MMX quantization functions, and optimization of the C versions.
authorLoren Merritt <pengvado@videolan.org>
Wed, 14 Sep 2005 17:20:17 +0000 (17:20 +0000)
committerLoren Merritt <pengvado@videolan.org>
Wed, 14 Sep 2005 17:20:17 +0000 (17:20 +0000)
commitc4ffed4986fe4706c0c5c2b514ce95668f0b8393
tree31ff42d62124d109ee86e688434ecb152bfe2cb1
parent16f423a00ee91f692d441a31fa99394543995582
MMX quantization functions, and optimization of the C versions.
about 3x faster quant_8x8, quant_4x4, quant_4x4_dc, and quant_2x2_dc. total speedup: 4-10%.
patch by Alexander Izvorski and Christian Heine.

git-svn-id: svn://svn.videolan.org/x264/trunk@293 df754926-b1dd-0310-bc7b-ec298dee348c
Makefile
common/amd64/quant-a.asm [new file with mode: 0644]
common/common.h
common/i386/quant-a.asm [new file with mode: 0644]
common/quant.c [new file with mode: 0644]
common/quant.h [new file with mode: 0644]
encoder/encoder.c
encoder/macroblock.c