After checking out the docs I see why. What i also noticed about xawtv is that it can compress at the same time as displaying the movie. The question then becomes exactly how does xawtv know that a new picture has been laid down into the frame buffer, is there a message sent to the application, or do they just set a timer to go off every 30 milliseconds (ntsc) or something? I think I would like to try what they are doing, but I also would like openGL to write on top of the buffer (to do some sweet overlays or some graphics effects). The ideal system would be to set up the capture device to write directly to an offscreen buffer, then have openGL use that buffer for its sampling and blending. Sound possible? Then the only semi-expensive call to swap the buffers. Perhaps you would also have to reset the driver pointer to write to a different section ofthe frame buffer or something, but anything is possible, eh? Anyway, just some thoughts, chris