| Ideas are definitely poppin' tonight.
Thanks Ross!,
Your thoughts definitely tie in with a current spatial sonic imagery project
here.
I wonder if this topic also relates to Terrain Mapping Synthesis. . . .in a
way ;-)
===
Found this recording at an internet CD warehouse on:
J. Cage - The 25-Year Retrospective Concert of the Music of John Cage
WERGO (GER) label 1994
01. Short Inventions (6)
02. First Construction (in Metal) for 6 Percussionists
03. Imaginary Landscape no 1
04. The Wonderful Widow of Eighteen Springs
05. She is Asleep
06. Sonatas and Interludes for Prepared Piano: Excerpt(s)
07. Music for Carillon no 1
08. Williams Mix
09. Concert for Piano and Orchestra
MadGello
Out
-----Original Message-----
From: owner-csound-outgoing@maths.ex.ac.uk
[mailto:owner-csound-outgoing@maths.ex.ac.uk]On Behalf Of Ross Bencina
Sent: Monday, March 22, 1999 10:52 PM
To: csound@maths.ex.ac.uk
Subject: Re: Cage's Williams Mix - an alternative approach
>I'm having a bit of difficulty thinking of a way to emulate how audio
>signals as laid on magnetic tape could be manipulated in the same manner
>using digital tools. I suppose one could lay the waveform down in a table
>and rotate the axis of reading in a graphical manner to some extent. I'll
>leave the hypothetical aspects of that discussion to those that are more
>knowledgable of the physics involved. ;-)
Because the waveform on tape is stored as a fluctuating magnetic field
representing _amplitude_, rotating has the effect of:
1. changing the playback speed - but I would also think it would smear the
high freqency information, so probably a table lookup combined with a simple
low pas filter would do the trick. If you think about the tape as a series
of vertical bands, darker ones meaning louder energy, then rotating the tape
will create diagonal bands. Since the heads are designed to read vertical
bands, multiple bands will be read at once. In the extreme case or 90 degree
rotation, a lot of bands will be read at once and you'll just get a click
relative to the avaerage level of the bands. To relate this to digital,
imagine looking at the waveform display in your sound editor from above -
you would be looking down at a corrugated surface, the closest parts of the
surface are loudest, these would be "darker" or "heavier" (more charged) on
the magnetic tape - hence my bands metaphor.
2. making only a segment of the tape accesible to the playback head, hence
creating a fade in and fade out effect. The spliced in tape would form a
diagonal strip accross the final tape, the thiner the strip, the quieter it
would be, so as the strip enters, crosses the final tape, and then leaves at
the other edge, it would fade in and then out again.
If the tape is stereo the signal will pass from one head to the other
creating some kind of panning.
I don't understand the physical details of constructing Williams mix enough
to say much more than this. If you wanted an "Authentic" version you would
need to know the amplitude characteristics of magnetic playback heads when
the tape is misaligned. A recent discussion on cec-discuss suggested that a
lot of artifacts(amplitude glitches etc) are introduced by the phisical
distortion created my tape splicing.
I don't think Cage would have a problem with performing spectral rotations
with a graphical tool though (it would sound different) - the main issue
would be that rotating magnetic tape creates a band that fades in and the
fades out again, you would have to apply that separately as an envelope - It
would be nice to do this in a spatial manner and have the sounds approach
and receed, as if we were sitting on the playback head.
Ross.
|