Re: Apple Computer changes its name

["Michael J. Mahon" <mjmahon@aol.com>]

mdj wrote:
> bieling@terra.es wrote:
>
> > mdj ha escrito:
> >
> >
> > > Just? You have to burn the tracks to an audio CD, which in the process
> > > loses the metadata (forcing manual re-entry) and causes considerable
> > > drop in quality.
> >
> > I'm not sure about this. I think that the product after
> > decompression/decodification (in order to play it) of a compressed (mp3
> > or whatever) music file is the same sample rate / bit depth as the AIFF
> > you burn into the CD, so where's the drop in quality ?
>
>
> mp3 (and similar) compression methods use psychoacoustic models to
> delete from the original source, information that is supposedly (or
> relatively) imperceptible to humans.
>
> Of course, this process is quite subjective, and there is a range of
> different mp3 encoders that perform the job to varying levels of
> success - although most are designed to perform best at either
> relatively high or relatively low bitrates.
>
> In many cases, just the decoding of the original encoded file results
> in audible artifacting, and these artifacts are present in the
> decompressed data you propose re-encoding.

Of course, unless some skullduggery is going on, the same artifacts
are present in any playing of the compressed file, authorized or not.

> Taking an audio file like the AAC encoded music on iTunes,
> decompressing it, then re-encoding with a different encoder will
> produce some "interesting" artifacts in the result.

Indeed, re-encoding material already compressed (and decompressed)
by another codec is virtually guaranteed to lose information, since
the designed losses of the second codec are not identical to the
designed losses of the first codec.

> You can test this yourself by doing a couple of transcode cycles using
> (insert arbitrary encoder). The results "speak" for themselves.

Some (not all) codecs are "information preserving" when re-encoding
information that they have once before encoded.  Of course, if any
change to the information occurred *after* the first decoding, then
further losses may result.

-michael

NadaNet networking for Apple II computers!
Home page:  http://members.aol.com/MJMahon/

"The wastebasket is our most important design
tool--and it's seriously underused."