Re: [PULL] cheese: overrun callback for slow machines

[Bastien Nocera <hadess hadess net>]

On Thu, 2010-09-09 at 10:33 -0700, Brandon Philips wrote:
> On 11:26 Wed 25 Aug 2010, Bastien Nocera wrote:
> > On Wed, 2010-08-25 at 12:23 +0200, Alexey Fisher wrote:
> > > Am Mittwoch, den 25.08.2010, 09:50 +0100 schrieb Bastien Nocera:
> > > > On Mon, 2010-08-23 at 18:07 -0700, Brandon Philips wrote:
> > > > > Hello-
> > > > > 
> > > > > This is one half of my solution to slow netbooks running cheese and
> > > > > trying to encode videos[1]. The basic idea here is to detect a pipeline
> > > > > stall and inform the user that their machine is too slow to encode video
> > > > > in real time. It suggests that lowering the resolution might help.
> > > > 
> > > > I would much rather that the uncompressed video was queued to disk, and
> > > > compression finished when we stop the capture.
> > > > 
> > > > Cheers
> > > 
> > > Probably cheaper for cpu and hdd will be mjpeg than raw.
> > 
> > Right, or as close as possible to whatever is handed by the camera,
> > although libv4l will probably do conversion anyway.
> > 
> > The point is that we should be able to capture directly to OGV, even if
> > the machine isn't fast enough to do this real time.
> 
> How do you capture directly to OGV if it can't encode to real time? By
> definition don't you need an intermediate cache on disk in some file
> format?

Yes, you do. But the capture shouldn't block the display, and it should
still be possible without popping up error messages, or asking the user
to change the video resolution.