Thanks! Interesting. At some point I want to play around with this myself. I'm curious to see what it would sound like to combine the x and y values together (e

Thanks! Interesting. At some point I want to play around with this myself. I'm curious to see what it would sound like to combine the x and y values together (eg: if the frequency is 10 samples between each x, randomly choose between x=0 to x=10 for the first, between x=10 to x=20 for the second). That would probably make it sound less like an FM tone, and more like hissing. I'm not much for math, but a lot of tutorials on colored noise are based on multiple filters. If this algorithm works, it would just need a single low-pass filter for smoothing.