[mythtv-users] Just how great can VDPAU be?

[Markus Schulz]

Am Mittwoch, 26. November 2008 schrieb Mark Kendall:
> 2008/11/26 David Engel <david at istwok.net>:
> > On Thu, Nov 20, 2008 at 06:21:48PM +0800, Mark Kendall wrote:
> > Are you using Isaac's patch, your own patch or something else? 
> > Isaac said OSD wasn't working in his first patch so I suspect
> > you're using a later version or doing your own thing.
>
> It's a modified version of Isaac's original patch. OSD,
> deinterlacing, PiP all implemented - though it looks like there are
> issues with VDPAU (i.e. not mythtv or mplayer) that are causing
> problems.
>
> > Does time-stretching work with VDPAU?  I thought I read something
> > about VDPAU controlling the timing of presented frames so I'm
> > concerned it might not work with time-stretching.
>
> Currently, time-stretch is fine (and pretty impressive on hidef
> h264!). Isaac is interested in looking into extending the use of the
> vdpau presentation queue and adding it to the vertical sync code -
> though I can't imagine anyone is going to implement it in such a way
> that it breaks time stretch (or anything else).

Anyone tried some graphic card benchmarking (not cpu-usage)?
Was an 8200 (or comparable) onboard chipset enough for really high 
bitrate h264 full-hd streams, or possible two of them (pip)?
If i understand right: with VDPAU _everything_ will be done on the 
graphic card, therefore the cpu can't help if the card was to slow for 
processing the video cause high bitrate/resolution/..?

-- 
Markus Schulz