>>>>> "M" == Mark Taylor <[EMAIL PROTECTED]> writes:
M> Just to explain a little more, scfi is a flag that tells the
M> encoder that some of the scalefactors for both granules (gr=0 &
M> gr=1) are the same. So it saves the encoding of some
M> scalefactors for gr=1, at the expense of requiring them to be
M> the same. I dont think it is a good idea to use scfsi for two
M> reasons:
M> 1) the usual excuse: FhG doesnt use it :-)
We can't overcome the FhG if we do the same as FhG does :)
M> To answer Q1 and Q2: I dont know! It should be in the dist10
M> encoder source, but I wouldn't trust it to be correct. The
M> dist10 source was buggy and would never turn scsfi on, so the
M> code was probably never tested. Better to take a look at the
M> mpglib decoder source in lame/mpglib/layer3.c and see what they
M> do.
OK, I checked them.... Uum, treating SCFSI is different from encoder
to encoder... even have a bug...(for example, in lame/mpglib/layer3.c,
it treat SCFSI as scalefactor = 0).
BTW, my new scfsi using algorithm is very simple. It provide only
little bit improvement, but never to loose the quality.
ISO code calculate total energy of an band and compare it with the
threshold, to see if we use scfsi. It needs some tuning of
threshold. And as it's very hard to find perfect setting of threshold,
we sometimes loose the quality when we use the scfsi.
see "recalc_scalebit" routine in my latest snapshot.
http://www.isoternet.org/~tominaga/lame-beta/lame-1121.tar.bz2
---
[EMAIL PROTECTED] // may the source be with you!
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
