>No, he is not correct. I guess you could be correct, it seemed to work correctly after making the symlink but on closer examination xine (or ogle) is not using the "raw" device like I thought it was (it is still using the buffered device). I was maybe to eager to admit that I was a dumbass :) >There is someting else "raw" related to I/O in Linux. >There is a "raw-io" capability in Linux that allows one to bypass >the page/buffer/etc cache, to allow certain application >(databases for example) to work on the "bare metal". yep, I figured that much out in my research in the subject. >See http://dvd.sourceforge.net/xine-howto/en_GB/html/howto-8.html for an >example of how to setup them. been over it, but the problem is I could not find the 'raw' executable they use in their examples to make the raw device.