| 2007-09-24 |
Loren Merritt | port sad_*_x3_sse2 to x86_64 |
tree | commitdiff |
| 2007-09-21 |
Loren Merritt | faster 4x4 sad |
tree | commitdiff |
| 2007-04-10 |
Loren Merritt | tweak ssse3 quant |
tree | commitdiff |
| 2007-04-05 |
Loren Merritt | 32bit version of ssse3 satd. |
tree | commitdiff |
| 2007-04-04 |
Loren Merritt | 2x faster quant. 2% overall. |
tree | commitdiff |
| 2007-04-04 |
Loren Merritt | cosmetics in asm macros |
tree | commitdiff |
| 2007-04-02 |
Loren Merritt | in hpel search, merge two 16x16 mc calls into one 16x17... |
tree | commitdiff |
| 2007-03-14 |
Loren Merritt | ssse3 detection. x86_64 ssse3 satd and quant. |
tree | commitdiff |
| 2006-12-15 |
Loren Merritt | New threading method: |
tree | commitdiff |
| 2006-11-06 |
Loren Merritt | 10% faster deblock mmx functions. ported from ffmpeg. |
tree | commitdiff |
| 2006-10-10 |
Loren Merritt | prefetch pixels for motion compensation and deblocking. |
tree | commitdiff |
| 2006-10-09 |
Loren Merritt | slightly faster chroma_mc_mmx |
tree | commitdiff |
| 2006-10-09 |
Loren Merritt | missing emms in plane_copy_mmx |
tree | commitdiff |
| 2006-10-06 |
Loren Merritt | merge center_filter_mmx with horizontal_filter_mmx |
tree | commitdiff |
| 2006-10-06 |
Loren Merritt | 1.5x faster center_filter_mmx (amd64) |
tree | commitdiff |
| 2006-10-06 |
Loren Merritt | mmx/prefetch implementation of plane_copy |
tree | commitdiff |
| 2006-10-03 |
Loren Merritt | simplify satd_sse2 |
tree | commitdiff |
| 2006-10-01 |
Loren Merritt | tweak motion compensation amd64 asm. 0.3% overall speedup. |
tree | commitdiff |
| 2006-10-01 |
Loren Merritt | move zigzag scan functions to dsp function pointers. |
tree | commitdiff |
| 2006-09-12 |
Loren Merritt | faster ESA |
tree | commitdiff |
| 2006-08-21 |
Loren Merritt | ssim broke on amd64 w/ pic. |
tree | commitdiff |
| 2006-08-17 |
Loren Merritt | SSIM computation. (default on, disable by --no-ssim) |
tree | commitdiff |
| 2006-07-16 |
Loren Merritt | slightly faster mmx dct |
tree | commitdiff |
| 2006-07-08 |
Loren Merritt | mc_chroma width2 mmx |
tree | commitdiff |
| 2006-05-10 |
Loren Merritt | faster intra search: filter i8x8 edges only once, and... |
tree | commitdiff |
| 2006-05-09 |
Loren Merritt | faster intra search: some prediction modes don't have... |
tree | commitdiff |
| 2006-04-24 |
Loren Merritt | mmx implementation of x264_pixel_sa8d |
tree | commitdiff |
| 2006-04-12 |
Loren Merritt | 3% faster satd_mmx |
tree | commitdiff |
| 2006-04-12 |
Loren Merritt | cosmetics in sad/ssd/satd mmx |
tree | commitdiff |
| 2006-04-11 |
Loren Merritt | yasm noexec stack |
tree | commitdiff |
| 2006-04-10 |
Loren Merritt | interleave multiple calls to SAD. |
tree | commitdiff |
| 2006-03-24 |
Loren Merritt | use FDEC_STRIDE instead of a parameter in mmx dct |
tree | commitdiff |
| 2006-03-17 |
Loren Merritt | some mmxext functions really only required mmx. |
tree | commitdiff |
| 2006-03-09 |
Loren Merritt | keep transposed dct coefs. ~1% overall speedup. |
tree | commitdiff |
| 2006-03-05 |
