[Dan Ebberts]

Sure, but it would be like this:

transform.position.valueAtTime(time-1)[1]

Dan

[Dan Ebberts]

You could convert the audio to keyframes and start with something simple like this:

threshold = 5.0;
level = thisComp.layer(“Audio Amplitude”).effect(“Both Channels”)(“Slider”);

if (level > threshold){

// mouth open code goes here

}else{

// mouth closed code goes here

}

Dan

[Dan Ebberts]

Try this:

L = comp(“3×1”).layer(“LIGNES”);

if (L.hasParent){
s1 = L.parent.transform.scale;
s2 = L.transform.scale;
[s1[0]*s2[0]/100, s1[1]*s2[1]/100]
}else{
L.transform.scale
}

Dan

[Dan Ebberts]

The problem is that expressions have no memory. No information is passed from one frame to the next, so an expression has no way of knowing if its value is increasing or decreasing.

What you have to do is loop frame by frame from time zero to the current frame, and recalculate the expression at each frame, accumulating the results as you go. Cumbersome, but possible.

Dan

[Dan Ebberts]

You can use JavaScript’s try/catch construct for this, like this:

try{
target=comp(thisComp.layer(1).name).layer(thisComp.name);
}catch(err){
target = // something else
}

In the catch section you have to define what value to use for target if the try section fails.

Dan

[Dan Ebberts]

The easiest (and best) way to do this is with Trapcode’s Sound Keys plugin and a time remapping expression.

Other than that, you’ll need to modify your expressions so that it loops through all previous frames and accumulates the audio amplitude at each frame. Then you would multiply the total by some scale factor, add that to the current time, and convert to frame number. That will cause the value to always keep increasing, and increase faster in response to louder audio.

Dan

[Dan Ebberts]

Play around with this:

decayTime = 1; // decay time (seconds)
mult = 5;
audLev = thisComp.layer(“Audio Amplitude”).effect(“Both Channels”)(“Slider”);

k = 4.605/decayTime;
currFrame = Math.round(time/thisComp.frameDuration);
decayFrames = Math.round(decayTime/thisComp.frameDuration);
sampleFrame = Math.max(0, currFrame – decayFrames);
decayTime = 0;
curMax = 0;

while (sampleFrame <= currFrame){ sampleVal = audLev.valueAtTime(sampleFrame * thisComp.frameDuration) * mult; curVal = curMax/ Math.exp(decayTime * k); if (sampleVal >= curVal){
curMax = sampleVal;
decayTime = 0;
}else{
decayTime += thisComp.frameDuration;
}
sampleFrame++;
}
curVal

Dan

[Dan Ebberts]

If I understand Glennser’s method, it seems like you could apply a layer transform expression to the pins of the second instance of CC Power Pin to tie them directly to the tracking points for the frame. It would look like this:

fromComp(thisComp.layer(“Frame Layer”).motionTracker(“Tracker 1”)(“Track Point 1”).attachPoint)

Dan

[Dan Ebberts]

Here’s an example for opacity. You’ll need to adjust the parameters to fit your situation:

minDist = 600;
maxDist = 800;
minOpacity = 30;
maxOpacity = 100;

d = length(toWorld(anchorPoint),thisComp.activeCamera.position);
linear(d,minDist,maxDist,maxOpacity,minOpacity)

Dan

[Dan Ebberts]

You must have left off the minus sign at the front of the expression (does the same thing as multiplying by -1). Glad you got it working though.

Dan