Re: electroacoustics - rap to tap to zap to frap


Subject: Re: electroacoustics - rap to tap to zap to frap
From: Richard Wentk (richard@skydancer.com)
Date: Thu Aug 05 2004 - 16:53:40 EDT


At 15:37 05/08/2004 -0400, you wrote:

>And yet, if all the computer is doing is analyzing prior works and
>mathematically determining new ones, resulting in a printed page of
>music, I'm willing to be that everything could be done (much, much more
>slowly) by hand. The computer ultimately is irrelevent to the work
>aside from speeding up the process. So basically you're calling it
>'Computer Music' just because it's done faster by a computer.

No, because it's the algorithm used in the analysis that matters.

The fact that you couldn't do it by hand - because it would take too long -
is exactly the point.

I could sit here with a calculator and calculate sine samples, file them to
an ancient mag tape drive (if I had one) and then play it back through a
DAC. But that's not a practical way to make music. The fact that a computer
can do it much faster, perhaps in real time, with far more complex
algorithms than a simple sine, is what makes it a new way of making music.

>On the other hand, real-time FFT-based convolution *must* be done by a
>computer (or some piece of electronic hardware). And even if I use a
>handy-dandy little VST plug in to do it and used only stock presets, not
>programming a thing myself, *I* still have to select which sound sources
>are most effective in the medium for my particular aesthetic goal

That's the key point. With certain kinds of computer music - the kinds
we're talking about here - exploring what a new algorithm can do *is* the
main aesthetic goal, and the main point of the exercise.

With other kinds of EA the properties of an algorithm - which have probably
been explored by someone else first - are now taken for granted, locked up
in a black box and used as a tool. Just as use someone might combine a
flute and high string line for a certain effect.

Exceptional work succeeds on both levels. But each level on its own is a
valid goal. And they are easily and clearly distinct.

It's not even the tool, so much as how it's used and with what intent.
Using Max/MSP as a softsynth in a generic and unimaginative way obviously
isn't done with the intent to create an interesting new algorithm. It may
or may not produce a good piece of music, but if it does the intent won't
have been to explore new algorithmic possibilities.

Using it to parse and process some element of live audio, with some
intelligent and adaptive memory-like features, and more or less high level
processing of musical structures would suggest a different intent, that's
more obviously oriented in a different aesthetic direction.

Maybe I'm missing something but this doesn't seem like a difficult
distinction to understand. The fact that we use 'computer music' to cover
many different kinds of work is maybe what's confusing the issue here. But
while the term may be poorly defined, I don't think actual practice is.
Historically there are various pieces that set out specifically to explore
techniques like FM, granular synthesis and various kinds of algorithmic
event shuffling. Some may be more interesting to listen to than others, but
for all of them it's the fact that they showcased a particular approach
that was as important as what they sounded like.

You can say the same about tape editing and fugue writing, of course. Some
people innovate techniques, and some people master them. Some do neither,
and flap around producing copycat work that's neither interesting nore
memorable. A select few both innovate and master.

Whether or not there's a computer involved is more of a side issue. The
process hasn't changed, it's just a new tool on the scene.

Richard



This archive was generated by hypermail 2b27 : Sat Dec 22 2007 - 01:46:03 EST