2010-01-14 |
Fiona Glaser | Periodic intra refresh |
blob | commitdiff | raw |
2009-12-09 |
Fiona Glaser | qpel RD no longer needs mbcmp_unaligned |
blob | commitdiff | raw | diff to current |
2009-12-05 |
Fiona Glaser | Actually do r1356 |
blob | commitdiff | raw | diff to current |
2009-12-05 |
Fiona Glaser | Significantly faster qpel-RD |
blob | commitdiff | raw | diff to current |
2009-11-12 |
Fiona Glaser | Fix all aliasing violations |
blob | commitdiff | raw | diff to current |
2009-11-09 |
Dylan Yudaken | Weighted P-frame prediction |
blob | commitdiff | raw | diff to current |
2009-10-29 |
Fiona Glaser | Further reduce code size in bime |
blob | commitdiff | raw | diff to current |
2009-10-29 |
Fiona Glaser | Take into account chroma MV offset during interlaced... |
blob | commitdiff | raw | diff to current |
2009-09-14 |
Fiona Glaser | Make MV costs global instead of static |
blob | commitdiff | raw | diff to current |
2009-09-07 |
Fiona Glaser | Improve x264 help |
blob | commitdiff | raw | diff to current |
2009-09-06 |
Fiona Glaser | Factor out a redundant RD call in qpel-RD |
blob | commitdiff | raw | diff to current |
2009-08-31 |
Loren Merritt | Faster me=tesa |
blob | commitdiff | raw | diff to current |
2009-08-23 |
David Conrad | GSOC merge part 2: ARM stack alignment |
blob | commitdiff | raw | diff to current |
2009-07-29 |
Fiona Glaser | Fix another 10L in QPRD |
blob | commitdiff | raw | diff to current |
2009-07-28 |
Loren Merritt | 15% faster refine_bidir_satd, 10% faster refine_bidir_r... |
blob | commitdiff | raw | diff to current |
2009-07-28 |
Fiona Glaser | Faster bidir_rd plus some bugfixes |
blob | commitdiff | raw | diff to current |
2009-07-26 |
Fiona Glaser | Fix a nondeterminism with threads and subme>7 |
blob | commitdiff | raw | diff to current |
2009-07-20 |
Anton Mitrofanov | New AQ algorithm option |
blob | commitdiff | raw | diff to current |
2009-05-27 |
Loren Merritt | a better way to keep track of mv candidates. |
blob | commitdiff | raw | diff to current |
2009-05-27 |
Loren Merritt | reorder some motion estimation patterns. |
blob | commitdiff | raw | diff to current |
2009-05-24 |
Fiona Glaser | Fix 10L in r1155 |
blob | commitdiff | raw | diff to current |
2009-04-17 |
Fiona Glaser | Remove various bits of dead code found by CLANG. |
blob | commitdiff | raw | diff to current |
2008-12-31 |
Steven Walters | factor mallocs out of hpel, ssim, and esa. |
blob | commitdiff | raw | diff to current |
2008-11-11 |
Fiona Glaser | Add subme=0 (fullpel motion estimation only) |
blob | commitdiff | raw | diff to current |
2008-10-22 |
Fiona Glaser | Sub-8x8 Qpel-RD in P-frames |
blob | commitdiff | raw | diff to current |
2008-10-22 |
Fiona Glaser | Faster qpel-RD |
blob | commitdiff | raw | diff to current |
2008-10-02 |
Fiona Glaser | Rework subme system, add RD refinement in B-frames |
blob | commitdiff | raw | diff to current |
2008-09-21 |
Fiona Glaser | Resolve possible crash in bime, improve the fix in... |
blob | commitdiff | raw | diff to current |
2008-09-20 |
Loren Merritt | r980 borked weighted bime |
blob | commitdiff | raw | diff to current |
2008-09-20 |
Fiona Glaser | Merge avg and avg_weight |
blob | commitdiff | raw | diff to current |
2008-09-19 |
Fiona Glaser | Rewrite avg/avg_weight to take two source pointers |
blob | commitdiff | raw | diff to current |
2008-09-05 |
Fiona Glaser | Add sad_aligned for faster subme=1 mbcmp |
blob | commitdiff | raw | diff to current |
2008-08-21 |
Loren Merritt | cosmetics |
blob | commitdiff | raw | diff to current |
2008-08-09 |
Fiona Glaser | Improve subme7 at low QPs and add subme7 support in... |
blob | commitdiff | raw | diff to current |
2008-07-06 |
Fiona Glaser | Various optimizations and cosmetics |
blob | commitdiff | raw | diff to current |
