Can wiggle be made less random using seed?

[paul_7484]

Using a wiggle expression to a film jitter effect, eg wiggle(12,7)

Is there a way to make the wiggle effect use each keyframe as the "seed" of the subsequent keyframe, instead of using the center as the point of reference, and in that way making it less random? I'm looking for an effect that produces "drift" while jittering. Is this possible using a math expression, with a set minimum and maxium excursion of position?

[Dan Ebberts]

You can stack multiple wiggles, like this (in this example w1 is a fast jittery wiggle, where w2 is slower, providing drift):

w1 = wiggle(12,7) - value;
w2 = wiggle(.5,25) - value;
value + w1 + w2

[paul_7484]

Great - thank you, that's a neat solution. So w2 is moving position based on where it is in w1, not just in parralel, right?

Can I use that same type of 'nested' expression of "wiggle" for other effects that wiggle, eg a flicker in level of exposure? 

[Dan Ebberts]

Both w1 and w2 are based on the pre-expression value, but they're added together so the result is cumulative. Not sure if that answers your question though. wiggle() is somewhat unique in that its result includes the pre-expression value, so to get just the wiggle part, you have to subtract value from the result. This technique should work for any property, if that's what you're asking.

[paul_7484]

I see what you're saying. It's a different approach to what I was expecting, but it looks like it pretty much does what I was seeking to do. Less linear than I was expecting, it's quite interesting how complex this is. Many thanks!

[Dan Ebberts]

Once you have two wiggles isolated, you can combine them in interesting ways. For example, you could multiply the two wiggles together, like this, to get a completely different feel:

w1 = wiggle(12,7) - value;
w2 = wiggle(.5,25) - value;
value + [w1[0]*w2[0],w1[1]*w2[1]]

[paul_7484]

Yes interesting. At risk of getting even more into the weeds, is it possible with a wiggle expression to radomize the frequency variable;  eg, instead of performing the w2 wiggle *every* instance 0.5 times per second, to instead perform that wiggle at a more randomized interval of let's say, once in a range of 0.1 to 24 per second?

In other words, w1 happens lets say 24 times per second

w2 happens at random intervals, within a range (no less often than 0.1 times per second, no more than 24 per/sec)

That would start looking a lot more organic. Hope that question is clear.. 

[Dan Ebberts]

Manipulating the frequency is surprisingly tricky, but possible. This might be helpful:

https://www.motionscript.com/articles/speed-control.html#wiggle

[paul_7484]

Wow okay, that's a lot of code. Popping this code in there with a slider and a few keyframes surprisingly worked well. Not exactly what I intended, but very interesting and definitely more non-linear. Hard to control but getting much closer to an organic film jitter. Thank you.

freq = effect("Slider Control")("Slider");
amp = 1;
n = freq.numKeys;
if (n > 0 && freq.key(1).time < time){
  accum = freq.key(1).value*(freq.key(1).time - inPoint);
  for (i = 2; i <= n; i++){
    if (freq.key(i).time > time) break;
    k1 = freq.key(i-1);
    k2 = freq.key(i);
    accum += (k1.value + k2.value)*(k2.time - k1.time)/2;
  }
  accum += (freq.value + freq.key(i-1).value)*(time - freq.key(i-1).time)/2;
}else{
  accum = freq.value*(time - inPoint);
}
wiggle(1,amp,.5,.5,accum)