Based on only one noise layer: A constrained noise reused as a dynamic mask. So the same noise with two differnt colormappings and differnt orientations + the mask showing you the first or the second set. Again, it is all based on the very same data…
And in the second part of the video a noise with coordinates, scaling and palette parameters filled with random values every second.
http://www.youtube.com/watch?v=7ohL8GX3Zu4
Awesome! I think that’s your best noise patterns yet.
Thanks, but which one do you mean? The “cross-fade” or the random value one?
Me, I especially like the cross wipes.
By the way, what would be a reliable way to call a function exactly once per second? My hack is too dirty to show it.
You know, I think the library needs that. Been thinking about it. Not interrupts, but just periodic resettable timers. “Once every N millis…”
store a “last run time” when you do each function call and if millis()-lastruntime >1000 (or indeed TIME_INTERVAL or similar) call it again. that’s what i tend to do, not sure if there’s a better method.
Oh yes, @Mark_Kriegsman ! Another thing I´m thinking about right now and related to that: a way to specify the speed of change of a variable. Like +4223 per 1000 ms, no matter, how often the value updating is called.
Yeah. I just want to package it up.
Tired of always witting it by hand. 
That looks great! Would you mind sharing the code? I’m very interested to know how you did the transitions.
I love what your’e doing with the noise functions!
That´s nice to hear. Whenever you might find the time to write an optimized ARM version one day - I´m ready to exploit it.
Very cool experimentation result:
Modify— z[0] += 50; to this x[0] += 50;
and then add colorshift++; in the animation.
So you got it working on your 16x16?! Good! Did you need to change anything beside my recommendations?
And congrats! You are the first person posting a fragment of a new animation, @Jon_Burroughs .
I assume, the animation looks pretty dynamic now… 
Oh yeah, everything you pointed out worked. Since I’m at half the screen resolution you are at, I tried halving all the scale_x[i] and scale_y[i] values. Then from there, I guess its all a matter of taste on what looks good.
Oh, and noise smoothing needed some adaption per animation too.
I adjusted the noisesmoothing variable because at a lower screen resolution, a greater amount of smoothing can make it look ‘blocky’ or ‘flickery’ around the edges of the colors. It’s hard to explain, but it’s just an aesthetic modification. Some people may not even care, but I like to experiment with every variable in the code.
I know what you mean. High values tend to eat up fine structures and gradients. Good to know, that the presets don´t work out of the box at a lower resolution. I´ll improve that. Very curious, what else you will discover on your studies. I really appreciate your feedback.