[Ffmpeg-devel] [PATCH] New grabbing interface

Guillaume POIRIER poirierg
Sat Feb 10 00:01:31 CET 2007


Hi,

On 2/8/07, Michael Niedermayer <michaelni at gmx.at> wrote:
> On Wed, Feb 07, 2007 at 11:29:51PM -0200, Ramiro Polla wrote:
> > Two choices attached:
> > grabbing_interface_1.diff makes input_sync an option (-isync).
> > grabbing_interface_2.diff adds a flag to AVFMT and sets input_sync if
> > any grabbing interface is selected.
> >
> > Both patches pass regressions tests.
> >
> > Tests made against unpatched FFmpeg.
> >
> > video4linux + audio:
> > Same results with second patch.
> > Same results with first patch using -isync.
> > Bad results with first patch not using -isync. Fewer video frames (now I
> > know why input_sync was added in the first place =).
> >
> > video4linux2 + audio:
> > Same results with second patch.
> > Same results with first patch using -isync.
> > I didn't notice any difference with the first patch not using -isync.
> > (This is probably what Luca mentioned).
> >
> > x11grab + audio:
> > Always same results.
>
> patch 1 looks ok

Though I can't test this patch because I don't have any grab hw, I've
tested that the patch builds fine and doesn't mess up regression
tests...

Michael, can I apply it, or should someone who have a grabbing
interface test it before commit?

Guillaume




More information about the ffmpeg-devel mailing list