Skip to main content
David BK
David BK
Participating Frequently
December 3, 2021
Answered

Expressions and vector math help

  • December 3, 2021
  • 2 replies
  • 621 views

Hello!

 

I have a few relatively basic questions about After Effects expressions and vector maths.

 

1. First off (just to get my naming right) when we're talking about data types and we use the word vector, does vector imply an array of 3 dimensions?

 

2. For arrays of 4 dimensions (ex. RGBA) do we call this a vector4 in After Effects? Or simply an array of 4 dimensions?

 

3. What's the easiest way to do math where the data types don't have the same dimension? Ex. float1 + vector3. Take a look at the following expression:

myVector = [0.1, 0.1, 0.1];
myFloat = 0.2;
myResult = myFloat + myVector;
[myResult]

The result here is [0.3, 0.1, 0.1]

 

So what's the easiest way to get a result of [0.3, 0.3, 0.3]? Or how do I convert 0.2 to [0.2, 0.2, 0.2]?

 

4. What's the easiest way to round a vector?

myVector = [2.59, 6.72, 15.34];
myResult = Math.round(myVector);
[myResult]

The result here is "NaN"

 

5. Where can I find references of all of After Effects expressions? I found this doc, but it seems to be incomplete.

https://ae-expressions.docsforadobe.dev/index.html

For example, there is no mention of Math.round().

 

6. Is there a way to explicitely define data type when declaring a variable in an expression?

 

Thanks!

This topic has been closed for replies.
Correct answer Mylenium

Long and short: None of this matters. You are making things way too complicated. AE's expressions simply don't care for explicit data types, hence you can declare whatever you want as float, int or vector and the precision will then strictly be determined by the context of the underlying function like the "real" vector functions (cross, dot, add etc.) being float as are color functions, but a lot of other stuff 16bit int. Type conversions are typically handled automatically in JS. Yes, some of those functions expect at least two values, but outside that an array is still just an array with individual numbers and as per your example they have to be rounded individually either inside the array or by referencing the array index. None of this is specific to AE, it's how JavaScript math workson a broader level. Math.round() never accepted more than one number argument. The expression functions are listed in the respective sections of the AE online help (Automation & Expressions, Working with Text), the rest is just generic JS knowledge on which you can find plenty of resources on the web relating to web-based JS usage or ActionScript in Flash/ Animate.

 

Mylenium

2 replies

Mylenium
Mylenium
Legend
December 4, 2021

Vector is the data type, array is the data structure. Vector functions ultimately want arrays, they just don't have to be explicitly declared. Your examples should pretty much work, as clearly you have programming experience.

 

Mylenium

Mylenium
MyleniumCorrect answer
Legend
December 4, 2021

Long and short: None of this matters. You are making things way too complicated. AE's expressions simply don't care for explicit data types, hence you can declare whatever you want as float, int or vector and the precision will then strictly be determined by the context of the underlying function like the "real" vector functions (cross, dot, add etc.) being float as are color functions, but a lot of other stuff 16bit int. Type conversions are typically handled automatically in JS. Yes, some of those functions expect at least two values, but outside that an array is still just an array with individual numbers and as per your example they have to be rounded individually either inside the array or by referencing the array index. None of this is specific to AE, it's how JavaScript math workson a broader level. Math.round() never accepted more than one number argument. The expression functions are listed in the respective sections of the AE online help (Automation & Expressions, Working with Text), the rest is just generic JS knowledge on which you can find plenty of resources on the web relating to web-based JS usage or ActionScript in Flash/ Animate.

 

Mylenium

    David BK
    David BKAuthor
    Participating Frequently
    December 4, 2021

    Thanks @Mylenium ! This clears a few things up. I haven't used JavaScript in a while so my head is still adjusting.

     

    Just to make sure I got this...

     

    We don't really call them vectors... we call them arrays.

     

    To add a float to each element of an array (of dimension 3) I could do this:

    myVector = [0.1, 0.1, 0.1];
myFloat = 0.2;
myResult = [myFloat, myFloat, myFloat] + myVector;
[myResult]

     

    Or this:

    myVector = [0.1, 0.1, 0.1];
myFloat = 0.2;
myResult = [];
for (i = 0; i < myVector.length; i++){
	myResult[i] = myVector[i] + myFloat;
}
[myResult]

     

    And to round each element of an array:

    myVector = [2.59, 6.72, 15.34];
for (i = 0; i < myVector.length; i++) {
	myVector[i] = Math.round(myVector[i]);
}
[myVector]

     

    ShiveringCactus
    Community Expert
    ShiveringCactusCommunity Expert
    Community Expert
    December 5, 2021

    Either approach works.  Personally I prefer the second option as it spells out each step.  Only thing you need to alter is to remove the  braces ( "[" and "]" ) from the final line.

    myVector = [0.1, 0.1, 0.1];
myFloat = 0.2;
myResult = [];
for (i = 0; i < myVector.length; i++){
	myResult[i] = myVector[i] + myFloat;
}
myResult
    Terms & ConditionsAccessibility statement
    Using the community Terms of use Privacy Cookie preferences Do not sell or share my personal information ad choicesAdChoices