I have an application that decodes MPEG transport streams and
outputs scaled images to a framebuffer. It's running on a 2 GHz
Core Duo, and there is some CPU power unused.
Currently, with 1280x720 input images there is some choppiness
at 1024x576 in the video, but not audio. There is none noticable
at 512x288 (preview). From the demuxer, I have mpeg2dec, videoscale,
ffmpegcolorspace, and my fbdevsink, separated by queues. The fbdevsink
takes RGB, in this case 32 bpp, depth 24, and the transport streams
decode to YV12 (or it can be I420).
I don't see any hints what the relative performance is for nearest
neighbor, bilinear, and 4 tap scaling. BTW, I have not been able
to get 4-tap scaling to work, even with the colorspace conversion after
the scaler (the fbdevsink color space property seems to be passed up to
the scaler, which cannot scale RGB, so it refuses).
Any links to help me choose between the scaling methods?
If 4-tap would be best for me, any idea how to get the scaler to output
YV12 to the color space converter (another filter element between the
color space converter and the frame buffer sink??)?
Thanks.
------------------------------------------------------------------------------
Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are
powering Web 2.0 with engaging, cross-platform capabilities. Quickly and
easily build your RIAs with Flex Builder, the Eclipse(TM)based development
software that enables intelligent coding and step-through debugging.
Download the free 60 day trial.
http://p.sf.net/sfu/www-adobe-com_______________________________________________
gstreamer-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gstreamer-devel
Locked
