Random Distributions for Unity

• Overview
• Usage
• Seeding
• Helpers

Overview

This asset is designed to make it easy to produce random numbers with a given distribution. It allows you to create independent, seedable generators.

Usage

The available classes are:

1. RandomLinear(min, max) - Use a linear distribution, i.e. all values are equally likely
   min: The minimum value
   max: The maximum value

2. RandomExponential(min, lambda) - Use an exponential distribution - higher values are exponentially less likely
   min: The minimum value
   lambda: The rate parameter (1 / expectation)

3. RandomGaussian(sigma, mu) - Use a Gaussian or Normal distribution - values' probabability drops off with distance from the centre
   sigma: The standard deviation; a measure of the width of the peak
   mu: The mean (expectation) value; the centre of the peak

using Voxus.Random;

var myGenerator = new RandomLinear(0, 100);

// Get a random float between 1 and 100
var randomNumber = myGenerator.Get();

Seeding

All the supplied generators have a SetSeed() method that can be used to seed each generator independently. This is especially useful for something like procedural generation.

Helpers

There's also a RandomHelpers class which provides Range() to convert a float between 0 - 1 to an int in a range (for example, to get a random element from an array) and OnUnitSphere() to get a point on a unit sphere using RandomLinear (which allows seeding, unlike Unity's built-in method).

using Voxus.Random;

var myGenerator = new RandomLinear(0, 1);

// Set a seed (optional)
myGenerator.SetSeed(0.12345f);

// Get a random element from an array
var randomEntry = myArray[RandomHelpers.Range(myGenerator.Get(), 0, myArray.Length)];

// Get a random point on a unit sphere
var randomPoint = Helpers.OnUnitSphere(myGenerator);