Home › API Reference › Mathf

Mathf

Type	Struct (only static members)
Namespace	`UnityEngine`
Works with	`float`

Mathf

Mathf is a toolbox of math functions for float numbers: clamping, rounding, smooth movement, trigonometry and more. It's Unity's version of C#'s System.Math, but works with float instead of double.

Constants

Name	Description
`Mathf.PI`	3.14159...
`Mathf.Deg2Rad` / `Mathf.Rad2Deg`	Multiply to convert degrees to radians, or back.
`Mathf.Infinity`	A number bigger than any other. Handy as a starting "closest distance".

Most-used methods

Method	Description	Example
`Clamp(value, min, max)`	Keep a value inside a range.	`Clamp(150, 0, 100)` = 100
`Clamp01(value)`	Clamp between 0 and 1.	`Clamp01(1.4f)` = 1
`Lerp(a, b, t)`	Value `t` of the way from `a` to `b` (`t` is clamped 0-1).	`Lerp(0, 10, 0.25f)` = 2.5
`InverseLerp(a, b, value)`	The opposite: how far along is `value`? Returns 0-1.	`InverseLerp(0, 200, 50)` = 0.25
`MoveTowards(current, target, maxDelta)`	Step toward a target, never overshooting.	`MoveTowards(0, 10, 3)` = 3
`SmoothDamp(current, target, ref velocity, smoothTime)`	Spring-like smooth approach.	Camera zoom
`Abs(x)`	Remove the minus sign.	`Abs(-4)` = 4
`Sign(x)`	1 if positive or zero, -1 if negative.	`Sign(-3)` = -1
`Min(a, b)` / `Max(a, b)`	Smaller / larger value.	`Max(2, 7)` = 7
`RoundToInt(x)`	Round to nearest int. Halves go to the nearest even number.	`RoundToInt(2.5f)` = 2
`FloorToInt(x)` / `CeilToInt(x)`	Round down / up to an int.	`FloorToInt(2.9f)` = 2
`Sqrt(x)`, `Pow(x, p)`	Square root, power.	`Pow(2, 3)` = 8
`Sin(rad)`, `Cos(rad)`	Trig. Takes radians.	`Sin(Time.time)` bobs -1 to 1
`Atan2(y, x)`	Angle (radians) of a direction. Great for 2D aiming.	See below
`PingPong(t, length)`	Goes 0 → length → 0 → ... as `t` increases.	Back-and-forth platforms
`Repeat(t, length)`	Like `%` (modulo) but works with floats and negatives. Wraps around.	`Repeat(370, 360)` = 10
`DeltaAngle(a, b)`	Shortest difference between two angles in degrees.	`DeltaAngle(350, 10)` = 20
`Approximately(a, b)`	Are two floats nearly equal?	See Common Mistakes

Code Samples

Health that can't go below 0 or above max

currentHealth = Mathf.Clamp(currentHealth - damage, 0, maxHealth);

A floating, bobbing pickup

using UnityEngine;

public class Bob : MonoBehaviour
{
    [SerializeField] private float height = 0.25f;
    [SerializeField] private float speed = 2f;
    private Vector3 startPos;

    void Start() { startPos = transform.position; }

    void Update()
    {
        // Sin goes smoothly between -1 and 1 as time passes.
        float offset = Mathf.Sin(Time.time * speed) * height;
        transform.position = startPos + Vector3.up * offset;
    }
}

Health bar fill amount

// Turn "37 out of 120" into a 0-1 value for a UI Image's fillAmount.
healthBar.fillAmount = Mathf.InverseLerp(0, maxHealth, currentHealth);

2D: rotate a sprite to face the mouse

Vector3 mouse = Camera.main.ScreenToWorldPoint(Input.mousePosition);
Vector2 dir = mouse - transform.position;
float angle = Mathf.Atan2(dir.y, dir.x) * Mathf.Rad2Deg;
transform.rotation = Quaternion.Euler(0f, 0f, angle);

Smooth camera zoom

private float zoomVelocity; // SmoothDamp needs a field to remember speed between frames

void Update()
{
    cam.fieldOfView = Mathf.SmoothDamp(cam.fieldOfView, targetFov, ref zoomVelocity, 0.2f);
}

Common Mistakes

Comparing floats with ==. Float math has tiny rounding errors, so 0.1f + 0.2f == 0.3f can be false. Use Mathf.Approximately(a, b) or check Mathf.Abs(a - b) < 0.01f.
Passing degrees to Sin/Cos. Mathf.Sin(90) is not 1. Use Mathf.Sin(90 * Mathf.Deg2Rad).
Integer division. currentHealth / maxHealth with two ints is 0 until it's 1. Cast one to float: (float)currentHealth / maxHealth.
Expecting Sign(0) to be 0. Mathf.Sign returns 1 for zero.

Mathf

Constants

Most-used methods

Code Samples

Health that can't go below 0 or above max

A floating, bobbing pickup

Health bar fill amount

2D: rotate a sprite to face the mouse

Smooth camera zoom

Common Mistakes

See also