You've already forked FFmpeg
mirror of
https://github.com/FFmpeg/FFmpeg.git
synced 2026-05-04 21:08:03 +02:00
76c7277385
233 to 105 cycles on Arrandale and Win64. Replacing the multiplication by s_m[m] by a pand and a pxor with appropriate vectors is slower. Unrolling is a 15 cycles win. A SSE version was 4 cycles slower. Signed-off-by: Michael Niedermayer <michaelni@gmx.at>