On 9 Okt, Gabriel Bouvigne wrote:
> For those interested, it seems that Berkely multimedia workload is a study
> centered around simd optimizations. One of the tests is Lame, for which some
> optimized simd code (including altivec) has been specifically produced.
> Unfortunately it seems that the Lame version is very old.
>
> It's not easy to have details because the pdf are nearly unreadable, the ps
> don't work correctly with my ghostview, and the downloads are huge.
I've downloaded parts of the mm_accel_suite (just enough to extract
mm_accel_suite/applications/lame/src_{accel,orig}/) and did a diff (diff
-durH -W 160 --side-by-side --suppress-common-lines src_orig
src_accel/src |less). I've found no important change, only code to
measure the numbers in mm_workload_publish.pdf.
Where do you have seen the optimized simd code?
Bye,
Alexander.
--
Failure is not an option. It comes bundled with your Microsoft product.
http://www.Leidinger.net Alexander @ Leidinger.net
GPG fingerprint = C518 BC70 E67F 143F BE91 3365 79E2 9C60 B006 3FE7
_______________________________________________
mp3encoder mailing list
[EMAIL PROTECTED]
http://minnie.tuhs.org/mailman/listinfo/mp3encoder
