Mercurial > libavcodec.hg
graph
-
Optimize 0 0 0-3 search, 45% faster on pentium dual. libavcodecTue, 23 Dec 2008 00:38:45 +0000, by michael
-
Simplify decode_cabac_mb_ref() a little bit, 2 cpu cycles faster on libavcodecMon, 22 Dec 2008 17:14:13 +0000, by michael
-
inline decode_cabac_mb_type for I & P frames, 9 cycles faster on pentium dual. libavcodecMon, 22 Dec 2008 16:46:17 +0000, by michael
-
Negate 2 more variables, 1 cpu cycle faster on pentium dual. libavcodecMon, 22 Dec 2008 16:20:13 +0000, by michael
-
Negate a few variables, this simplifies the code and makes it 5 cycles faster libavcodecMon, 22 Dec 2008 16:10:35 +0000, by michael