log

age author description
Mon, 01 Oct 2007 14:23:36 +0000 diego Remove const vector macro indirection that is useless and obfuscating libavcodec
Mon, 01 Oct 2007 14:16:07 +0000 diego Remove Metrowerks compiler workaround. libavcodec
Mon, 01 Oct 2007 14:10:49 +0000 diego Remove Metrowerks compiler workaround. libavcodec
Mon, 01 Oct 2007 05:56:30 +0000 lorenm div -> mul libavcodec
Sun, 30 Sep 2007 03:36:13 +0000 lorenm simplify lpc libavcodec
Sun, 30 Sep 2007 03:01:56 +0000 lorenm 20% faster lpc, 6% overall flac decoding libavcodec
Sun, 30 Sep 2007 02:12:03 +0000 lorenm replace FIR with finite differences. libavcodec
Sat, 29 Sep 2007 23:44:34 +0000 lorenm unroll finite differences to avoid swapping registers. libavcodec
Sat, 29 Sep 2007 23:21:57 +0000 lorenm encode_residual_fixed(): replace FIR with finite differences. libavcodec
Sat, 29 Sep 2007 22:31:18 +0000 lorenm sse2 version of compute_autocorr(). libavcodec
Sat, 29 Sep 2007 15:20:22 +0000 vitor AMV video decoder. libavcodec
Sat, 29 Sep 2007 15:09:34 +0000 andoma Use emms_c() instead of ifdef libavcodec
Sat, 29 Sep 2007 12:52:56 +0000 vitor Add variables to prepare to the AMV decoder patch. libavcodec
Sat, 29 Sep 2007 07:43:52 +0000 lorenm was computing one more autocorrelation coefficient that was actually used libavcodec
Sat, 29 Sep 2007 05:41:27 +0000 lorenm replace brute force find_optimal_param() with a closed-form solution. libavcodec
Sat, 29 Sep 2007 01:54:25 +0000 lorenm gcc isn't smart enough to factor out duplicate stores libavcodec
Fri, 28 Sep 2007 20:38:32 +0000 iive On bitstream error is_intra_more_likely() could use dsp.sad[0]() that will segfault when xvmc is in use. libavcodec
Fri, 28 Sep 2007 06:06:18 +0000 lorenm 2.5x faster compute_autocorr() libavcodec
Thu, 27 Sep 2007 11:29:21 +0000 takis cosmetics: typos libavcodec
Thu, 27 Sep 2007 10:45:16 +0000 takis Remove redundant test. libavcodec
Thu, 27 Sep 2007 06:21:57 +0000 vitor AMV audio decoder libavcodec
Thu, 27 Sep 2007 03:16:25 +0000 lorenm oops, revert unrelated change libavcodec
Thu, 27 Sep 2007 02:42:00 +0000 lorenm unroll encode_residual_lpc(). speedup varies between 1.2x and 1.8x depending on lpc order. libavcodec
Wed, 26 Sep 2007 13:28:13 +0000 andoma h264/PAFF preparation: use DELAYED_PIC_REF to mark non-refs frames as held for delayed output libavcodec