[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       |  klaussfreire
             Priority:  normal       |                   Status:  open
              Version:  git-master   |                Component:  avcodec
             Keywords:  aac          |               Resolution:
  regression                         |               Blocked By:
             Blocking:               |  Reproduced by developer:  1
Analyzed by developer:  0            |
-------------------------------------+-------------------------------------

Comment (by atomnuker):

 Kamedo2: yes, all encoder options now start with "-aac_":
 {{{
 #!div style="font-size: 80%"
 ffmpeg -help encoder=aac
   {{{
 AAC encoder AVOptions:
   -aac_coder         <int>        E...A... Coding algorithm (from -1 to 3)
 (default 2)
      faac                         E...A... FAAC-inspired method
      anmr                         E...A... ANMR method
      twoloop                      E...A... Two loop searching method
      fast                         E...A... Constant quantizer
   -aac_ms            <boolean>    E...A... Force M/S stereo coding
 (default false)
   -aac_is            <boolean>    E...A... Intensity stereo coding
 (default auto)
   -aac_pns           <boolean>    E...A... Perceptual noise substitution
 (default auto)
   -aac_tns           <boolean>    E...A... Temporal noise shaping (default
 auto)
   -aac_pred          <boolean>    E...A... AAC-Main prediction (default
 auto)
 }}}
 }}}

 Any option set to automatic means that the profile will determine it by
 default, unless it is set via the command line. Any option not set to a
 default 'auto' means the default value indicated will be set. Also,
 "-aac_ms" is not boolean as indicated but can be set to '-1' which means
 it will be automatically used when there will be an encoding gain.

 Keep in mind the psychoacoustic system currently doesn't account for the
 cutoff which the new coder introduced, leading to bits being wasted and
 the quality being decreased. Claudio will be pushing a patch to fix that.
 This only affects heavy synth samples but should fix a lot of bugs which
 might be related currently. This is also what's currently blocking us from
 removing the 'experimental' flag.
 Also, I still have to merge my LTP patches, which will happen later today.

 Kamedo2: Not sure how but I got an email invitation from Shion to Slack
 (Audio Video Encoding Community) which you are apparently a member of. I
 understand enough Japanese to kinda understand the email and I'd love to
 join but I'm still learning, let alone understanding technical jargon.
 Sorry :|
 Maybe after I know a little more.

--
Ticket URL: <https://trac.ffmpeg.org/ticket/2686#comment:428>
FFmpeg <https://ffmpeg.org>
FFmpeg issue tracker