2008-07-04 |
Fiona Glaser | Update file headers throughout x264 |
blob | commitdiff | raw | diff to current |
2008-06-12 |
Fiona Glaser | More tweaks to me.c |
blob | commitdiff | raw | diff to current |
2008-06-11 |
Fiona Glaser | Tweaks and cosmetics in me.c |
blob | commitdiff | raw | diff to current |
2008-06-03 |
Fiona Glaser | memzero_aligned_mmx |
blob | commitdiff | raw | diff to current |
2008-06-03 |
Fiona Glaser | omit redundant mc after non-rdo dct size decision,... |
blob | commitdiff | raw | diff to current |
2008-05-17 |
Fiona Glaser | explicit write combining, because gcc fails at optimizi... |
blob | commitdiff | raw | diff to current |
2008-04-27 |
Loren Merritt | earlier termination in SEA if mvcost exceeds residual |
blob | commitdiff | raw | diff to current |
2008-03-25 |
Loren Merritt | fix an arithmetic overflow that disabled SEA threshold... |
blob | commitdiff | raw | diff to current |
2008-03-22 |
Loren Merritt | cosmetics in DECLARE_ALIGNED |
blob | commitdiff | raw | diff to current |
2008-03-22 |
Loren Merritt | more alignment |
blob | commitdiff | raw | diff to current |
2008-01-27 |
Loren Merritt | remove a useless bit of me=umh (originally copied from... |
blob | commitdiff | raw | diff to current |
2008-01-27 |
Loren Merritt | satd exhaustive motion search (--me tesa) |
blob | commitdiff | raw | diff to current |
2008-01-08 |
Loren Merritt | slightly wrong memory allocation in r717, fixes a poten... |
blob | commitdiff | raw | diff to current |
2008-01-06 |
Loren Merritt | convert absolute difference of sums from mmx to sse2 |
blob | commitdiff | raw | diff to current |
2008-01-06 |
Loren Merritt | round esa range to a multiple of 4 |
blob | commitdiff | raw | diff to current |
2007-12-01 |
Loren Merritt | faster removal of duplicate mv predictors |
blob | commitdiff | raw | diff to current |
2007-12-01 |
Loren Merritt | avoid a division in umh. |
blob | commitdiff | raw | diff to current |
2007-11-19 |
Loren Merritt | cosmetics: reorder mc_luma/mc_chroma/get_ref arguments... |
blob | commitdiff | raw | diff to current |
2007-11-16 |
Loren Merritt | cosmetics: use symbolic constants for frame padding... |
blob | commitdiff | raw | diff to current |
2007-11-01 |
Loren Merritt | fix a possible nondeterminism with me=umh + threads. |
blob | commitdiff | raw | diff to current |
2007-10-29 |
Loren Merritt | use hex instead of dia for rdo mv refinement. ~0.5... |
blob | commitdiff | raw | diff to current |
2007-07-17 |
Loren Merritt | move os/compiler specific defines to their own header |
blob | commitdiff | raw | diff to current |
2007-07-06 |
Loren Merritt | cosmetics |
blob | commitdiff | raw | diff to current |
2007-05-26 |
Alex Izvorski | replace alloca with malloc everywhere. per manpage... |
blob | commitdiff | raw | diff to current |
2007-04-02 |
Loren Merritt | in hpel search, merge two 16x16 mc calls into one 16x17... |
blob | commitdiff | raw | diff to current |
2006-12-15 |
Loren Merritt | New threading method: |
blob | commitdiff | raw | diff to current |
2006-09-24 |
Loren Merritt | workaround: on some systems, alloca() isn't aligned |
blob | commitdiff | raw | diff to current |
2006-09-12 |
Loren Merritt | faster ESA |
blob | commitdiff | raw | diff to current |
2006-09-12 |
Loren Merritt | faster ESA |
blob | commitdiff | raw | diff to current |
2006-04-24 |
Loren Merritt | Before evaluating the RD score of any mode, check satd... |
blob | commitdiff | raw | diff to current |
2006-04-24 |
Loren Merritt | When evaluating predictors to start fullpel motion... |
blob | commitdiff | raw | diff to current |
2006-04-10 |
Loren Merritt | more interleaved SAD. |
blob | commitdiff | raw | diff to current |
2006-04-10 |
Loren Merritt | more interleaved SAD. |
