[FFmpeg-devel] Anybody has a Core 2? [PATCH] Small SSSE3 optimization

[Zuxy Meng]

Hi,

2007/5/9, Michael Niedermayer <michaelni at gmx.at>:
> Hi
>
> On Wed, May 09, 2007 at 06:16:36PM +0800, Zuxy Meng wrote:
> > Hi,
> >
> > Cool! Can u do a small unit-test to compare the MMX2 and SSSE3 version
> > of hadamard8_diff? Intel don't give the latency of pabsw in their
> > manuals (while AMD always give ALL instructions' latency & throughput)
>
> see http://www.agner.org/optimize/instruction_tables.pdf

Thanks! According to this document, pabsw is very fast: the latency is
merely 1 cycle and it has a throughput of 2 per cycle so the SSSE3
version should be considerably faster.
-- 
Zuxy
Beauty is truth,
While truth is beauty.
PGP KeyID: E8555ED6