Path: utzoo!attcan!uunet!midway!ncar!zaphod.mps.ohio-state.edu!uwm.edu!csd4.csd.uwm.edu!info-high-audio-request
From: drm2@mvuxn.att.com (David R Moran)
Newsgroups: rec.audio.high-end
Subject: Re: 1/3 Octave EQ query
Message-ID: <7389@uwm.edu>
Date: 2 Nov 90 13:52:33 GMT
Sender: news@uwm.edu
Lines: 29
Approved: tjk@csd4.csd.uwm.edu
Originator: tjk@csd4.csd.uwm.edu

In article <7360@uwm.edu>, erich@inmet.inmet.com writes:
> 
> Is anyone aware of a 2 plus channel 1/3 Octave EQ with spectrum analyzer.
> It would be desirable to have both analog and digital inputs and outputs
> and I would prefer digital rather than analog internal circuitry.

Why, for heaven's sake? It is probably possible to simulate in software
and with DSP all of the work done in a good 1/3-octave RTA, but a
good Reticon filter bank built around a PC (as in the dbx RTA1, now the
Sound Technology 4000 RTA) or equivalent hardware architecture is 
much more than accurate for 1/3-octave work, at least with pink noise.
And good analog constant-Q 1/3-octave equalizers are available...
Maybe there are other arguments for digitizing this kind of analysis
but I cannot think of what they are.
There are some arguments for digitizing equalization, but the usual
argument is that 1/3-octave is not fine enough (it is, but you can't
convince anyone who believe otherwise). The "phase" arguments usually
show that the arguer does not understand what happens in the so-called
time domain when transducers play back within an enclosed space. But 
there is some validity to doing EQ digitally.

Analog is very, very good for many thing in audio -- quiet, versatile,
easy to manipulate, sensibly expensive or inexpensive, etc.; many things 
that digital is not, or not always, or not easily. If this question 
arises from an uninformed pro-digital bias (and it may well not, it may 
have sounder reasons behind it that I am missing), I would point out that 
sometimes it is useful, whenever you see the word digital or thrill to 
using it yourself, to substitute "numerical" instead. As in numerical-ready 
speakers, or in this posting, numerical inputs and outputs...