[FFmpeg-devel] [PATCH] fftools/ffplay: 240M matrix is not the same as BT.601

[Marton Balint]

On Tue, 8 Jun 2021, Valerii Zapodovnikov wrote:

> You are **very** wrong. This is YCbCr 101: 420 has all the same colors as
> 444 does. Just if one pixel is fixated the entagled pixels have less than
> all possible colors. This is also not corners issues, it is reproducable on
> one color all over the plane.

Then do reproduce with that and attach the sample having the same metadata 
(colorspace/primaries/transfer) to the ticket, with mpv and ffplay 
screenshots with the rendered output.

I am asking this, because based on the tests I have made I still believe 
there is no color space issue, but feel free to prove me wrong.

Thanks,
Marton