[FFmpeg-user] Developer "cehoyos" closed my bug without any explanation, and without solving it

Alexey Eromenko al4321 at gmail.com
Tue Oct 11 03:30:40 EEST 2016

> Hello,
> sorry you feel this way, but looking at your bug report do you seem to have
> a misunderstand regarding who is responsible for which bugs. You state that
> it's Apple's decoder that cannot play the videos. And at the same time do
> you state VLC does play them. VLC uses ffmpeg's libraries so it obviously
> works with ffmpeg, just not with Apple's decoder. Further more do you state
> that the videos were created with libx264, which if it was the source of
> your problem, is also not part of ffmpeg but only an external library and
> has its own project. In short, you have no case.
> You first need to accept that the problem is very likely with the Apple
> decoder. Just because you cannot find anyone at Apple to listen to your
> problem doesn't mean you can shove it onto ffmpeg. You have also reopened
> the bug report several times, which suggest a certain stubbornness in
> accepting these facts. You then have to provide evidence of this being an
> ffmpeg bug, and you have so far only provided evidence against it. Cehoyos's
> responses were also quite clear and not unfriendly at all, because he told
> you how to solve your problem.
> I suggest that you change your stance in this matter and stop acting like
> Cehoyos would owe you an apology.
> User problems are handled here on this maling list and not on the bug
> tracker. Or do you disagree?

It well may be an Apple bug, but since I can't find anyone at Apple to
fix it, I think fixing it in the encoder is way simpler, and it is the
way to go. Especially because it's open-source and because Apple
players are popular.

Discovering after-the-fact that some of my videos are incompatible
with Apple decoders is a no-no, and a total shock for me. And no
documentation on such fundamental issues _AT ALL_. Even the workaround
parameter he suggested is not documented either.

Like I know for sure that a BMP file is a Bitmap and is displayed
correctly every time, on all computers and operating systems, and MP3
file is an audio file, and is played back correctly every time, on all
supported platforms, I expect the same from MP4 video file -- it must
be played perfectly on all supported platforms, and since those
poipular decoders are well-known, the encoder can target their
limitations too.

For ffmpeg developers fixing this bug maybe simple, just parsing the
container tags properly, and yes, testing resulting MP4 files with the
popular media players is super-critical; This means at least Windows
(Windows Media Player and VLC), Android (Google Nexus and Samsung
Galaxy) and Apple (iOS + Mac OS). Perhaps not for every minor release,
but for major releases QA is a must-have.

I can do some of the testing myself, being a QA guy, I'm capable of
this task, but when there is a severe and critical issue, alarms must
be sounded. It's my duty. (+I can add testing on Chromebook and Debian
Linux and other platforms)

-Alexey Eromenko "Technologov"

More information about the ffmpeg-user mailing list