[sdiy] Gene's method works great (was:Linear detuning !)

[jhaible at debitel.net]

> You can do my trick on most Minimoogs that have not recently been
> calibrated, ha ha.
>
> I had a hunch they would not be the same end result, only close. So - if
you
> tweak the scale and the detune so that C1 had a 1Hz beat, and C5 had a 1Hz
> beat, would C2, C3, and C4 beats be above or below 1Hz?
>
> - Gene


I've made some simulations of Gene's method. It works great within several
octaves. You don't come close to the pure linear detuning mode (which is
boring anyway), but you can make that mix of linear and expo detuning (i.e.
beat frequency increasing on higher notes, but less than proportional to the 
signal frequency) very nicely.

I evaluated the following equations:

(1) standard expo detuning

f1 = 55Hz * 2 ** CV            (= standard expo VCO)
f2 = 55Hz * 2 ** (CV + 10mV)    (= expo detuned VCO)

This gives B = 0.38Hz @ f1 = 55Hz 
and B = 12Hz @ 1.76kHz 

(2) pure linear detuning

f1 = 55Hz * 2 ** CV            (= standard expo VCO)
f2 = 55Hz * 2 ** CV + 0.38Hz   (= linear detuned VCO)

This gives B = 0.38Hz for any f1 

A mixture of (1) and (2) lets beating arbitrarily increase 
with frequency. For instance, B = 0.38Hz @ f1 = 55Hz,
gradually rising to B = 3Hz @ f1 = 1.76kHz

(3) scale misadjust and expo detuning

f1 = 55Hz * 2 ** CV            (= standard expo VCO)
f2 = 55Hz * 2 ** (0.9985 * (CV + 10mV))
 (= scale misadjustment of -0.15%, plus expo detuning
  of 10mV)

This gives B = 0.38Hz @ f1 = 55Hz 
and B = 3Hz @ 1.76kHz 
Between this, the beat frequency is continuously rising.

So you really can get that desired effect with Gene's method!

I still favor the linear detuning method for several reasons
(mistracking method will need different optimising when
octaves are switched, and the CV scaling is very sensitive 
to start with), but a pleasant beating along several octaves
can really be made without it.

JH.











>
>
>
> > Lurking a lot these days.
>
> Hi Gene - nice to see you're among us.
>
> As far as I understand it, what you're describing is another trick to make
> VCO pairs sound
> richer: slightly misadjust the tracking (V/oct scale), and then
(partially)
> compensate
> this by detuning. With this, you have two degrees of freedom, so you can
> also
> set your desired beat rate for two different notes - a high note and a low
> note,
> if you like. However, the behaviour in _between_ (change of beat rate
> accross the
> keyboard) is much different than  with my method.
>
> A third method would be running the whole (V/Oct) CV thru a waveshaper
that
> bends the response of the VCO ever so slightly.
>
> JH.
>
>



-------------------------------------------------
debitel.net Webmail