naturerecordists
[Top] [All Lists]

Re: ATRAC compression

Subject: Re: ATRAC compression
From: Walter Knapp <>
Date: Thu, 11 Jul 2002 13:32:06 -0400
 wrote:
> 
> Music recording is definitely a major part of our recording business,
> although that is more in terms of rack mount products. In portables , we
> sell to many, many different markets, and the ones that tend to dislike
> ATRAC are the speech and language recording professionals, linguists,
> speech therapists, forensic scientists etc. particularly those who
> undertake detailed computer analysis, because any kind of compression can
> leave artefacts in the recorded material , not necessarily audible but
> which affect this kind of analysis work.  In Marantz's defence we have
> been supporting and manufacturing cassette portables for over 20 years ,
> and even if we do stop manufacturing MD at some point in the future, we do
> continue to support all of our discontinued products.

Over the last couple years I've had extensive contact with several
speech specialists. In all cases they were picking my brains about ATRAC
and the HHb Portadisc considering getting one for their work and
contacted me as I was one of the first ones with a Portadisc. Every one
eventually bought a Portadisc, and have been delighted with the
recordings. They have told me they have had no problems with analysis.
The only one I know to have problems managed to drop her Portadisc and
damage it. She did a lot of whimpering while it was off being repaired.
I think that these folks were previously using cassette recorders for
their work, though I did not ask.

The detailed analysis that these folks do is often not much different
than the sonograms we do. In fact most of the sonograms I've seen from
those groups look like pretty simple stuff which gives less info than
the stuff I use. Some of the other exotica they do can be really strange
and be dependent on custom, calibrated equipment and therefore would not
use any general purpose recorder. Like any group they get stuck on
favorite equipment and techniques, often attributing special abilities
to it that are not so special.

I think ATRAC does so well because it does not try to bring selected
samples through, but synthesizes a whole new audio track from what
amounts to control data. That track is very little different from any
other D/A converter output. Many people get trapped into thinking
there's missing samples, and it does not work that way. You get a full
set of samples, they just are not the original ones, and may differ
slightly in their values. For most analysis software this presents no
problem, and in terms of numeric analysis is generally within the range
of error of the mic's and pre's used. The problems I've seen have mostly
been ones of clock sync, one of the new problems with the advent of
digital sampling. And that's not really a ATRAC problem.

I would expect that you have a stock of repair parts and would continue
to do repairs as long as the stock remains. I've not heard of any
manufacturer restarting the manufacture of parts just for repair stock,
so there's bound to be a limit. What sort of timeframe do you plan on as
far as parts for discontinued equipment?

Walt



________________________________________________________________________
________________________________________________________________________

<Prev in Thread] Current Thread [Next in Thread>
Admin

The University of NSW School of Computer and Engineering takes no responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the naturerecordists mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU