
}
}
}else{
if(IS_8x8DCT(mb_type)){
s->dsp.h264_idct8_add4(dest_y, block_offset, h->mb, linesize, h->non_zero_count_cache);
}else{
s->dsp.h264_idct_add16(dest_y, block_offset, h->mb, linesize, h->non_zero_count_cache);
}
}
}
}else{
for(i=0; i<16; i++){