MersenneTwister Class

public sealed class MersenneTwister : RandomSource<BlockGenerator32<MersenneTwisterEngine>>, 
	IRandomSourceFactory<MersenneTwister>

Public NotInheritable Class MersenneTwister
	Inherits RandomSource(Of BlockGenerator32(Of MersenneTwisterEngine))
	Implements IRandomSourceFactory(Of MersenneTwister)

public ref class MersenneTwister sealed : public RandomSource<BlockGenerator32<MersenneTwisterEngine>>, 
	IRandomSourceFactory<MersenneTwister^>

[<SealedAttribute>]
type MersenneTwister = 
    class
        inherit RandomSource<BlockGenerator32<MersenneTwisterEngine>>
        interface IRandomSourceFactory<MersenneTwister>
    end

Inheritance

Object  →  RandomSource<BlockGenerator32<MersenneTwisterEngine>>  →  MersenneTwister

Implements

IRandomSourceFactory<MersenneTwister>

Use the MersenneTwister class to represent a pseudo-random number generator that uses the Mersenne Twister MT19937 algorithm of Takuji Nishimura and Makoto Matsumoto.

This generator has a very long period of 2¹⁹⁹³⁷-1. This period is a Mersenne prime number (since 19937 is prime), whence its name. The algorithm is very fast and produces very good random sequences.

MersenneTwister can be used in place of the Random class to obtain pseudo-random numbers of a higher quality.

The Mersenne Twister is a linear random number generator. This makes it unsuitable for use in cryptography applications. To generate a cryptographically secure random number suitable for creating a random password, for example, use a class derived from RandomNumberGenerator such as RNGCryptoServiceProvider.

Note: This generator is provided for legacy compatibility and historical reproducibility. For new applications, consider using modern generators like Pcg64 or Xoshiro256StarStar. This implementation uses the new Layer 3 architecture with BlockGenerator32<TEngine>.

Seed Profile support:

• Standard (default): Hiroshima University 2002 reference seeding
• CPlusPlus: Partial C++11 std::mt19937 compatible seeding (scalar seeds only)
• Numpy: NumPy BitGenerator-compatible seeding

Seed profiles affect only the seeding behavior. The generation algorithm remains MT19937.

Reference: M. Matsumoto and T. Nishimura. "Mersenne Twister: A 623-dimensionally equidistributed uniform pseudorandom number generator." ACM Trans. Model. Comput. Simul. 8(1), 1998.

Reference