[sdiy] Sample&Hold thoughts

[Happy Harry]

You will have a problem with aliasing, the undersampling of
the audio will make weird difference tones. think ring modulator
type sounds...

I'd use a good parallel A/D converter (12-16 bit), clocked at maybe a fixed 
44khz rate... then feed it to a D/A converter (12-16 bit) but
add switches to selectively disable bits starting with the LSB...

so its 12bit, 11bit, 10bit, .... 4bit ?

That should do what you are trying to do in real time.

A digital delay (an OLD one) could probably be hacked to do this.
Don't try with a DeltaLab (1 bit) or a modern one using DSP...

H^) harry


>From: Plutonique9 <Plutonique9 at symbiocom.com>
>To: <synth-diy at node12b53.a2000.nl>
>Subject: [sdiy] Sample&Hold thoughts
>Date: Thu, 1 Mar 2001 06:00:45 GMT
>
>Hey there people,
>I'm sure this has come up before in the history of sample and hold
>circuits, though I haven't read about it anywhere, the idea has been
>knockin' around my head for the last couple of days.
>
>Ok, normally S&H's derive their sample voltage from a white noise
>source, or a LFO as with my EML. The max clock speed on most synth
>S&H's is probably around 20hz-30hz. In this fashion, they are
>normally used to add random deviations to VCF's and VCO's, I like
>them a lot, especially if they offer an external trig. input :)
>
>What I'm wondering now is, is it possible to clock sample n' holds at
>extreme clock speeds, say around 11khz-22khz, or even higher.
>I guess you know where I'm going with this.
>
>I'd like to hear the effect of using an audio source, like a drum
>loop or a voice for example, as a "sample" source for the S&H. You
>know, in place of where white noise or an LFO is usually used.
>Now if you clocked the S&H at a high enough rate, anywhere between
>4khz up to 48khz, or something, what you would hear at the output of
>the S&H would be the audio source, drum loop or whatever, the only
>difference would be that you could change the fidelity in real time
>over an extremely wide range by changing the clock rate, right?
>
>I'm just drooling over the thought now of "Sample rate modulation",
>say if you modulated the clock speed from an external LFO or
>something.
>
>Ok, so that's the plan, I'm still not "down enough" with electronics
>to know whether their are limitations which would make this
>impossible, let me know what ya think :)
>
>Seeya
>-- Plutonique9, Plutonique9 at symbiocom.com
>
>http://www.mp3.com/Plutonique9
>

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com