[FFmpeg-devel] [PATCH] ALSA: use a different time filter.

[Michael Niedermayer]

On Wed, Dec 28, 2011 at 09:23:11PM +0100, Nicolas George wrote:
> L'octidi 8 nivôse, an CCXX, Michael Niedermayer a écrit :
> > Can you explain how the current timefilter becomes divergent ?
> 
> If I understand things correctly, it can be seen as a second order
> differential equation solved using the Euler method. This method is not very
> efficient at the best of time and requires quite a small time step. With the
> selected parameters, the characteristic time of the homogeneous solution
> seems to be about 150 ms while the time step on a typical ALSA setup would
> be something like 23 ms.

could you post the input values with which the timefilter misbehaves ?
iam curious what exactly happens ...

[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

I do not agree with what you have to say, but I'll defend to the death your
right to say it. -- Voltaire
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20111228/c86998f6/attachment.asc>