[FFmpeg-devel] [PATCH] FFV1 rectangular slice multithreading

[Michael Niedermayer]

On Sat, Oct 16, 2010 at 02:24:42AM +0200, Reimar D?ffinger wrote:
> On Sat, Oct 16, 2010 at 12:49:57AM +0200, Michael Niedermayer wrote:
> > On Fri, Oct 15, 2010 at 07:31:00AM +0200, Reimar D?ffinger wrote:
> > > On Fri, Oct 15, 2010 at 04:23:09AM +0200, Michael Niedermayer wrote:
> > > > On Thu, Oct 14, 2010 at 04:34:50PM -0700, Baptiste Coudurier wrote:
> > > > > On 10/14/2010 04:22 PM, Michael Niedermayer wrote:
> > > > >> On Thu, Oct 14, 2010 at 02:59:31PM +0200, Michael Niedermayer wrote:
> > > > >>> Hi
> > > > >>>
> > > > >>> Following patchset makes ffv1.2 much faster on multiprocessor systems
> > > > >>> (requires version to be set to 2, which needs you to edit the source if you
> > > > >>>   want to try as 1.2 bitstream is not finalized yet)
> > > > >>>
> > > > >>> Compression wise 4 slices with foreman and large gops (300 frames) perform
> > > > >>> slightly better (0.05% IIRC) than 1 slice.
> > > > >>> With small gops (50 frames) compression is worse with the rangecoder and the
> > > > >>> large context model by 0.8% otherwise better too.
> > > > >>> (its quite obvious why its worse in that case and ill be working on that ...)
> > > > >>>
> > > > >>> Comments welcome, bikesheds not, and ill apply this soon
> > > > >>
> > > > >> applied, flame me on svnlog
> > > > >
> > > > > I quickly tested it, very nice speedup, could we get speedup for  
> > > > > decoding too ? It's now encoding faster than decoding :)
> > > > 
> > > > It speeds up decoding too you need new v1.2 bitstreams and you need to put
> > > > -threads x before -i
> > > 
> > > May I suggest a regression test using all that?
> > 
> > That would require us to enable v1.2 bitstream generation with some flag even
> > though its not final and could not be decoded in the future
> 
> Sorry, I misunderstood which stage this is in. Adding this when it becomes
> actually available sure is soon enough

> (though adding a commented-out one
> as soon as you think it is reasonably stable might be useful in the unlikely
> event that someone else wants to work on it).

feel free to add that

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

The worst form of inequality is to try to make unequal things equal.
-- Aristotle
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20101016/0219e8a1/attachment.pgp>