.de PP
.sp 6p
.ti +3n
..
.nh
.sp 1i
.ll 6.25i
.in 6.25i-\w'Computing Laboratory'u
.nf
Computing Laboratory
The University
Canterbury
Kent\0\0CT2\07NF

Wed 18 Apr 1990
.in 0

Dear Ian
.fi
.PP
Hi.
Well, much has happened since we last met.  I took a job in Paris this summer,
at IRCAM, Pierre Boulez' underground music research centre by the Pompidou
Centre.  Unfortunately, my boss there was a paranoid maniac
(the words of one of the composers there, which seem to fit most closely)
so I have been back in England since September, trying to decide what to do
next.
.PP
We are getting to the point where there is enough processing power available
to make it realistic to synthesise sound, or process sound digitally,
in real time on more or less general-purpose processors.
.PP
I have ideas for a project I would like to try, to build a synthesiser
from more or less general-purpose processors.  Most currently-available
synthesisers give you a paradigm for generating sound, and then let the user
fiddle with the parameters.  With this device, you would build your own
algorithms for generating sound, out of oscillators, phase vocoders, filters
and whatnot, and the system would generate code and arrange to run it on the
processors being used.
.PP
The interface would be graphical: boxes for processing elements and lines
for signal (or control) paths, it would be able to take audio input directly
from analog-to-digital converters, or from CD-ROM, DAT etc, and control input
from sliders on objects on the screen and from MIDI controllers.
Ideally, you should be able to modify the patch while it is running.
.PP
I am working on how to do it technically, but would appreciate ideas on
the workings of business, so that I can think about ways of making it happen.
.PP
I wonder if I could bounce ideas off you some time.

Best wishes to all
