naturerecordists
[Top] [All Lists]

Re: ATRAC don't get no respect

Subject: Re: ATRAC don't get no respect
From: Dan Dugan <>
Date: Thu, 24 Apr 2003 12:06:48 -0700
Walt, you're mis-using terminology here, and I feel the need to 
explain for those who are here for education:

>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.

By "lossy compression" above, you seem to be meaning "imperfect." 
This imprecise way of speaking is a bad thing to do in an educational 
forum. "Lossy compression" refers to certain psychoacoustic data 
reduction techniques, and it's confusing when you use it to describe 
the other losses that happen in recording processes.

Digital recording on DAT or hard disk is imperfect to the degree that 
it is limited by the quality of the converters, the 16-bit dynamic 
range, and 44.1 or 48K sampling rate. With today's converters, all of 
these limitiations are of little consequence. Many of the finest 
classical music CDs you buy were recorded on DAT! The problem with 
DAT is reliability, not fidelity. Most recordists are changing to 
hard disk recording because they can use higher standards for 
conversion (24-bit, 96K), always desirable in master recording, and 
it's more reliable.

Those who are into the engineering know that a DAT recording system 
is constantly making errors, and correcting them completely by using 
a redundant data coding scheme. This process is invisible to the 
user. The result is perfect up to the point where there are too many 
errors, and failure. The same thing goes for CD reproduction, with 
the addition of a "gentle" first failure mode called error 
concealment.

Lossy compression systems (ATRAC, MP3, Dolby Digital movies, the new 
AAC) are a whole different kettle of fish. They analyze the sound, 
record the analysis parameters, and reconstruct it from that, rather 
than recording the straight digital audio samples. It's unfortunate 
that the term "compression" seems to have stuck, because it makes it 
difficult to explain to newcomers that the word now means two 
completely different things in audio engineering, depending on the 
context!

-Dan Dugan


________________________________________________________________________
________________________________________________________________________

<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