Feedback in FM synthesis

[WeAreAs1 at aol.com]

In a message dated 8/6/00 5:56:24 AM Pacific Daylight Time, debus at cityweb.de 
writes:

<< On the DX7 there are some algorithms that offer feedback over several
 operators. And on the SY/TG77/99 you can freely set the feedback path
 from anywhere to anywhere. >>


If I am not mistaken, the SY/TG series allows the user to freely set the 
feedback path, but the feedback still originates from only one operator 
(please correct me if I'm wrong about this).  In the DX series, the feedback 
path is preset within each of the 32 algorithms, and the feedback always 
comes from only one operator (I'm sure about this).  However, in many of the 
algorithms, the feedback may come from an operator that is "at the bottom" of 
a stack of operators, and it may feedback to the "top" of the stack, 
essentially giving a result that would be similar (but not identiacl) to 
having multiple feedback paths within the same "stack"

I have always dreamed of having analog VCO's that could do linear sine-wave 
FM like the Yamaha system does.  Maybe not six VCO's, with multiple 
algorithms, but at least two VCO's, with feedback path.  So far, I have never 
played an analog VCO that did this as well as the Yamaha operators.  The main 
problem seems to be dynamically scaling the modulation depth so that the 
modulated VCO plays in tune over a wide range.  The Oberheim Xpander does a 
*reasonably* good job at this, but not good enough.  I don't quite understand 
the technical aspects of the problem - Would someone care to clarify it for 
me?

Michael Bacich

.