blob | commitdiff | raw | diff to current |
2006-04-10 |
Loren Merritt | interleave multiple calls to SAD. |
blob | commitdiff | raw | diff to current |
2006-03-20 |
Loren Merritt | RD subpel motion estimation (--subme 7) |
blob | commitdiff | raw | diff to current |
2006-02-27 |
Loren Merritt | copy current macroblock to a smaller buffer, to improve... |
blob | commitdiff | raw | diff to current |
2006-01-18 |
Loren Merritt | early termination in UMH search |
blob | commitdiff | raw | diff to current |
2006-01-18 |
Loren Merritt | split mv_range enforcement from edge-of-frame clipping... |
blob | commitdiff | raw | diff to current |
2006-01-17 |
Loren Merritt | cosmetics: simplify #includes |
blob | commitdiff | raw | diff to current |
2006-01-09 |
Loren Merritt | early termination within large SADs. ~1% faster UMH... |
blob | commitdiff | raw | diff to current |
2005-12-30 |
Loren Merritt | MSVC compatibility fixes |
blob | commitdiff | raw | diff to current |
2005-12-30 |
Loren Merritt | joint bidirectional motion refinement (--bime) |
blob | commitdiff | raw | diff to current |
2005-12-20 |
Loren Merritt | Successive elimination motion search: same as exhaustiv... |
blob | commitdiff | raw | diff to current |
2005-12-03 |
Loren Merritt | faster subpel motion search. |
blob | commitdiff | raw | diff to current |
2005-11-22 |
Loren Merritt | subpel search: always check mvp. |
blob | commitdiff | raw | diff to current |
2005-11-03 |
Loren Merritt | skip a few duplicate candidates in qpel search. |
blob | commitdiff | raw | diff to current |
2005-11-03 |
Loren Merritt | skip a few duplicate candidates in fullpel hex&umh... |
blob | commitdiff | raw | diff to current |
2005-11-03 |
Loren Merritt | X264_ABS => abs |
blob | commitdiff | raw | diff to current |
2005-10-15 |
Loren Merritt | Use SAD instead of SATD for halfpel motion search. |
blob | commitdiff | raw | diff to current |
2005-10-08 |
Loren Merritt | new option: --mixed-refs |
blob | commitdiff | raw | diff to current |
2005-06-17 |
Loren Merritt | lossless mode enabled at qp=0 |
blob | commitdiff | raw | diff to current |
2005-06-14 |
Loren Merritt | rate-distortion optimized MB types in I- and P-frames... |
blob | commitdiff | raw | diff to current |
2005-06-05 |
Loren Merritt | SATD-based decision for 8x8 transform in inter-MBs. |
blob | commitdiff | raw | diff to current |
2005-05-24 |
Loren Merritt | UMHex motion seach (but no early termination yet) |
blob | commitdiff | raw | diff to current |
2005-05-11 |
Loren Merritt | New ratecontrol options: |
blob | commitdiff | raw | diff to current |
2005-05-11 |
Loren Merritt | use a predicted mv as starting point for subpel refinement. |
blob | commitdiff | raw | diff to current |
2005-04-19 |
Loren Merritt | Allow manual selection of fullpel ME method. New method... |
blob | commitdiff | raw | diff to current |
2005-03-15 |
Loren Merritt | macroblock_analyse: simplify cost comparisons. (cosmetic) |
blob | commitdiff | raw | diff to current |
2005-03-14 |
Loren Merritt | Chroma ME (P-frames only). |
blob | commitdiff | raw | diff to current |
2005-03-02 |
Loren Merritt | Fix a buffer overrun with very long MVs. |
blob | commitdiff | raw | diff to current |
2005-02-23 |
Loren Merritt | Preliminary adaptive B-frame decision (not yet tuned). |
blob | commitdiff | raw | diff to current |
2005-02-14 |
Loren Merritt | Precalculate lambda*bits for all allowed mvs. 1-2%... |
blob | commitdiff | raw | diff to current |
2005-02-14 |
Loren Merritt | Save some memcopies in halfpel ME. |
blob | commitdiff | raw | diff to current |
2005-02-13 |
Loren Merritt | Cache half-pixel interpolated reference frames, to... |
blob | commitdiff | raw | diff to current |
2005-02-10 |
Loren Merritt | cosmetics |
blob | commitdiff | raw | diff to current |
next |