[FFmpeg-devel] v210 decoder patch

Francois Oligny-Lemieux eucloid
Wed Jul 18 01:41:54 CEST 2007


Thanks for your answers, it is very appreciated.

On 7/17/07, Michael Niedermayer <michaelni at gmx.at> wrote:
> Hi


> how does AVI/RIFF would
> interface with swscale. From what I know it matches a CODEC_ID_XXX from a
> > RIFF code, then load this codec and pass the packets to it. How would
> this
> > work with swscale ? Would it use CODEC_ID_RAWVIDEO ? CODEC_ID_RAWVIDEO
> is
> yes
> > implemented in raw.c. How would the convertion be triggered in raw.c ?
> it would not
> > Would
> > that mean I would have to make a new pixel format with 10-bits per pixel
> > similar to YUV422. Then at some place in raw.c I would need to include
> my
> > convertion routine (see patch) to fill up this theorical 10-bpp YUV
> picture.
> no, you must extend swscale to do the convertion

Do you mean adding swscale functions calls inside libavcodec/raw.c ?
If yes, I would have to look how I can do this to fit my needs as in my
project I don't include libswscale (and I guess others too) so I was
thinking to put the generic convert function in a seperate file under the
libswscale realm with only dependencies on .h of libswscale so libavcodec
can still compile without libswscale easily (instead of including "swscale.h"
in raw.c, you would include this "newfunc.h" and include that newfunc.o in
the project. Although libswscale has a nice value my guess was to keep
libavcodec free (or almost free) of libswscale dependencies.

What I'm thinking to write will just take weird/silly pixel format and place
them in one of the well-known pixel format (YUV420P, YUV422, YUV422_16BIT),
i.e., it will not convert YUV422 to YUV420 or stuff like that. Let me know!

the only convertion which i would accept is to 16bit per component YUV422
> but that will be more work as you still must add 16bit YUV422 support to
> swscale
> convertion to any RGB format is totally unaccpetable, so is to any <10bit
> yuv format or yuv format of different subsample factor

Ok, I'm keeping this as a backup solution if A doesn't work for me.


More information about the ffmpeg-devel mailing list