On Wed, 30 Aug 2006 09:31:13 +0100
Dieter <[EMAIL PROTECTED]> wrote:
> > or you have to transfere back the data into host memory
> > which makes the whole acceleration useless.
>
> It seems that this would depend on how many CPU cycles you are saving
> by doing work in the graphics chip verses the overhead of transferring
> the data.
Actually very little. CPU speed is for 8x8 DCT based systems
not the problem. They are fast enough. The problem lies in
shuffle the data fast enough to and from the host memory.
Newyer systems that are based on wavelets exhibit CPU limitations,
but also cache starvation due to accessing large areas of the
image (aka memory).
> This seems like a good point to post a feature suggestion/request I have
> received:
>
> } It would be nice to be able to have access to the raw video data after a
> } frame has been displayed.
I don't know how hard this is to do, but i would very much like
to see this too.
Attila Kinali
--
心をこめて聞け心をこめて話せ
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)
