Thanks for answering, Ivan! > Are you sure this file plays ok if you decode it in software only mode > with -vc ffmpeg12 ? There are no indication for any kind of errors. Yes, just double-checked that again, it plays nicely. The only idea I have left concerns the LVDS screen which xrandr claims to be connected, but it isn't. If the video was played to the not-connected-display it would look just like it looks... The LVDS connector however isn't oficcially documented, so it's not easy to connect a display, and I can't persuade the graphic chip to think there was no LVDS display. I think I might have to wait for someone writing an intel driver with vaapi :-) > BTW, using MC only level of acceleration (aka no hw idct) is usually > much slower on all modern CPUs. Ah, ok, so it would probably be futile after all. Yet wikipedia says intel 8xx/9xx chips would support idct for MPEG2 via xvmc... I've got no idea if that's even possible. And it might depend on the meaning of "modern CPU". The intel atom works an in-order-concept which seemed to be out of date for ten years... Greets, Kiste _______________________________________________ MPlayer-users mailing list MPlayer-users@xxxxxxxxxxxx https://lists.mplayerhq.hu/mailman/listinfo/mplayer-users
