On Mon, 28 Aug 2006 10:38:00 -0400
"Timothy Miller" <[EMAIL PROTECTED]> wrote:
> On 8/27/06, Dieter <[EMAIL PROTECTED]> wrote:
>
> > I've been researching this a bit. Preliminary conclusion is that the
> > OGC card needs to provide XvMC.
>
> Any patents apply here?
The question should be rather, which patent does not apply.
Video coding is so full of patents, that it is impossible
to design a codec without violating at least a dozen patents.
Of course, most of these patents are trival patens, have prior
art or just patent mathematical algorithms known for decades.
> Anyhow, I don't want to support any one specific codec. If we can
> provide something that's as useful for Theora as it is for MPEG1, 2,
> and 4, then I'll consider it, depending on how much logic is required.
I'd say don't consider it for OGA1. If you want to stay
generic, all you can do is to implement an IDCT. But that does
not save any bandwidth. Anything else would be need to feed the
compressed bitstream which requires knowledge of the codec on the
card (yes, this can be done with some sort of co-processor and uploading
software to it, but that has other difficulties).
If we talk about bandwith problems because we use PCI instead
of PCI-e or AGP, then just state that the PCI version is
not meant for these kind of high-bandwith[1] operation.
PCI-e should cleanly solve that issue (given that the chip
can handle the data rate)
Attila Kinali
[1] a 1080p30 16:9 content needs around 120MByte/s of
data rate. For smooth playing and various other considerations,
a minimum of 150 to 180MByte/s bus speed is required.
--
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)
