[net.audio] Science and art

jj@alice.UUCP (06/01/85)

I've read several articles lately, by Craig Dory, Phil Karn, and
others, and I've noticed a strong theme throughout, that being
that audio is art, and that engineering isn't art.  I disagree
substantially with both, so I suspect there's some room
for discussion here...

<Enter reminisce mode, type 'n' if you're easily bored, yawn>

I started out in audio like a lot of people, namely, I listened,
I liked, and I bought. Then, I listened a little more, thought
a while, and built, eliminating the problems that I had bought
the first time around.  This went on for a while, and
led to my beginning to look for good models from control
systems (the mathmatical part, not the "concept") that would
describe some of the things I heard.  Needless to say,
the needs for math soon went far afield of control
systems.

To make a long story short, I worked on reinforcement,
recording (for a while), and decided that the entire 
schmeer was too opinionated (I grew excessively weary
of hearing "this is stupid, that is stupid" by all
the :"authorities") and too poor to make a good living
at, and started working on digital signal processing.

The applications were obvious from the start, but the
requsite convertors were out of sight, i.e. the designers
of the few 14 bit digital convertors (either direction) that 
were fast enough didn't even admit that "time behavior" was
a problem.  So, I worked on digital speech for a while, while
doing a lot of analog consultation (intra-company) for
various sorts on an informal basis.  

<exit boring mode>

Some of the things I've tried were tests on what sorts of distortions
were caused by what sort of signals in what sort of equipment.
This led to several designs of input modifying devices that were intended
to make one sort of equipment (triode, pentode, transistor) sound like
another.  For the most part, it was easy to make a GOOD (notice
the strong emphasis) transistor amplifier sound like any of the others,
by making various modifications of the input signal to the amplifier.
Such things as a transformer in parallel with a shunt on the
other winding, several diodes to give either center clipping
(of a very mild variety), a few class A transistors and
caps to do some slew-rate limiting, and so on, and the
deception seemed to work very well, i.e., I convinced
several people that the MacIntosh with nothing but the
heaters glowing was what they were listening to, that
it had a switch for the screen-grid proportioning circuit
that made it act triodish, and that I had switched the
switch, when the ony working amp in the room was a 
!!!Southwest Technical Products Tiger .01!!!
<well, actually a pair of them, the signal being stereo...>

So, I must say my faith in audiophiles, etc, sort of completely
evaporated. <These tests were made in school days, I could
make some better ones now, but I really don't have the time
to trash, and I don't seem to have very many audiophile
friends, either...>

When CD's came out, I listened to a few, and it was evident
from my digital (work) experience that the recording process
was crocked.  Not just badly done, totally crocked.  It was
also obvious that the digital medium didn't have the problems,
the recording engineer did.  (I meant to "" recording
engineer, oh, well, I don't want to call an editor at this
stage of the game...)  The various glaring problems wiht
miking, hall acoustics, etc, were almost painful, as was
the bad EQ. <I must admit I didn't know WHY the EQ was bad,
Bob Carver seems to have covered that one.>

I'll pass, for the time being, on the artificial enhancements
on vinyl that aren't on CD as a result of there being no
mastering process.


Anyhow, back to the premise, that audio isn't totally
art and engineering has artistic (in the sense of elegance)
facets.  

I claim that audio is a VERY inexact field, and that many
people have learned very well to listen differently, and
judge differently, depending on whether or not the
source is live or recorded.  This is due to the
various inexaxtnesses in the process, and so on.

Anyhow, it's clear to me, at least, that the question
is NOT "lifelike" reproduction, it's "reproduction
in the way that I'm used do", that drives MOST audiophiles.
It's also clear that such differences can usually be easily 
examined and captured, IFF one is allowed to make the necessary
measurements, and then goes away and analyze the results
and comes back to do some subjective tests.  The problem
is that EVERY setup has different variables <and I count
moving the speakers, etc, as variables here> and that
the evaluation for any one individual and system is 
good ONLY for that individual and system.  
Hence the conclusion that engineering can detect and
manipulate the effects heard by any given listener.
I certainly agree that the recording engineer,
etc, shouldn't interfere with the performance aspects!

(For example, a recent IEEE paper (I forget which society we sent it to)
<First author Cox, RV> that describes tests on several 
lower quality audio signals found that, in a small,
and QUITE simple test, there were several distinct
kinds of listeners.)

The conclusion that engineering has artistic possiblities
I think I'll leave, it's clearly less controversial in this
forum, I'm tired of typing, and this is too long already.
-- 
TEDDY BEARS HAVE LIMITED PATIENCE! THEY DO EVENTUALLY GET HUNGRY!
"What's the use of bearing bracers, hats or spats or shoes
with laces, or the things they buy in places down on Brompton Row?

(ihnp4/allegra)!alice!jj