PPG/Waldorf/Wolfram Franke

[Paul Maddox]

Dear all, 

 Here is what I received from wolfram franke on the subject of PPG's 
and Wave's...


>  I am a big fan of wavetable synthesis, like the wave/uWave and PPG. 
>I have been trying for years to get the sound of a PPG wave into a 
>module for a modular synth. To some degree I have suceeded, I managed 
>to get a copy of the eproms in the PPG which had the 'waves' on. Only 
>to discover that there were 127, not 2048 as I expected (32 tables, 
64 
>waves in each table). I managed to obtain a copy of the schematics 
for 
>the PPG 2.3 and discoevered that there was a bank of ram which would 
>hold the 2048 waveforms, upon closer examination I noticed this was 
>made of 4164 chips, which are DRAM and as such need constant 
>refreshing. This would mean that the data would be lost at each loss 
>of power (turning of the unit). Do I assume correctly that each time 
>you power up the PPG 2.3 it generates all 2048 waveforms?

Not quite. In fact, the wavetables are calculated as soon as you 
select a
sound that uses a particular wavetable.

>   The module I have has the 127 waveforms (originally half cycle) 
>expanded to full cycle waveforms on. From this I have a CV selecting 
>wave and a Highspeed VCO (800Khz) driving a counter (with a reset 
>input for sync, just for fun). I have a theory about "transforming" 
>from one wave to another, which involves takeing the difference in 
>waveforms at the same point, dividing it by a variable amount 
(related 
>to how far through the table you want) and then adding this figure 
>onto the first. This gives me the ability to "transform" from a 
simple 
>square wave to a sine or saw or complex wave. Am I going going down 
>the right path? or is some kind of harmonic analysis used?

The interpolation scheme is quite simple although it is difficult to
explain. In its basis, it is only a linear interpolation of two waves,
filling up the empty slot between the first and the second wave.

>   I notice that your 'wave' (a synth I would dearly love to be ble 
to 
>afford) has 300 waveforms in its ROM, again I assume you most do some 
>'on the fly' transforms or something at powerup to generate the 
>tables.

The Wave has the same ROM waveforms in it (plus some new ones) and 
performs
the same interpolation thing as soon as you select a sound. This is
furthemore true for the Microwave II, XT and XTk.

BUT: the easiest and cheapest way for you to get a "modular" PPG, is 
to get
hold of a CreamWare Pulsar. Check out http://www.creamware.de



Best Wishes,

Wolfram Franke

---------------------------------------------------------------
Wolfram Franke                                         TSi GmbH
Product Support                                    Neustr. 9-12
wmf at tsi-gmbh.de                                 D-53498 Waldorf
http://www.tsi-gmbh.de          New Phone #: +49-(0)2636-9764-0
http://www.waldorf-gmbh.de       New Fax #: +49-(0)2636-9764-99
---------------------------------------------------------------


I thought some of you might find this interesting...
it completely squashes my theory about haveing all 2048 waves in 
ram/rom.. just goes to show that the PPG really was WAAAYYY ahead of 
its time in struture and concept.

At least I now know I'm on the right lines with my 'morph' module, I 
can't wait now... out with the soldering for the next few weeks...

Paul Maddox


 
******************************************************* 
*  Modulus DIY Web Page ;-                            * 
*   http://www.geocities.com/siliconvalley/park/2707  * 
*  E-Mail ;-                                          * 
*   Space_banana at hotmail.com                          * 
******************************************************* 
 


______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com