Wiardgroup

Hi Mathieu
after reading Prof. Richter's comment (below) a sensible clarification
 occurred to me. We have been discussing noise in two contexts:
1) as an AUDIO signal
and
2) as a source for derivation of CONTROL VOLTAGES (ie via a sample/hold)  

and as i read back over the discussions i notice that i jump back and
forth between these two contexts very freely and this is prone to
cause confusion. So please allow me to restate a couple of points clearly.

As an AUDIO signal, analog and digital noise are different but to my
ear, the difference is not well defined and with processing the two
forms of noise can 'cross-dress' pretty convincingly (at least in a
dark bar when i'm drunk)
;'>
As the professor points out, digital noise has sonic parameters that
are easily changed (especially on the N.R.) and shifting these
parameters will change the AUDIBLE characteristics of the noise quite
a lot. But i would add that Analog white noise ('hiss') can be changed
quite easily too with processes like filtering and balanced modulation.
The point is that in the AUDIO realm the difference can be very blurry.

To my ear, the most apparent differences between analog and digital
noise are exposed when the two forms of noise are sampled by a
sample/hold circuit and output as patterns of control voltages. That
context is what yields the 'character' difference to which i was
referring in my earlier posts.

-doc

PS: i want to state for the record that Grant & i did NOT collude on
referring you to the Federal Scientific chart on musicsynthesizer.com
. It's just that little chart is such a gem in it's succinctness and 
thoroughness that it's hard to beat as a resource.... Kudos to Grant
for noticing that chart in the back of some cast-off technical
reference in  the 2nd-hand bookstores he haunts, and giving it a new
lease on life as a billboard on the information highway! Bravo Professore!


--- In wiardgroup@yahoogroups.com, "Grant Richter" <grichter@...> 
> Because analog and digital noise generators sound different. Analog
noise is "hiss" and has a 
> gaussian level distribution. Digital noise can sound like "hiss"
under some circumstances, but 
> can be changed a lot, it has a uniform level distribution.
> 
> Here is a chart from Federal Scientific, which shows statistical
properties of waveforms 
> including analog and digital noise:
> 
> http://www.musicsynthesizer.com/Circuitry/probability.htm
> 
> The Noise Ring outptus waveforms 6 and two different versions of 10.
The "Change" control 
> essentially controls auto-correlation.
>