Limitations with Math.pow?

[Dan Ebberts]

Both of those work fine for me when applied to a slider. What does the rest of your expression look like?

Dan

[Allen Ellis]

You’re right – it does.

The expression is applied to a point control. Here is the code:

x = this_comp.layer(“X”).effect(“X”).param(“Slider”);
y = Math.pow(x,3); // y = x^3

[x,y]

It complains about the error on the last line, not line 2.

Here is the project file:
https://www.allenellis.com/public/Graph.aep

Look at layer 3: “Point Control”.

Thanks in advance.

[Dan Ebberts]

Hmmm… that’s strange. I don’t get any expression errors when I open the project, and it renders fine. I notice that it’s a 5.5.1 project, so I’m wondering if it’s something that got fixed between 5.5 and 7.0 (the oldest version I have installed right now).

Maybe somebody can try it in 5.5.

Dan

[Allen Ellis]

Well that project opens fine for me, too – I get the error when I change layer 3’s expression to be

y = Math.pow(x,2.5);

Or some other decimal/fraction number.

Sorry that I forgot to mention that. I didn’t want to submit a project with errors because I assumed it wouldn’t show the error that way.

[Dan Ebberts]

Ah, OK – here’s the deal. When you feed Math.pow() a negative number for x and a non-integer value for y, it tries to generate an imaginary or complex number, and the actual result you get is NaN. That’s what’s causing AE to throw the error. So – you’ll need to do something like this to get the curve you’re after:

x = this_comp.layer(“X”).effect(“X”).param(“Slider”);
if (x < 0){ y = -Math.pow(-x,2.5); }else{ y = Math.pow(x,2.5); } [x,y] Dan

[Allen Ellis]

Thanks so much – that’s exactly what I needed.

I’m running into other similar issues now but they’re more minor and I think I’ll know how to deal with them now. Thanks again :).