[sdiy] new project in the works

[Fredrik Carlqvist]

Levels will be quantized to 2^N levels, where N is the number of bits
remaining in the mask. For example, anding the A/D result with 0xE0 will
give eight levels. If the sampling rate is kept at 36kHz, no time
alaising will be noticable.

I used it directly as an audio processor. The alising was what I was
after. I wanted to be able to alias the signal in both space and time
domain, therefore I added the sample rate control.

With four bits of sampling and a lower frequency (perhaps 8kHz), I get
something that sounds like a toy. With two bits it's a really cheap toy.
At eight bits and 36kHz, the effect is barely noticable. Changing the
sampling frequency on the fly will give bell-like soft ring modulation
effects, not as extreme as an analog multiplier but very useful. 


Fredrik C



-----Original Message-----
From: owner-synth-diy at dropmix.xs4all.nl
[mailto:owner-synth-diy at dropmix.xs4all.nl] On Behalf Of Czech Martin
Sent: Wednesday, May 28, 2003 14:59
To: Fredrik Carlqvist; synth-diy at dropmix.xs4all.nl
Subject: RE: [sdiy] new project in the works


I think the original idea was to quantize levels,
but not time. So no alias effects.
But perhaps other time related effects (e.g. slew rate).

How can you describe the effects you got
without much alias involved?

m.c.

-----Original Message-----
From: Fredrik Carlqvist [mailto:ifrc at iar.se]
Sent: Mittwoch, 28. Mai 2003 14:42
To: synth-diy at dropmix.xs4all.nl
Subject: RE: [sdiy] new project in the works



I built one of those some years back with a different approach. I used a
PIC16 to sample the incoming signal to 8 bits, then mask the result to 1
through 8 bits. I could also control the sampling rate from 36kHz down
to 1kHz. This gives a ring modulation effect due to the Nyquist effect.
The two parameters (bit count and sample rate) were set by two voltages,
directly sampled by the PIC. So you could call it voltage controlled. 

Fredrik C