Rando

Four random number algorithms:

• Xoshiro128Plus
• PCG32
• TauswortheLCG
• WangHash

Refer to the screenshots, but these perform anywhere up to almost 2x faster than the RNG inside of Mathematics (or 3.14 times faster than the old Engine RNG).

Four Platforms:

• Managed C#
• Direct calculation in any multithreading system such as Tasks, or using Burst/Jobs
• Examples with named instantiation or by using generic Jobs<T>
• Pre-calculate and save data to arrays to use anywhere (and/or use an IJob for calculation)
• Helper methods to advance the array index and refill it at the end of the frame
• GPU, both compute or generic shaders
• Simple use case is to just include one of the RNG header hlsl files
• Also supports the more complex case of using keywords to disable/enable a specific RNG algorithm

Twenty Methods:

• UInt [0, uint.maxValue) where [ denotes inclusion and ) exclusion
• Int [0, int.maxValue)
• Float [0, 1)
• Bool True/False
• Sign 1/-1
• UIntRange [min, max)
• IntRange [min, max)
• FloatRange [min, max)
• Float2 (all these with each component the same 0-1 as Float)
• Float3
• Vector2
• Vector3
• InsideUnitCircle
• OnUnitCircle
• InsideUnitSphere
• OnUnitSphere
• quaternion (as per the newer math library)
• Quaternion
• ColorHSVDefault (Random Color via HSV with some default ranges for HSVA)
• ColorHSV

Code:

• As always our code in all of the C# and HLSL files are fully documented, the same goes for the demo

Documentation/Demo:

• A comprehensive 13-page document to walk you through all the use cases, modes, algorithms and the demo
• The demo is a comprehensive 1.2k (documented) lines that covers all the algorithms and platforms, as well as several of the methods