The continous-time equation is : (d²x/dt²)/F² +(dx/dt)/(Q*F)+x = E, where E stands for the excitation, F the "proper" pulsation (not the frequency) of the system, and Q the quality coefficient.

the derivations are discretized using :

d²/dt² --> (Z - Z^(-1))Fe²,

and d/dt --> (Z - Z^(-1))Fe/2 (Fe is the samplerate);

After some algebra, this leads to the difference equation:

x(n) = 1/(Fe²/F² + Fe/(2QF))* [ (2Fe²/F² - 1)x(n-1) + (Fe/(2QF) - Fe²/F²)x(n-2) + E(n-1)];

anyway, i did a mistake in the biquad~ coefficients (the excitator coefficient), and now highest pulsations are louder than the lowest, but the difference is smaller ...

here is the new patch. François.

interesting problem, could you describe the equation you used?

cyrille

Accédez au courrier électronique de La Poste : www.laposte.net ; 3615 LAPOSTENET (0,34 €/mn) ; tél : 08 92 68 13 50 (0,34€/mn)