Formula is a powerful unit that allows you to evaluate logical and mathematical expressions directly via a textual Formula and a list of Arguments

Performance: At the moment, using the formula unit is significantly slower than using the operator units individually, because of the binary tree traversal overhead (despite caching attempts). It is preferable to avoid using this unit at every frame.

Most often, you'll want a formula to return either a boolean for logic, or a number for math, but formulas can return any type of value.

A new formula looks like this:

The first text field in the header is the formula itself. 

The second text field is the amount of arguments. It's set to 2 by default, giving us A and B as inputs. Formulas can have up to 10 arguments, which are always ordered alphabetically. If you add more, they will be called B, C, D, E, and so forth.

For example, this formula returns a boolean indicating whether at least 10 seconds have elapsed since the start of the game and the current object's name is Player: (because... reasons).

Variable Names

You can also directly use variable names in the formula. For example, if you have a graph variable named health, you could return a boolean just by typing the formula health > 50. The argument names are evaluated in the following order of priority:

  1. Alphabetical argument names (a - z)
  2. Graph variable names
  3. Object variable names
  4. Scene variable names
  5. Application variable names
  6. Saved variable names

Properties and Methods

You can retrieve the value of a property on an argument or variable by using the [arg.prop] notation.

For example, if position is a  Vector 3 object variable, you can check if it is equal to zero with: [position.x] = 0 .

You can also get the return value of parameterless methods with the [arg.Method()] notation. 

Note that accessing properties and methods is not guaranteed to be compatible with AOT platforms, because the AOT pre-build cannot generate stubs for members that are only accessed by name.


You can use the following literals:

Literal Description Example
Number An integer or float. 3.5
String A piece of text between apostrophes. "Hello World!"
Boolean A boolean value. true, false
Null The null constant. a != null
Delta Time The Unity frame delta time. 30 * dt
Invert Delta Time The inverse of the delta time. 30 / second


You can use every common logical and mathematical operator in formulas, even the ones defined through custom operator overloading in script. Here's an overview:

Operator Operation Rank Result Example
not, ! Logical Negation Unary The opposite of the operand. not true
- Numerical Negation Unary The negative of the operand. -5
and, && Logical And Binary True if both operands are true. (a < 5) and (b > 3)
or, || Logical Or Binary True if either operand is true (a < 5) or (b > 3)
=, == Equality Binary True if the two operands are equal. a = b
!=, <> Inequality Binary True if the two operands are not equal. a != b
<, <=, >, >= Numeric Comparison Binary The result of a numeric comparison a >= 10
+ Addition Binary The sum of the two operands. a + 5
- Subtraction Binary The difference between the two operands. b - 3
* Multiplication Binary The product of the two operands. 12 * a
/ Division Binary The quotient of the two operands. b / 2
% Modulo Binary The remainder of the division of the two operands. a % 2
?: If Ternary The left operand if the condition is true, otherwise the right operand. (health > 0) ? "Alive" : "Dead"

All common bitwise operators like ~ and >> are also supported.


You can also use any function from this table:

Name Result Example
abs The absolute value of a specified number. abs(-1)
acos The angle whose cosine is the specified number. acos(1)
asin The angle whose sine is the specified number. asin(0)
atan The angle whose tangent is the specified number. atan(0)
ceiling The smallest integer greater than or equal to the specified number. ceiling(1.5)
cos The cosine of the specified angle. cos(0)
exp e raised to the specified power. exp(0)
floor The largest integer less than or equal to the specified number. floor(1.5)
log The logarithm of a specified number. log(1, 10)
log10 The base 10 logarithm of a specified number. log10(1)
max The larger of two specified numbers. max(1, 2)
min The smaller of two numbers. min(1, 2)
pow A specified number raised to the specified power. pow(3, 2)
round Rounds a value to the nearest integer or specified number of decimal places. round(3.222, 2)
sign 1 if the number is positive, -1 is if it negative. sign(-10)
sin The sine of the specified angle. sin(0)
sqrt The square root of a specified number. sqrt(4)
tan The tangent of the specified angle. tan(0)
truncate The integral part of a number. truncate(1.7)
v2 Creates a 2D vector. v2(0, 0)
v3 Creates a 3D vector. v3(0, 0, 0)
v4 Creates a 4D vector. v4(0, 0, 0, 0)

This article was helpful for 11 people. Is this article helpful for you?

[position.x] = 0

No Idea how it is used in a formular, no real example, well not really actually anywhere.

Typing this expression will return true if the X component of the position is zero, false otherwise.


Could you guys move the "Performance" box to the top -- and maybe when a newer Bolt is released you can let us know the performance issues are fixed?   When I saw this I was really tempted to use it.

The "truncate" function doesn't work here. (ArgumentException: Function not found)

Also feeding in different values always gives me back the first result, when I use functions. If for example, I feed in the index (0-100) of a For Loop and then do the function "round (a)" I get "0" back a hundred times. Is that intended?