Interesting Paper available on Moog VCF

[Thierry Rochebois]

>You seem to be suggesting that this algorithm is used in the Nord
>Lead,

No, I just say that the NordLead is an example of this kind of implementation.
I do not think that the researches from Clavia are published anywhere.
You know, many "industrial" researchers are not famous and never published
anything in any conference... My example is Ralph Deutch, he worked for
Kawai and patented more than 136 devices... And nobody knows him.

>that the Nord Lead has a good simulation of the Moog filter,

I played the Nord Lead in a shop once.
I tried many things to hear some problems related to digital.
For example, I opened the filter and transposed any waveform at its
maximum (including transpose and pitch bend)... and I didn't hear any
aliasing... You can try this experiment with any digital stuff (exepted
Waldorf, PPG and Fairlight... because they use a DAC per voice)

>and that a digital simulation of a filter is a good thing because a DSP
>evaluation board costs $200 (huh?).

This price must be compared to analog development equipment including
scope, iron... $-)
I own an Atari Falcon030, it includes a 56001DSP and it costs less than
$700 (and its price is going down) second hand... vintage computer ;-)

A DSP can emulate more than one module at the same time.
One can imagine to build up a virtual modular system. The number
of modules is only limited by the power of the DSP (about 60,000,000 Multiply
Add per second /44100 = 1360 MA per sample for the 56002, the half for
the 56001 in the Falcon). 

The difficult thing is to implement good oscillators and filters, without
aliasing and with good tracking curves.

>There is no attempt made to deal with nonlinearities or other interesting
>behavior responsible for the characteristic sound of the filter.

Nonlinearities are not very hard to implement in digital form.. Perhaps another
paper will show this kind of implementation.
By the way, it would be interesting to study a real Moog filter to show if they
really exist ;-)

Maybe, the punchy sound of the Moog filter can be explained in other ways :
Phase relationships between each stage, Envelop dynamics and transient 
behaviour of the filter... ?-)


Thierry Rochebois

+-------------------------------------------------------------+
|                                                             |\
|  Thierry Rochebois      Tel:69 41 79 39  Piece 353          | +
|  IEF   AXIS/TSA(ARCEMA)   Bat 220 UPS   91405 ORSAY Cedex   | |
|  thierry at ief-paris-sud.fr               FRANCE              | |
|  http://www.ief.u-psud.fr/~thierry/welcome.html             | |
|                                                             | |
+-------------------------------------------------------------+ |
 \                                                             \|
  +-------------------------------------------------------------+