Re: analog sound demonstrations

[James Beauchamp <jwb@xxxxxxxxxxxxxxxxxxxxx>]

Richard Parncutt wrote:
>I'm thinking of having an old-fashioned analog device made for teaching
>purposes. It would have a single loudspeaker and about 8 oscillators each
>with adjustable frequency and amplitudes (making, say, 16 knobs on the
>panel) and a master volume so the whole thing can be turned up and down. The
>idea it to easily demonstrate things like beats, roughness, masking, fusion
>of spectral components into complex tones, pitch effects and so on in
>lectures and labs, and have the students play with the machine themselves to
>get a feel for how things really work in complex tones. I prefer this
>approach to computer software because I couldn't be bothered setting up the
>right kid of computer and peripherals in various locations. And in any case
>I think the chunky portable analog approach might have pedagogical
>advantages.
>
>So my question is: Do other people want, or already have, similar devices?

I still use an old Putney EMS synthesizer from 1969 for class demonstrations.
It's quite a challenge to keep it in operation, however. I understand they can
still be purchased. (My last lead on that was Lawrence Casserley, Lawrence
Electronic Operations -Tel +44 1494 481381 -FAX +44 1494 481454, Signal
Processing for Contemporary Music -email leo@chiltern.demon.co.uk) Still, it
doesn't do all of things that Richard Parncutt would like.

Looking at my old files, I see that Starkey Lab Inc. (Minneapolis) advertised
a tabletop Hearing Science Laboratory containing 3 sine generators, a buzz
generator, a noise generator, and various filters. This sold for $5000 in 1979.
Pasco Scientic (San Leandro, California) offered a very compact model 9307
Fourier Synthesizer 9307 which synthesized 8 harmonics on 440 Hz with
individually controllable amplitude and phase for $540 in 1974.

It would be great to have a machine like Mr. Parncutt suggests; however, hardly
anyone is doing analog electronics these days. Also, a digital solution would
be much more accurate. The 16 knob outputs could be mutliplexed into a single
voltage feeding an A/D converter which in turn would drive a high speed CPU
(e.g., a Pentium) behind the scenes . . .

Perhaps the easiest approximate solution to Mr. Parncutt's quest would be to
buy one or more monochrome NeXTStations (current used market price approx.
$300). Free software exists for Fourier synthesis (e.g., Jean Laroche's
WaveFormEditor and Perry Cook's TwoWaves) to demonstrate many of the things Mr.
Parncutt mentions. Perhaps there are other auditory demonstration programs for
the NeXT or other computers that other people could mention.

Jim