[FFmpeg-user] Ticket#5936: Purple haze on ffmpeg mpeg4 decoding

[Toerless Eckert]

Thanks a lot, Andy

Mind explaining how you did that trick with git bisect ? 

Just verified:

Applying c5fc8ae12622a507d7b9ee30ddcd3734e6de6b1d to a working older version
of ffmpeg introduces the purple haze.

Reverting c5fc8ae12622a507d7b9ee30ddcd3734e6de6b1d even in latest 3.2 makes
purple haze go away.

Of course, i have no idea how/what type of problem with vsynth3-wmv2.avi
the fix was meant to correct. Just looking at the clip i couldn't figure it out.

Coulnd't make heads or tails out of the fate testing for vsynth3-wmv2, with
3.2, the output i see is:

    tests/data/fate/vsynth3-wmv2.rep 
      vsynth3-wmv2:0::

and with e12622a507d7b9ee30ddcd3734e6de6b1d, it shows:

vsynth3-wmv2:0::KysgZW5jX2Rl..... (and so on, long).


On Thu, Nov 10, 2016 at 06:33:28PM +0000, Andy Furniss wrote:
> Toerless Eckert wrote:
> >Just opened https://trac.ffmpeg.org/ticket/5936
> 
> >git diff 03bb99ae1a99fa315621308b885a8fc70702c9bc 9236f7b5a23b4907f7b2bf6346ecd88e6d76f1e0
> >produce 700k diff and i can not figure out what a likely cause is ;-(
> 
> Git bisect - can sometimes be tricky, but this was easy (he says
> hoping he found the right commit).
> 
> c5fc8ae12622a507d7b9ee30ddcd3734e6de6b1d is the first bad commit
> commit c5fc8ae12622a507d7b9ee30ddcd3734e6de6b1d
> Author: Michael Niedermayer <michaelni at gmx.at>
> Date:   Fri Jun 13 01:45:23 2014 +0200
> 
>     avcodec/mpegvideo: fix edge emulation with uvlinesize below 25
> 
>     Fixes fate-vsynth3-wmv2
> 
> _______________________________________________
> ffmpeg-user mailing list
> ffmpeg-user at ffmpeg.org
> http://ffmpeg.org/mailman/listinfo/ffmpeg-user
> 
> To unsubscribe, visit link above, or email
> ffmpeg-user-request at ffmpeg.org with subject "unsubscribe".

-- 
---
tte at cs.fau.de