strengths of partials

Edward wrote:

>While I've got the expert here, I've been trying to
>duplicate your dissonance curves in Derive. I noticed that the two codes
>(BASIC and MATLAB) differ in the strengths of the partials, with the BASIC
>code using amplitudes that diminish by a factor of 0.88 each time and the
>MATLAB code using amplitudes of 1.0 for all the partials. And neither of the
>codes is structured the same way as the equations, although they are
>equivalent except for some scaling factors. My main question is, "Where did
>the 0.88 factor come from?" Is it meant to emulate some real musical
>instrument, or did you pick it because it generated reasonable curves and
>"interesting" sounds?

the two sets of codes are from different times, and I was doing
different examples - one with all partials the same (maybe a prototypical
organ with all stops pulled out) and the other a rough imitation of
a string instrument in which the partials die away at some rate -
and yes, I guess I picked this exact value because I liked the
curve and the sound...

of course, the intent is that one can change the relative strengths
and frequencies of the partials at will...