[FFmpeg-trac] #2686(avcodec:open): Native AAC encoder collapses at high bitrates on some samples

[FFmpeg]

#2686: Native AAC encoder collapses at high bitrates on some samples
-------------------------------------+-------------------------------------
             Reporter:  Kamedo2      |                    Owner:
                 Type:  defect       |                   Status:  open
             Priority:  normal       |                Component:  avcodec
              Version:  git-master   |               Resolution:
             Keywords:  aac          |               Blocked By:
  regression                         |  Reproduced by developer:  1
             Blocking:               |
Analyzed by developer:  0            |
-------------------------------------+-------------------------------------

Comment (by Timothy_Gu):

 Replying to [comment:295 Cigaes]:
 > Replying to [comment:294 Timothy_Gu]:
 > > @cehoyos: this patch makes ANMR worse than default twoloop, even it is
 theoretically better and takes more time. While Claudio expresses interest
 to work on ANMR later, I don't think committing a patch that makes
 something worse than they should be is a good idea.
 >
 > IMHO, at this point, the question is not whether the ANMR coding is
 worse than it ''should be'' but whether it makes it worse than it
 ''currently is''.
 >
 > If we blocked patches because something could be done ''even better'',
 then the only acceptable patch series would be “[PATCH 0/85042] Make
 FFmpeg the ultimate multimedia software”.
 >
 > As I understand, this patch makes some modes work much better than now,
 with very little or no degradation on the little that did work: in my
 book, this is very good for inclusion. Knowing ways of making even better
 is good too, but for later patches.

 I agree with you. However this behavior now contradicts the behavior
 originally in the documentation, which should be either fixed in the code
 or documented.

 On a side note, can anyone check if ANMR with patch is better than
 without? If so then I have no problem landing the patch (except its nits)
 with the documentation changes.

--
Ticket URL: <https://trac.ffmpeg.org/ticket/2686#comment:296>
FFmpeg <https://ffmpeg.org>
FFmpeg issue tracker