"Cat Bag" – a piece for electronic recording based on electronic keyboard by Greg
Fox, 17th of January 2005

Equipment: electronic keyboard, 2-note polyphonic with mono output – 8 voices,
presumed FM, 8 programmed rhythm-loops with tempo control, four custom drum
sounds playable by the user, pitch-range 3 octaves, non touch-sensitive; amplitude
control with three settings. Unknown make, presumed Asian origin.

Procedure: 12-note row : d g e g#c b c#a a#f f#d#
Undergoes same procedure as "New Miracle Scenes" to produce:

d b e c#f#f d#a a#g#c g
a d f g b c c#e g#f#d#a#
f#g#d c a#a f b c#g e d#
c#d#a#d c g#b g e f f#a
e a c a#d d#g f f#b c#g#
g#a#b f g d e f#d#c#a c
g c#g#a d#f#d a#f c b e
b f#a d#g#e c d g a#f c#
(last four permutations not used as there are only eight 'instruments')

Each instrument improvises based on the raw permutation data plus tonal filters
thereof, for example permutation one, D Major = d b e c#f# a g

This material is recorded to PC and saved.

The master track starts off with a bass-line made up of the pre-programmed rhythm
section of the keyboard with tempo and beat changes designed for maximum
instability. This forms the basis for the piece. It also contains the pitch-classes of the
original row, and therefore lends at least a conceptual/internal consistency to the piece
from the very beginning, given that everything else is derived from this material,
albeit in (most likely) an imperceptible way.

The other material is either pasted direct to the master track, or is first subjected to a
random "action-painting" cut and paste operation to move parts of the track forward
and backward in the section. The over-riding principles are of breaking up silent
passages and of breaking up long notes, so a cut could be made to start in the middle
of a longish note and end in the middle of a rest. It would then be pasted either nearby
or far-away depending on the whim of the operator according to action painting
stylistic considerations. (What I mean is that randomness is emulated rather than
contrived, and so what we see is a human impression of what randomness might be
like – ie. pieces moving in extreme ways, avoidance of obvious pattern, etc. etc.)

Once the cut-up is performed (or not bothered with), the tracks are overlaid onto the
master track. Again this is performed with an action-painting style approach, and with
consideration to the dramatic use of the left and right channels to imitate ensemble
playing.

The overall track was listened to after each large-scale paste operation, and decisions
were made as to what might happen to it next, bearing in mind that essentially this
additive process happened vertically not horizontally (though the duration grew from
13 to 21 minutes so some compromise was required to avoid contrapuntal overload).

The final element is the ghastly demonic voice of the ZX Spectrum's "Currah
Microspeech" unit repeating the title with increasing unpleasantness and increasing
hesitance between the two words (with throw-back moments of tick-like hurry).
Most aesthetically obvious in this relationship is the point near the end of the piece
where the robot voice interrupts loudly when the rest of the virtual "ensemble" (of
one) are being uncharacteristically calm and understated.

The piece was written in one evening on the 17th of January 2005 with only the minor
influence of alcohol towards the end (20cl of standard vodka at 37.5% alcohol by
volume = 7 units).
It is dedicated to Melloney Robinson, who bought me the keyboard as a Christmas
present.

Concept: In terms of links to the wider world and its ideas, I suppose the main link is
to the kind of technological nostalgia we're starting to see since the 90s really; the
idea that instead of an inner yearning for live music or pre-electronic working
methods, or times when children played in the street, instead of all that, we yearn for
radio, for old computers, for times when computer games were less realistic and
technology was less all-powerful. Musically this yearning for "the retro" hasn't much
happened; there's always been the Harry Partch approach of using non-musical items
for musical purpose, carried through the whole John Cage movement, but this is a bit
different and is about purposefully using more "complete" technology (software
capable of moving sound around and reconstructing it instantly, and of mixing layers
together in whatever way the user wants – I'm not assuming it can't bear
improvement, but am saying that for the purpose it has no practical limitations;
something analagous to a modern camera being used to take a photo at a wedding) to
'play with' the older technology (though the keyboard in question is 'retro' in its
functionality mainly due to cost considerations rather than its year of production) and
make it do new things.
The use of the voice of the ZX Spectrum highlights this point very well – the
spectrum could be used as a sampler by the use of its low-bandwidth input socket
used usually for reading data from audio tapes – but the sample-rate is something like
8kb of data per second (however many samples that actually is per second)…. and
with the processor running at (approx.) 10Mhz, instead of the (low by 2005 standards
where average is around 1.5Ghz) 533Mhz of my home PC, the spectrum would have
"struggled" to have performed the FFTs used on some of the instrumental tracks to
compensate for my poor playing, and would have "struggled" to perform all the mix
pasting, let alone handle the 234Mb waveform using its casette interface that stored
data at around 8 minutes for each 100kb of data. (My figures here aren't 100%
accurate but illustrate the point.)

So a kind of 21st century "neo-classicism" – old technology being recycled using
modern technology and implemented using techniques created for a different purpose
in the last century and long since disregarded by the majority. But how "neo-
classical" does "Cat Bag" sound?

Delivery: "Cat Bag" is designed to suit a standard audio compact disk (or equivalent,
eg. mini-disk, DVD) or delivery via electronic means in mp3 format. There is no
ideological opposition to the idea of reducing the quality of the samples but this
should be reflected in the title of the performance/file. (eg. Cat Bag 16khz/8-bit
mono)

At present a standard mp3 of the piece is a little large for a dial-up connection and a
little slow on an average 'broadband' connection, but as this is what I'd deem
"incomplete" technology (it imposes limitations on the user), I don't see this as a
problem when defining the piece as apt for electronic deployment.