How To

Non-Linear Values of Sound Envelope Time and LFO Rate

0

nathanielbabiak 2020-01-12 03:49

I saw in the manual that the envelope times vary from 0.02 to 12 seconds, and the LFO rate varies from 0.12 to 18 Hertz. It also says it's nonlinear, so I figured I'd just use an exponential function to interpolate, but does anyone know if this is correct? I'd rather not figure out a way to confirm this (seems difficult), especially if someone else has solved it already.

For the envelope, my equation is X = 0 TO 15, Y = 0.02 * EXP( 0.426462 * X ), which produces this data: 0.02, 0.0306, 0.0469, 0.0719, 0.1101, 0.1687, 0.258, 0.396, 0.606, 0.929, 1.423, 2.18, 3.34, 5.11, 7.83, 12.

For the LFO, my equation is X = 0 TO 15, Y = 0.12 * EXP( 0.334042 * X ), which produces this data: 0.12, 0.1676, 0.234, 0.327, 0.457, 0.638, 0.89, 1.244, 1.737, 2.43, 3.39, 4.73, 6.61, 9.23, 12.89, 18.


was8bit 2020-01-12 04:49

Wish i could help, its all a mystery to me.... maybe Timo will pop in and comment :)


Timo 2020-01-12 09:15

It's a hard coded list of values. More or less exponential, but adjusted. I took this idea from the SID chip, but my values are different. Here is a super interesting interview I used as inspiration: http://www.ausretrogamer.com/interview-with-bob-yannes-sid-chip-inventor/

In the source code you can find the exact values of NX: https://github.com/timoinutilis/lowres-nx/blob/master/core/machine/audio_chip.c


was8bit 2020-01-12 13:50

Cool tech interview :) alot above my ability but still fun to read :)


nathanielbabiak 2020-01-13 00:37

Thank you so much - this'll really help! :-)


Log in to reply.