> > > Do you (or anyone) have any thoughts on JRT's suggestion of using a > > > DSP chip? The decoder needs to handle the most common formats, > > > mpeg at the very least, to avoid transcoding. Transcoding takes a > > > lot of CPU, and Theora is lossy. > > > > True. What would be nice would be a CPU with a built-in DSP, so that the > > CPU can handle the network stuff and OSD and so on, and the DSP could > > do the heavy-duty decoding work.
Timothy> How "realtime" does this need to be? Can we not implement the Timothy> networking stuff on the DSP? Realtime requirements probably vary with Ethernet chip? Nick> Well, are we still thinking about having a 1000Mbit network, or does Nick> 100Mbit offer enough bandwidth? If HD mpeg2 is the worst case, then 100Mbit is plenty. Timothy> I'm planning on implementing RGB <-> YUV conversation, at least. Timothy> Rather than spending bunches of CPU time doing the matrix algebra, we Timothy> can do it in hardware and save at least SOME converstion time. IIRC, OGC will also do scaling. Anything else it can offload? Nvidia claims their GPU does 95% of the decode for mpeg2. (page 9 of the pdf from the other day.) They claim to save 9 Watts decoding HD on gforce7 vs on CPU. (page 19) I don't see a mention of what CPU they used. If OGC can't offload enough, are there HD mpeg decoder chips that we could consider? (e.g. documented) _______________________________________________ Open-graphics mailing list [email protected] http://lists.duskglow.com/mailman/listinfo/open-graphics List service provided by Duskglow Consulting, LLC (www.duskglow.com)