Loren Merritt | more amd64 mmx intra prediction |
tree | commitdiff |
| 2006-02-27 |
Loren Merritt | copy current macroblock to a smaller buffer, to improve... |
tree | commitdiff |
| 2006-02-27 |
Loren Merritt | h->mc.copy() |
tree | commitdiff |
| 2006-02-10 |
Loren Merritt | amd64 mmx for some intra pred functions |
tree | commitdiff |
| 2006-02-10 |
Loren Merritt | dequant_mmx made incorrect assumptions about extreme... |
tree | commitdiff |
| 2006-02-08 |
Sam Hocevar | * Additional fixes to the PIC versions of assembly... |
tree | commitdiff |
| 2006-02-08 |
Loren Merritt | cosmetics in mc_chroma |
tree | commitdiff |
| 2006-02-06 |
Loren Merritt | x86 mmx for some intra pred functions |
tree | commitdiff |
| 2006-02-06 |
Loren Merritt | cosmetics: reorganize intra prediction dsp |
tree | commitdiff |
| 2006-01-09 |
Loren Merritt | early termination within large SADs. ~1% faster UMH... |
tree | commitdiff |
| 2005-11-11 |
Sam Hocevar | * common/amd64/quant-a.asm: added missing GLOBAL... |
tree | commitdiff |
| 2005-11-06 |
Loren Merritt | mmx dequant. up to 3% speedup w/ RD. |
tree | commitdiff |
| 2005-11-03 |
Loren Merritt | amd64 sse2 8x8dct. 1.45x faster than mmx. |
tree | commitdiff |
| 2005-10-26 |
Loren Merritt | mmx deblocking. |
tree | commitdiff |
| 2005-10-26 |
Loren Merritt | r336 broke amd64 x264_pixel_sad_16x16_sse2 (though... |
tree | commitdiff |
| 2005-10-25 |
Loren Merritt | Windows 64bit asm. |
tree | commitdiff |
| 2005-10-10 |
Loren Merritt | cosmetics: remove declarations of nonexistent asm functions |
tree | commitdiff |
| 2005-10-04 |
Sam Hocevar | * common/amd64/mc-a.asm: use RIP-relative addressing... |
tree | commitdiff |
| 2005-10-01 |
Loren Merritt | mmx avg (already existed by not used for bipred) |
tree | commitdiff |
| 2005-10-01 |
Loren Merritt | cosmetics: move avg function ptrs from pixf to mc. |
tree | commitdiff |
| 2005-09-25 |
Sam Hocevar | * common/amd64/*.asm: use RIP-related addressing... |
tree | commitdiff |
| 2005-09-25 |
Sam Hocevar | * common/amd64/mc-a.asm: removed useless global variables |
tree | commitdiff |
| 2005-09-24 |
Loren Merritt | faster mmx quant 15bit, and add 16bit version. total... |
tree | commitdiff |
| 2005-09-24 |
Loren Merritt | faster mmx satd. *x16: 20%, *x8: 10%, total: 2-4%. |
tree | commitdiff |
| 2005-09-14 |
Loren Merritt | MMX quantization functions, and optimization of the... |
tree | commitdiff |
| 2005-09-10 |
Loren Merritt | SSE2 pixel comparison functions |
tree | commitdiff |
| 2005-08-29 |
Loren Merritt | mmx 8x8 dct. |
tree | commitdiff |
| 2005-07-13 |
Loren Merritt | faster SAD_INC_2x16P for amd64. |
tree | commitdiff |
| 2005-06-03 |
Loren Merritt | 8x8 transform and 8x8 intra prediction. |
tree | commitdiff |
| 2005-05-11 |
Loren Merritt | port recent MC asm changes to amd64. |
tree | commitdiff |
| 2005-04-25 |
Loren Merritt | cosmetics. |
tree | commitdiff |
| 2005-04-19 |
Loren Merritt | amd64 asm patch, part2. |
tree | commitdiff |
| 2005-04-19 |
Loren Merritt | amd64 asm patch, part1. |
tree | commitdiff |
|
projects / x264 / history