Simpler handling of random number generators

All random number generators managed by the NEST kernel are now seeded by providing a single seed \(s\) with \(1\leq s \leq 2^{31}-1\). The kernel automatically seeds the random number streams on the various parallel threads and processes:

NEST 2.x

NEST 3.0

msd = n_threads * seed + 1
   'grng_seed': msd,
   'rng_seeds': range(msd+1,
               'rng_seed': seed})

Changing the type of random number generator to use is easy now:

NEST 2.x

NEST 3.0

# too difficult to show here
          'rng_type': 'mt19937'})

Which random number generator types are available can be checked by:


Details about the new random number generators can be found in the guide on random number generators.

Counter-based random number generators

In addition to the conventional random number generator types, there is added support for “counter-based” random number generators (CBRNGs) with the Random123 library. Where conventional RNGs use N iterations of a stateful transformation to find the Nth random number, the Nth random number of a CBRNG can be obtained by applying a stateless mixing function to N.

The Random123 library is included in NEST, and random number generator types like Philox and Threefry are available out of the box.


On some systems or with some compilers, the CBRNGs may not give reliable results. In these cases, they are automatically disabled during compilation and will not be available.