"Steven Sullivan" <(E-Mail Removed)> wrote in message
news:(E-Mail Removed)
> out-(E-Mail Removed) wrote:
>> The current stereophile online has an article of the
>> subject line. It addresses in part the question of if
>> it is possible to distinguish 320 and cd sources.
>> Reading between the lines me thinks the owner of Naxos
>> really does not think one can but bows to the author of
>> the article so as not to dissuade readers of the mag
>> from becoming customers. He wants readers to consider
>> themselves the 2 percent, which is part of the self
>> defined and self admiration fringe of the hobby with the
>> golden ears of course. Here is the relevant part:
>
>> Even Naxos founder Klaus Heymann acknowledges the
>> inaccuracy of the claim, in a Naxos press release,
>> that, at 320kbps, "speed and quality make MP3 files
>> indistinguishable from CDs."
>
>> "To the average consumer," he told me via telephone
>> from his home in Hong Kong, "320kbps is virtually
>> indistinguishable from CD sound. You would be
>> surprised how many people listen to 128kbps
>> selections from the Naxos Music Library played back
>> through my computer and hi-fi system and ask, 'Where
>> is the CD player?' 128kbps is not wonderful. The
>> overtones are missing, and the bass is not well
>> defined. But at 320kbps, I believe that 98% of people
>> think it sounds like a CD. I personally think it is not
>> exactly like a CD, but it sounds damn good."
> All Mr. Heymann would ahve to do to cap this, would be to
> present some ABX test result.s
If he did, I suspect he'd characterize the audible degradation differently.
The statement that the overtones are missing is IME completely bogus.
Perceptual coding does not remove audible overtones, as a rule they are
among the sonic entities that are preserved the best. Again, bass defintion,
requiring a relatively low data rate to reproduce effectively, is unlikely
to be harmed. The usual artifacts of audibly inaccurate perceptual coding
take the form of a loss of articulation in the middle and high frequencies
during complex passages.
Even when there are audible problems, almost all of the musical work is
audibly accurate.
> And even Mr. Heymann shoudl realize that there is no such
> thing as '128 kbps' monlithically; there are different
> codecs that can compress to 128kbps
> with different levels of success, also dependent on the input music.
The myth that is purpounded here is that all perceptual coders operating at
a certain bit rate sound the same.