[FFmpeg-devel] [RFC]Remove experimental flag from aac encoder

Rostislav Pehlivanov atomnuker at gmail.com
Mon Aug 31 13:49:12 CEST 2015


The only real major feature the encoder doesn't quite have working fine yet
is TNS, for which I have a patch which I will push later today.
Everything else works fine. There are some infinite loops happening with
non-default extreme settings (very high bitrates with low frequency cutoffs
with the faac-like coder), but I have an idea how to fix them and will push
the fix for that probably tomorrow.
Claudio also has some changes which he still hasn't merged yet, not sure if
he has the time right now. I'll write to him to see what he's up to.

PNS, Intensity Stereo and maybe M/S need to get enabled by default since
they definitely improve the quality a ton. Might do that for PNS and IS
when I push the last TNS changes tonight.

Bottom line is, there are still some things which need to be pushed and I
think after that, in 4-5 days depending on what's happening with Claudio's
changes, I will personally be comfortable with removing the experimental
flag and declaring the encoder stable.

On 31 August 2015 at 10:59, Robert Krüger <krueger at lesspain.de> wrote:

> On Mon, Aug 31, 2015 at 10:44 AM, Carl Eugen Hoyos <cehoyos at ag.or.at>
> wrote:
>
> > Hi!
> >
> > I didn't test myself but iiuc, the aac encoder produces better
> > quality than all other libavcodec audio encoders and better
> > quality than the libvo-aac encoder that is not marked as
> > experimental, so the flag should be removed.
> >
>
> I haven't done scientifically thorough testing but it has improved a lot
> over the past few months and is definitely in a very usable state
> quality-wise (in 256kbit I couldn't hear differences to the Apple encoder,
> which regularly gets very high test scores, tested with HQ source material
> and equipment) and probably much better maintained than a lot of other
> pieces of non-experimental code. If there are no known, significant
> stability issues +1.
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>


More information about the ffmpeg-devel mailing list