Octave and Fifth Quantizer

[WeAreAs1 at aol.com]

Jorgen wrote:

<< Yesterday I made some mesurements on my ASM-1 VCO clone, 
to test my new CV quantizer.  The CV quantizer outputs CV for octaves 
and fifths. It is based on John Simonton's 4 bit quantizer 
and consists of two TL074, one 4066 and a bunch of resistors.
The following voltages was output by the quantizer:

0.000
0.498
0.996
1.494
1.997
2.495
etc..   >>

Don't you mean octaves and flatted fifths?  Raising the CV by one-half volt 
raises pitch by 6 semi-tones, but a fifth would actually be seven semi-tones 
up.  Six semi-tones will give a flatted fifth (or tri-tone).  Is this what 
you were going for?  Obviously, the resistor matrix for octaves and perfect 
fifths would be a bit more complex - maybe not even possible using Simonton's 
flash converter (I haven't thought it through, however).  

A quantizer that gives octaves and tri-tones would be very useful for quickly 
setting a VCO's basic range, as an alternative to a mechanical octave switch 
or a coarse tune knob, but a quantizer that gave octaves and perfect fifths 
could also be used for interesting musical effects.  (I guess you could 
always play "satanic" sequences and arpeggios with those tri-tones...)

On a related note, I have been working on creating a set of scale tables that 
will be burned into an EPROM.  The EPROM will be placed between the DAC 
address counter and the DAC in a voltage quantizer in an analog sequencer 
(I'm going to try it on my Oberheim Minisequencer first, since it already has 
a nice quantizer).  Right now, the Minisequencer's quantizer quantizes to 
half-steps, but the EPROM tables will remap the incoming addresses to a 
pre-defined series of scale notes, such as major scale, blues scale, harmonic 
or melodic minor, etc - yes, even octaves and fifths.  (Or weird stuff, like 
a keyboard note map that centers around middle C, but inverts the table for 
voltages below middle C, so that pitch goes UP no matter which direction you 
move away from middle C!  This wouldn't have to be half-steps, it could be 
pentatonic scales or some other idiot-proof mode, so all you'd have to do is 
hit keys, and you'd get something musical)  This scale quantizer would be a 
lot of fun to improvise through, either with the sequencer, or playing into 
it from a keyboard (making it impossible to play "wrong" notes, if you 
wanted).  LFOs and envelopes would also be useful CV inputs.

A 2764 EPROM has 12 address bits, and the six lowest addresses need to be 
used for each scale table (one address for each step in the six-bit 
quantizer's range.  Each step calls up a different map).  The other six MSB's 
can be used as bank selects to select different scales, giving access to as 
many as 64 different scale tables.  Obviously, you could put twice as many in 
a 27128, and so on, although I'm not sure that more than 64 *tonal scales* 
would be useful to me.  I guess I could also use the EPROM to hold non-tonal 
tables, such as interesting waves and patterns, but I havn't gotten around to 
thinking too hard about that - yet.  I am open to any and all suggestions on 
this topic.

This idea could be made to work on other quantizer devices, such as the one 
in the ARP sequencer, or the Polyfusion quantizer (schematic at the Synthfool 
site).

Jorgen, I realize that you were probably going for something completely 
different than what I've brought up here (tri-tones would be more practical 
than perfect fifths in a range-selecting quantizer, since they are 
symmetrical relative to the octaves, and perfect fifths are not), so I'd like 
to know how you plan to use your octave/fifth quantizer.

Michael Bacich