Matthew Hodgson wrote:
So, I decided to give John Harvey's exciting YUV decoding patch a go on 0.3.4p - and so far it's all working fairly well :) I've had a few problems, though, which I thought I'd share:

a) I'm feeding the decoder 25fps progressive PAL video to /dev/video48. However, the YUV decoder renders both fields of each frame using only the odd field's buffer (i.e. lines 1,3,5.. counting from the topmost line being 1). This obviously halves the vertical resolution of the luminance, which makes the visual quality considerably worse than the OSD - especially noticeable on aliasing artefacts in animation. This occurs after initialiasing the MPEG decoder with a typical MPEG2 recording sent to /dev/video0.

I just encoded a dedicated MPEG2 clip at PAL 25fps progressive rather than interlaced and used it to initialise the YUV decoder by sending it to /dev/video0. Sure enough, I can now see both fields of progressive YUV I send to to /dev/video48 (woo! :D). However: my new problem is that one field (the odd field) is about 20% darker than the other, yielding migraine-inducing flickering. The data in the framebuffer itself is absolutely fine - but something in the card seems to be using different gains on the luminance or chrominance (or both) components of the two channels.


Does anyone know what that could be, or have suggestions on how to tame it? The progressive MPEG2 plays back without flickering...

M.


------------------------------------------------------- This SF.Net email is sponsored by Oracle Space Sweepstakes Want to be the first software developer in space? Enter now for the Oracle Space Sweepstakes! http://ads.osdn.com/?ad_id=7412&alloc_id=16344&op=click _______________________________________________ ivtv-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/ivtv-devel

Reply via email to