[FFmpeg-devel] [PATCH] AAC: use table for cbrtf(n)*n

Michael Niedermayer michaelni
Mon Jan 11 23:05:08 CET 2010


On Mon, Jan 11, 2010 at 08:22:11PM +0000, M?ns Rullg?rd wrote:
> Alex Converse <alex.converse at gmail.com> writes:
> 
> > 2010/1/11 M?ns Rullg?rd <mans at mansr.com>:
> >> Alex Converse <alex.converse at gmail.com> writes:
> >>
> >>> On Mon, Jan 11, 2010 at 8:31 AM, Mans Rullgard <mans at mansr.com> wrote:
> >>>> Up to 10% faster on Cortex-A8.
> >>>> ---
> >>>> ?libavcodec/aac.c | ? 13 +++++++++----
> >>>> ?1 files changed, 9 insertions(+), 4 deletions(-)
> >>>>
> >>>
> >>> Otherwise LGTM
> >>
> >> So OK to apply this one?
> >>
> >
> > The only other potential issue I can think of is that MP3 uses this
> > same 4/3 quantizer and has it's own set of hacks to deal with it. But
> > I am OK with committing this as is and leaving reconciling the two as
> > a fight for another day.
> 
> Done.
> 
> MP3 is also fixed-point, so it's probably quite different.  I haven't
> looked at it in detail.

someone should add a float path to mp3 as that would make it much faster
on normal desktop hardware.

[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Opposition brings concord. Out of discord comes the fairest harmony.
-- Heraclitus
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20100111/06e57b47/attachment.pgp>



More information about the ffmpeg-devel mailing list