-
Simulate chasing over a terrain
I’ve got a pointer that stays in the center of a comp, and a shape that wiggles around within the bounds of the comp. The pointer is always looking at the wiggling shape using this expression:
target = thisComp.layer("Target");
delta = thisLayer.toComp([0,0]) - target.toComp([0,0]);
radiansToDegrees(Math.atan2(delta[1],delta[0])) + valueWhat I’d like to do next, is have a texture underneath moving in the opposite direction to the pointer to simulate the pointer ‘chasing’ the shape around a terrain. Ideally involving change in speed too. I’m not sure where to start with it though, any help would be much appreciated. Cheers.
replied 5 years, 4 months ago 2 Members · 5 Replies
-
5 Replies
-
Maybe something like this applied to the Offset Turbulence parameter of Fractal Noise (or the position property of a large texture layer):
refPoint = [thisComp.width/2,thisComp.height/2];
distanceFactor = 2;
delta = thisComp.layer("Target").position - refPoint;
value - delta*distanceFactor
-
Hey Dan, that’s definitely on the right track. The problem with that though is that the terrain stops moving when the target does, I’d like the terrain to be moving continuously and only angled via the direction of the pointer/target. The speed variation could be that the terrain moves faster when the target is closer to the pointer (as if it was catching up).
I have a feeling the answer may lie somewhere in your guide here:
https://www.motionscript.com/articles/speed-control.html
Although I’m not sure how to integrate the concepts here, because it feels like it needs ‘time’ in play to keep it moving constantly (with no keyframes).
-
Yeah, OK. Maybe something like this then:
maxSpeed = 10;
minSpeed = 0;
maxDistance = 500;
target = thisComp.layer("Target").position;
pointer = thisComp.layer("Pointer").position;
accum = d = [0,0];
f = timeToFrames(time);
for (i = 0; i <= f; i++){
t = framesToTime(i);
d = target.valueAtTime(t) - pointer.valueAtTime(t);
factor = linear(length(d),0,maxDistance,maxSpeed,minSpeed);
accum += normalize(d)*factor;
}
value - accum;
-
that’s so cool Dan, thank you!
I will take the time to dissect and absorb this. It means a lot.
One thing, is there a reason the ‘accum’ variable needs a second name ‘d’? Does that establish two separate variables?
-
Just a lazy way to initialize two variables to the same value.
Reply to this Discussion! Login or Sign Up