On Thursday 10 February 2005 18:36, Isaac Richards wrote:
> On Wednesday 09 February 2005 02:03 am, Mark Anderson wrote:
> > Folks,
> >
> > Attached is a patch that works around the problem caused by the via
> > unichrome XvMC and AC3 patches not liking each other. I have posted this
> > fix before but it won't patch cleanonto cvs now, this will.
>
> Applied, but I had to spend quite a bit of time fixing this to work
> properly. You were leaking quite a bit of ram on each channel change, and
> it would  crash after 6 or 7 channel changes in a row.
>
Sorry about that, should have used a few more brain cycles on that one. 
Although I have been running my frontend with this code for three or four 
days now with no obvious problems, although I have plenty of spare ram. 

> I _really_ don't like this method of detecting stream changes.  Why can't
> you just check for the current audio track not sending data anymore?  That
> would work a hell of a lot better, and would properly deal with audio
> streams disappearing while on the same channel.
>
I don't think I would go so far as to say a "hell of a lot better" but it 
would be a better solution. I simply haven't had any spare time to look at 
it. I would have to count video frames in GetFrame and re-init audio after ?? 
frames received without audio. This will probably introduce audio lag when 
changing channels and so we would probably end up with the code in 
Reset to get the audio back quicker. Anyway it needs some more thought and 
some experimentation which I haven't got around to at the moment.
  
Cheers
Mark.
> Isaac
> _______________________________________________
> mythtv-dev mailing list
> [email protected]
> http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev
_______________________________________________
mythtv-dev mailing list
[email protected]
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev

Reply via email to