[sdiy] fixing the modular sequencer concept, epiphany & solution

[john mahoney]

> well... I don't often stick up for Paul Perry

Well, I don't often disagree with Harry Bissel! (Because he knows, oh, a few
orders of magnitude more than I do about this stuff. :-)

I didn't mean to totally poop on the idea, it's just not the solution I'm
looking for.

It would definitey be fun to slice and dice the CVs!

So, the latency won't totally suck, that's good to know. But you obviously
can't go *too* high with the frequencies because you're using audio gear to
handle the signals. And you need many octaves of range. The frequencies can
all be scaled higher (we don't have to use 1V/octave scaling) but scaling
tends to magnify non-linearities in the conversion processes.

Speaking of non-linearity, the required VCO precision (how good need they
be?) is a question of the required system precision! The old "chain is as
strong as its weakest link" theory applies. For pitch CVs, it's got to be
real accurate and of course it needs excellent temperature stability. Also,
I'm not sure how far we can go with multiplexing, either. I suppose there's
a sort of Nyquist limit that applies here, like "the multiplexing frequency
must be less than the lowest frequency that will be recorded (divided by
some constant?)" but we also have to allow for the VCO to reach its new
pitch... well, okay, I guess that the VCO could get a sync signal from the
multiplexing circuitry... hmmm....

So again, this is certainly an interesting approach but it's not the way I'd
design a programmable synth module.
--
john


----- Original Message ----- 
From: "harrybissell" <harrybissell at prodigy.net>
To: "john mahoney" <jmahoney at gate.net>
Cc: <synth-diy at dropmix.xs4all.nl>
Sent: Saturday, February 11, 2006 10:50 PM
Subject: Re: [sdiy] fixing the modular sequencer concept, epiphany &
solution


> well... I don't often stick up for Paul Perry
>
> (prlly because he's an AH dep... and "I fought the Law"... :^)
>
> This IS a viable solution. Latency could be as little as one
> cycle of audio with a ramp/hold converter... or if you use a high
> frequency you can do a tachometer style with very short delays...
>
> Sure, maybe not enough for 130bpm sequences... but it would work
> for everything else.
>
> The V/F converter need not be a musical 'vco'... there are many
> easy to use chips out there...
>
> slicing and dicing could be fun.  Maybe record to tape and do the
> original 'cut and paste' technique.
>
> (btw Paul... you know tomorrow I'll be on your @ss again... ok ??   :^)
>
> H^) harry