[sdiy] preferred noise source circuits

[David G. Dixon]

This has been a pretty interesting thread on noise so far.  I'm still
deciding what to do, although I will most likely use either an analog source
(tranny with reverse-biased e-b junction) or something digital but
simplistic using counters.

One thing I came across lastnight which I found intriguing was the "shimmer"
generator on Ken Stone's Cynare drum synthesizer:

http://www.cgs.synth.net/modules/cgs47_cynare.html

This is basically six schmitt trigger oscillators at different frequencies
multiplied through four exclusive OR gates (basically, a 40106 and a 4030)
and averaged at a divider.  I gather that he uses it to provide a shimmering
sound like a cymbal.  I've simulated it and the output is pretty "noisy".
One could replace the feedback resistors in the oscillators with pots for
all sorts of interesting effects.  Also, if the oscillators were slowed
down, this looks like an interesting source for random gates and triggers.
Finally, by taking the output (which is 10Vpp) and dropping it 5V through an
adder, one gets a pulse wave with pseudo-random +/-5V positive and negative
pulses.  Hence, this would make a pretty cool kind of LFO, especially if one
put an integrator on it to generate random triangles, and polarity switches
to rectify the output in either direction (my triple LFO has this feature,
and I have found it to be extremely useful).

Has anyone else played with this shimmer generator?



> From: synth-diy-bounces at dropmix.xs4all.nl [mailto:synth-diy-
> bounces at dropmix.xs4all.nl] On Behalf Of Adam Schabtach
> Sent: Friday, June 26, 2009 12:08 PM
> To: 'sdiy DIY'
> Subject: RE: [sdiy] preferred noise source circuits
> 
> It's also far simpler to generate mathematically correct pink noise with a
> microprocessor (i.e. with software) than with analog circuitry. I learned
> this years ago when I wrote the pink~ object for Cycling '74's MSP.
> Unfortunately I don't have any references handy to cite, and I suppose it
> would be a violation of their copyright for me to post the source code
> here
> (if I still have it, which I'm not sure that I do). Google should help the
> curious as always, though.
> 
> So, I'm in agreement with Tom. You could build a dandy white and pink
> noise
> source with one microprocessor. OTOH if hardware's your thing and software
> isn't, then you're welcome to use many more parts to create an
> approximation
> of the same result.
> 
> --Adam
> 
> > -----Original Message-----
> > From: synth-diy-bounces at dropmix.xs4all.nl
> > [mailto:synth-diy-bounces at dropmix.xs4all.nl] On Behalf Of Tom
> > Wiltshire
> > Sent: Tuesday, June 23, 2009 7:55 AM
> > To: Amos
> > Cc: sdiy DIY
> > Subject: Re: [sdiy] preferred noise source circuits
> >
> > Well, if we're talking about "white" noise, then I mean noise
> > that is genuinely white - i.e. has an equal distribution of
> > energy across all the frequencies of interest (the audio band
> > in our case). That's not really subjective at all. Likewise
> > if you're generating random pulses or gates, then "better
> > quality" means statistically indistinguishable from random.
> > You don't want any unwanted weighting creeping in unless you
> > put it there.
> >
> > Obviously whether white noise is the best kind for many
> > synthesis purposes is a much more subjective issue. Perhaps a
> > 6dB rolloff of the high end makes a better audio noise source
> > since it doesn't sound so harsh. If you've got good quality
> > white noise to start with, then you can choose what you do to
> > it. If it's got it's own flavour already, it's much harder to
> > remove that.
> >
> > T.
> >
> >
> >
> > On 23 Jun 2009, at 12:38, Amos wrote:
> >
> > > On Tue, Jun 23, 2009 at 6:09 AM, Tom Wiltshire
> > <tom at electricdruid.net>
> > > wrote:
> > > "a digital noise source with a simple 8-pin DIP microprocessor that
> > > will give you much better noise quality than most analogue noise
> > > sources."
> > >
> > > see now, I don't mean to stir the pot, but "much better
> > noise quality"
> > > seems pretty subjective to me.  This is what it (ought to,
> > > IMO) come down to when choosing your noise source.  What
> > qualities do
> > > you evaluate when determining "goodness" of noise?
> > > This is more an open question for anyone who has an opinion on such
> > > things.
> > >
> >
> >
> > _______________________________________________
> > Synth-diy mailing list
> > Synth-diy at dropmix.xs4all.nl
> > http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
> >
> 
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy