It appears as if the devs still don't get (or care about) all the useful things you could do with VCRTCs. The biggest point that came up is that it doesn't (yet?) do something it wasn't designed to do, enable optimus GPU switching.
I'm still convinced that virtual CRTCs in one form or another are a very useful tool to have. Stick to your guns, Ilija!
Originally Posted by ZedDB
Can this perhaps be used as a way to record the screen and save the output to a video file on disk?
probably, though I don't see any architectural advantage over current methods (like RecordMyDesktop). While it's nice to access the video data as a v4l-stream for compatibility with video chats etc, it shouldn't be faster than grabbing pixel data directly from the onscreen framebuffer. At least not until GPU video encode is possible in OSS drivers.
Originally Posted by alazar
Has this any advantages over VirtualGL?
Both are solutions to different problems. Right now I can't come up with a use case where one would replace the other.