[FFmpeg-devel] H.264 Lossless Decoding Patch (Profile 244) for 8x8 Intra Prediction

Yogender Kumar Gupta yogender.gupta at gmail.com
Tue Mar 18 11:31:15 CET 2014


Please find attached the patch that takes care of x264 builds, basically
calls an alternate function only when x264_build = -1.

Thanks,
Yogender


On Sat, Mar 15, 2014 at 5:44 PM, Ronald S. Bultje <rsbultje at gmail.com>wrote:

> Hi,
>
> On Sat, Mar 15, 2014 at 7:43 AM, Reimar Döffinger
> <Reimar.Doeffinger at gmx.de>wrote:
>
> > On Wed, Mar 05, 2014 at 08:15:50PM +0100, Hendrik Leppkes wrote:
> > > On Wed, Mar 5, 2014 at 8:14 PM, Reimar Döffinger
> > > <Reimar.Doeffinger at gmx.de> wrote:
> > > > On 05.03.2014, at 20:05, Jason Garrett-Glaser <darkshikari at gmail.com
> >
> > wrote:
> > > >> This will break decoding on all x264-encoded lossless files (the
> vast
> > > >> majority of lossless files) produced in the past half decade or so.
> > > >
> > > > Any chance of auto-detecting this? (reliably?)
> > > > Either way I guess it should be configured by a bug workaround flag
> as
> > we do for other codecs.
> > >
> > > The decoder already parses the x264 version string and sets
> > > H264Context->x264_build. Could make it switch implementation based on
> > > that. That is assuming the person encoding didn't go out of their way
> > > to purge this information.
> >
> > We can add an option to also force it, but in general I expect
> > autodetection based on x264_build should be good enough.
> > IMHO it will at least be better than not matching reference decoder.
> >
>
> Autodetection sounds good. Can someone write a patch for this?
>
> Ronald
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: lossless_patch
Type: application/octet-stream
Size: 5067 bytes
Desc: not available
URL: <http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20140318/51343e50/attachment.obj>


More information about the ffmpeg-devel mailing list