--- In John Campbell
<> wrote:
> >John Campbell wrote:
> > > Walter Knapp wrote:
> > >
> > >
> > >>oryoki2000 wrote:
> > >>[edit]
> > >>
> > >>>This is because ATRAC will always be
> > >>>criticized for creating, by design,
> > >>>an approximation of the original source.
> > >>>Saying the result is "indistinguishable
> > >>>from the original" doesn't change the
> > >>>fact that 80% of the original digital
> > >>>material is discarded, and other data
> > >>>added, when creating the ATRAC version.
> > >>
> > >>As I've noted it's a error to think ATRAC even keeps 20% of
the original
> > >>data. The more I dug into it, the more it became clear that it
should
> > >>rather be thought of as synthesizing a whole new set of data,
and
> > >>storing instruction on how to do this. Yes, a approximation,
but so is
> > >>anything that comes out of a A/D - D/A cycle.
> > >
> > >
> > >
> > > Maybe the reason you have to keep preaching this gospel is due
to
> > > your wilful mis-reading of other's comments. The original
statement,
> > > and those preceding it, are figurative - you are being literal
(or
> > > trying to be, at least). You seem determined to hide from the
fact
> > > that ATRAC and other audio data reduction systems are not
called
> > > lossy compression for nothing.
> >
> >Or maybe I understand how others will read it. Posts into the
internet
> >are near immortal. I look at these statements and ask how all
possible
> >readers will interpret what's said. What was said contained a
error that
> >leads to false conclusions.
>
> A couple of people used imprecise terminology, but to say there
was
> "a [sic] error" is somewhat mis-stating the case. Data reduction
via
> perceptual encoding is achieved through requantisation of the
> original data. That means reduction of the number of bits, or
what
> some might think of as a process where "material is discarded" or
> "only 20% of the original information" is saved. Those statements
> are vague, from a scientific perspective, but so too is your
> "synthesizing a whole new set of data". Not exactly a textbook
> explanation, is it?
>
> >
> >DAT is lossy compression, hard disk digital is lossy compression,
every
> >single digital sampling method for sound is lossy. The question
is not
> >is it lossy, they all are. The question is what does that mean.
Saying
> >it's lossy is just the beginning, not the end of the discussion.
>
> DAT and HD are not "lossy compression" at all. Having captured
bits
> via analogue to digital conversion, they do not then purposefully
> reduce the bit-rate through requantisation - and that is, I
repeat,
> what lossy compression means. So the issue IS that ATRAC is lossy.
>
> I recommend chapter 3 of Francis Rumsey's "The Audio Workstation
> Handbook" (Focal Press) to anyone seeking clarification on audio
data
> reduction.
>
> John
In a serious discussion it would also be good to use citations of
others comments in a fair manner. It would be problematic to pick
out single phases as "only 20% of the original information"
or "material is discarded" out of a larger context, where much more
precise explanations followed.
Terminology is always difficult to judge. Each of us has a different
background. When I used the term "information" I thought on
communication engineering. In that context, this statement is
absolutely not vague (all the bits stored in a binary file do carry
information).
Also, the short citation "material is discarded" taken from oryoki's
posting was isolated out of this context :
"Saying the result is 'indistinguishable from the original' doesn't
change the fact that 80% of the original digital material is
discarded, and other data added, when creating the ATRAC version."
So, it is important to note, that the word "digital" preceded the
citation. The complete sentence was absolutely correct from the
scientific view (this is my personal impression only).
As we can see, there are not only those psycho-acoustic effects in
human perception - there are also other, much more subjective
perception issues in speech when we are discussing a certain issue...
Raimund
________________________________________________________________________
________________________________________________________________________
|