[FFmpeg-devel] [RFC]lavc/ffv1dec: Scale msb-packed output to full 16bit

[wm4]

On Thu, 17 Nov 2016 21:16:13 +0100
Carl Eugen Hoyos <ceffmpeg at gmail.com> wrote:

> 2016-11-17 0:49 GMT+01:00 Kieran Kunhya <kierank at obe.tv>:
> >>
> >> It was lossless before and is lossless after the patch but it
> >> output the 16bit format that you don't like and of which you
> >> claim it breaks with find_best_pix_fmt() - I still believe this
> >> should be trivial to fix.
> >>
> >> Currently white is gray, the patch tries to fix this.  
> >
> > erm...in what way does it fix that?  
> 
> White output stops looking gray.

Stop trying to add bugs just because your trial&error seemed to have
shown an improvement.

This things don't "look" like anything - they have discrete output
values. kierank posted a useful link explaining how it works and why
the old code was correct - go read it.

> > A white/black value in 16-235 needs to remain like that when shifted.  
> 
> Since this patch isn't about 8bit (what 16-235 indicate),
> I didn't know what to do with your comment, sorry.
> 
> >This isn't RGB-land.  
> 
> As just written in the other mail:
> AV_PIX_FMT_GRAY8 and AV_PIX_FMT_GRAY16 are full-scale
> (because of repeated user-request).
> If you want to change this, please suggest a patch but this patch
> fixes current behaviour of one decoder compared to the scaler
> and all other decoders.

The GRAY formats are not marked as RGB formats.