Pierre Hugonnet schrieb am Mon, 04 Sep 2000:
> Robert Hegemann wrote:
> >
> > hmm, this was the meaning of -V in the very first beginning
> > of LAME's VBR code.
> >
> > Actually you influence the absolute threshold of hearing aka ATH
> > and the masking thresholds computed by GPSYCHO. VBR tries to
> > avoid distorted bands, but sometimes, when there are not enough
> > bits, they will survive.
> >
> Not enough bits ? Then VBR should select a higher bitrate ?
Well, even the highest bitrate is finite ;-)
> >
> > This is a little bit problematic, because the number of distorted
> > bands does not tell you the weight of distortions you will get.
> > What do you think sounds uglier:
> > 20 distorted bands, each 0.1 dB
> > or
> > 1 distorted band by 2 dB
> > ???
> >
>
> I would say that 20 distorted bands at 0.1dB is preferable, and assume that this is
>the choice of lame's algorithms. Is this assumption wrong ?
The default quant_compare routine in LAME -X0 would say one
distorted band is better than 20. But there are another seven
different quantization comparing possibilities you can select
from. Maybe we should document them, but you can dig in the
mail archive, early this year I described most of them.
> Pierre
Ciao Robert
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
