[FFmpeg-soc] ALAC Benchmarks

[Michael Niedermayer]

On Wed, Aug 20, 2008 at 10:22:57PM +0530, Jai Menon wrote:
> Hi,
> 
> On Wednesday 20 Aug 2008 9:33:25 pm Guillaume POIRIER wrote:
> > Hello,
> >
> > On Wed, Aug 20, 2008 at 5:15 PM, Jai Menon <realityman at gmx.net> wrote:
> > > Mike, did you by any chance get time to do the benchmarks with the alac
> > > encoder in FFmpeg-svn? I was thinking of posting a news item on
> > > Hydrogenaudio and it would be nice if I could link to that spreadsheet. I
> > > would have tried but I don't have the samples and know how of the
> > > methodology ;-). Thanks in advance.
> >
> > I have a question about your ALAC encoder: since this is a lossless
> > compression, is it safe to assume that no listening test is needed to
> > compare your encoder to other encoders?
> >
> > Guillaume
> 
> Yes. The decoded output is bitexact to the original source. You could do ABX 
> tests to convince yourself though :-)
> And please do not pay heed to the doubting hordes at Hydrogenaudio who say 
> that the encoded audio "lacks bass" and assorted nonsense :-) [1].

Actually it is possible in principle to hear a difference between lossless and raw
because the cpu may need more power or may change its power
consumtion in a different pattern depending on what is run on it.
If now the DAC and amplifier run on a common source of power/ power supply
then given low quality DAC, amplifier and power supply and high quality
headphones or speakers iam pretty sure it could be audible. ...

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

I hate to see young programmers poisoned by the kind of thinking
Ulrich Drepper puts forward since it is simply too narrow -- Roman Shaposhnik
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-soc/attachments/20080820/51a1c591/attachment.pgp>