[sdiy] Uniformly distributed noise generator?

cheater00 . cheater00 at gmail.com
Mon Jun 10 16:06:29 CEST 2013


On Mon, Jun 10, 2013 at 3:15 PM, Scott Gravenhorst <music.maker at gte.net> wrote:
>
> Tom Wiltshire <tom at electricdruid.net> wrote:
>>Hi Damian,
>>
>>On 10 Jun 2013, at 12:37, "cheater00 ." <cheater00 at gmail.com> wrote:
>>
>>> - analog noise is truly random, not pseudo-random
>>
>>In theory. In practice, analog noise is unlikely to be totally
>>pure, and a good digital implementation would give better
>>quality.
>
> I agree with Tom.
>
> It has been sooo long since the implementation of the chuff chuff LFSR with only a
> few bits and low clock rate that it is truly a ridiculous comparison to analog noise

That's not really good enough in my application. This is for a
measurement fixture. I need true, never-repeating randomness. If my
tests keep on repeating the same lie over and over, then I'll be
forced to believe them. It's not inconceivable that the specific
random sequence which happens right after power-on will in some way
change the result of the test.

I agree that this never ever matters in music and isn't even a consideration.

> I Am curious to know why you want this particular distribution.  What is the
> application if I may ask?

I would like to build a statistical time base plugin for a Tek 7000
oscilloscope in order to experiment with Monte Carlo style
measurements.

Another use for this would be as a time base for a scanning electron
microscope (SEM). Raster scanning (like in a TV) means that the full
intensity of the beam is focused on a single point for fairly long,
which ends up destroying the specimen. You want a random scan, and you
want it uniformly distributed otherwise you can't see anything other
than the centre point. An LFSR would be suited here, but if I have the
above then I can use it here as well.

Cheers,
D.


More information about the Synth-diy mailing list