SynthModules

Another option would be to rescale the incoming pitch CV to be more
compatible with with 10mV steps of the ADC. A voltage divider will do this,
right? So, it's not much of a hassle from a hardware standpoint, as the main
ingredients are a pot or some precision resistors. But it would make your
program slightly incompatible with a "stock" PSIM, so to speak; that's not a
big deal, IMHO.

Scaling 1V/Octave to 0.96V/Octave yields an even 80mV/semitone. Is it worth
it? Probably not for a mere 0.33% error.

I've imagined a similar application. Before doing an arpeggiator, I want to
make a CV delay ("echo").
--
john


----- Original Message ----- 
From: "djbrow54" <davebr@earthlink.net>
To: <SynthModules@yahoogroups.com>
Sent: Wednesday, May 19, 2004 3:55 PM
Subject: [SynthModules] Re: semi-tone math


> I'm not really trying to 'quantize' the keyboard.  I just want to be
> able to 'sample' it and be able to output the same voltage.  This is
> the same question as Andrew Scheidler's - what is the multiplier
> between the input and output?  I think the multiplier is 3.752984.
> If I use this in my programs, I can output the correct voltage.
> Depending on the order of the math, I can also get it to quantize or
> not.
>
> 64/17 (which is 3.7647) tracks my keyboard much better.  This is
> only a 0.3% error and I know my keyboard is built with 1% resistors
> so this is certainly within the tolerance band.
>
> The application I have mostly running scans four inputs: trigger,
> gate, CV, and delay control.  Whenever I see a 0 to 5V or 5V to 0
> transition on the gate or trigger, I timestamp the event, add the
> delay value to it, and put it into a buffer.  If the trigger
> transition was 5 to 0, then I also sample the CV and put it into the
> buffer as well.  When the timestamps match, I will output the
> appropriate trigger, gate, or CV on the outputs.  I have been
> running delays up to a couple of seconds as fast as I can play the
> keyboard.  I'm driving some VCOs and ADSRs with my keyboard and
> other VCOs and ADSRs with the PSIM. This is why I need the output
> voltage to be the same as the input voltage.  In this application,
> the order of the math does not quantize the output but keeps the
> VCOs in tune with each other.  I eventually want to turn this into a
> real-time arpeggiator.  Anyway, I got the inputs and outputs to
> match using this multiplier.
>
> Dave
>