back to list

FANS of ANS

🔗jpehrson@rcn.com

4/11/2001 6:22:56 PM

Jay Williams sent the following conjecture regarding the ANS
synthesizer in Moscow... probably he's figured it out!

Well I confess, when I read Anton's article, I understood the
principle of the ANS but didn't have any physical model of the thing
in my head. Your questions prompted me to deduce just how this
puppy's constructed and here's how I think it must look:

So, we have this rectangular glass plate, one side of which is
covered with this mastic goop. Let's assume that, if standing in
front of this machine, this "score" plate goes lengthwise from left to
right and lies horizontally. Now, below it is a light source, the
mastic surface is on top. When it's ready to be played it is
positioned so that immediately above it is an opaque plate of metal
or
wood with a narrow crack at the left end that serves as the
light-window. It runs from front to back so it extends the full width
of the "score" plate. Above that crack are the 5 wheels he mentioned,
and these are arranged in a column corresponding to the length of the
light-window. Each wheel,in turn, is made of glass and also coated
with mastic. The 144 sound tracks on the wheel are each a sine wave
drawn in the mastic so light can get through to a photo cell. Thus
you
have 144 undulating tracks like the end grooves on lp records with
the
higher pitches at the outer edge and the lower ones toward the
center.
That's why he refers to them as "phonograms." These wheels are always
turning at a constant speed. If I were building it I'd have a
variable
spped motor running them so I could change the "Pitch Standard" or
create mass glissandi without having to do freaky things with the
drawings which, in any case, would produce a graduated, rather than a
continuous gliss. Now, you can draw your sounds on the "score" plate
and a rack-and-pinion moves it from right to left at whatever speed is
right for the tempo of the piece. I assume that motion as he's got
all those coders and cutters at the left end so the score would
likely start there. So, you draw the score and then lower the rest of
the mechanism so the plate is not subject to outside light. I'd
think
some sort of rulers would be essential. This puppy's botta be of
some size. If we assume a millimeter for each of those 720 sound
tracks, we have this glass plate that must be about 30 inches wide
and lord knows how long in order to accommodate a 5 minute piece!
Think what we could do with a modern version using a cd-type
principle and computer-generated drawings. That way, you could have a
roll of plastic on which you'd have ample width for a drawing and
more length for the piece. But anyway, what a damned clever
invention.

Jay Williams

________ ______ _____ ___
Joseph Pehrson