<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> <html> <head> <meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type"> <title></title> </head> <body bgcolor="#ffffff" text="#000000"> Ricardo Baratto wrote: <blockquote cite="mid20040912150906.184020f9.ricardo@cs.columbia.edu" type="cite"> <pre wrap="">hi, newbie question here, </pre> <blockquote type="cite"> <blockquote type="cite"> <pre wrap="">Surely if not send bitstream over the protocol </pre> </blockquote> <pre wrap="">This would have been nice, since it would allow decoding to take place in the local X server of a cheap EPIA board. Unfortunately, the protocol implementation in the X server is not complete, and just implements surface and context creation / deletion and capability querying. </pre> </blockquote> <pre wrap="">is it possible to do all the decoding in hw with existing cards? in other words, if the protocol was complete could we just send the stream directly to the remote card and get remote hw video decoding? or one would still need to do some of the work in software in the local X server? </pre> </blockquote> Hi! Yes- and no. For, example, the CLE266 can do almost all decoding in hardware. Software overhead as low as 1-5% CPU. However, XvMC implies that the decoding is done either on the same computer the X server resides at (using DRI or direct rendering), or by the X server itself (indirect rendering). The latter is impossible right now because of the incomplete protocol, even though the XvMC standard allows it. Accelerated indirect can, for example, be used for playing xine on a remote computer while displaying on your local computer, which actually does all the decoding. (Watch your neighbour's DVD) Audio has to be transferred also, though. /Thomas <blockquote cite="mid20040912150906.184020f9.ricardo@cs.columbia.edu" type="cite"> <pre wrap=""> thanks, ricardo </pre> </blockquote> </body> </html>