[FFmpeg-devel] [PATCH] AAC Encoder, Round 2
Mon Aug 25 16:36:47 CEST 2008
On Mon, Aug 25, 2008 at 03:22:09PM +0200, Gabriel Bouvigne wrote:
> Michael Niedermayer a ?crit :
> >>1. Perform some input filtering (lowpass, highpass, stereo attenuation, whatever)
> > Its debateable in how far this should be here or seperate and outside of the
> > encoder.
> While highpass (mainly in order to remove DC) might belong to a general
> "preprocessing" step, highpass and stereo attenuation, while being
> filtering operations, might/should actually be driven by the encoder
> itself (psymodel or encoder core). So even if right now they are using
> fixed parameters, it might be a good thing to provide provisions within
> the interfaces to set up parameters of those filtering operations per frame.
> *stereo attenuation depends of current complexity of input. If the
> encoder is easily able to encode the input, there is no point in
> performing stereo atenuation.
> *lowpassing might increase temporal smearing in case of transcients, so
> it might depend on attack detection.
I agree partially, that is when the parameters are changed its not an
independant filter but iam not that sure if the parameters should be
the reason is that significant differences in lowpass or
stereo attenuation will lead to audible inconsistencies.
I have not done any listening tests but this is something that would need
such listening tests IMHO.
Because IMHO constant stereo attenuation of a signal at a low bitrate that
could not otherwise be encoded at such bitrate is likely not very annoying
even if its rather strong, but switching it on and off is almost certainly
Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB
The educated differ from the uneducated as much as the living from the
dead. -- Aristotle
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Size: 189 bytes
Desc: Digital signature
More information about the ffmpeg-devel