[FFmpeg-user] zimg colorspace yuv -> rgb possible?

Andy Furniss adf.lists at gmail.com
Wed Jan 27 22:26:17 CET 2016


Carl Eugen Hoyos wrote:
> Carl Eugen Hoyos <cehoyos <at> ag.or.at> writes:
>
>> Andy Furniss <adf.lists <at> gmail.com> writes:
>>
>>> ffmpeg -loglevel debug -i snk-420i.y4m -vf scale=interl=1
>>> ffmpeg.png
>>
>> Do you understand this message to add the missing options?
>> http://thread.gmane.org/gmane.comp.video.ffmpeg.devel/181474/focus=181547

Been
>>
AFK, haven't had time to think about that one yet.

> Or maybe you should just tell scale that you first want yuv422p, then
> - in a second pass - rgb:
> http://thread.gmane.org/gmane.comp.video.ffmpeg.devel/163152/focus=187817

So
>
there's a -

"scale filter probably can be used instead of ilpack"

an

"i believe yes"

and

"I didn't test, but apparently ilpack can be dropped"

:-)

To be fair I will never be certain about anything - and it's possible
that ilpack "wins" by doing something different in the weighting.

The author clearly knew that the samples were 25/75% distant -

http://mplayerhq.hu/pipermail/mplayer-users/2004-March/043508.html

Yet the code seems to weight 3/8 5/8 (maybe I misunderstand!)

earlier in the thread he mentions doing "proper weights"

http://mplayerhq.hu/pipermail/mplayer-users/2004-March/043455.html

> -vf scale=interl=1,format=yuv422p (or -vf
> format=yuv422p,scale=interl=1 ?)

I think I have already posted that this is better but not as good a ilpack.

> Note that this is still exactly the information you pass to MPlayer.

I fail to understand this sentence!

I know the output of MPlayer + ilpack is 422 if that's what you mean,
but it is visibly different from that made when doing

ffmpeg -i in-420i  vf scale=interl=1,format=yuv422p ...


More information about the ffmpeg-user mailing list