rng

Initialising the built-in random number generator

Usage

rng(seed)

• The input seed should be a non-negative integer scalar.
• This function has no return value.
• It initialises the random number generator to a given initial state (determined by seed) using the algorithm discussed in William H. Press, Numerical Recipes, 3rd edition, Cambridge University Press, 2007, p. 342.
• Since all pseudo-random number generating functions in this app (e.g., randi, normrnd, exprnd, poissrnd) utilise the same generator, executing rng(seed) would simultaneously reset all of the functions to the initial state determined by seed.
• 💡 The default seed is the current time, i.e., number of seconds since 00:00:00 UTC, January 1, 1970. The default seed is used when no existing seed can be found. This usually happens the app launches for the first time after a fresh installation.
• 💡 Once a seed has been set, it will not change until a new seed is set by rng or by uninstalling the app.

Example

• When it is initialised with 17, rand() generates the same sequence of pseudo random numbers. When it is initialised with a different number, a different sequence is generated.

Input

rng(17)
rand(1,5)
rng(17)
rand(1,5)
% Different seed number
rng(16)
rand(1,5)

Output

ans = 1e-1 × 
 0.1463   4.0537   8.8335   9.2374   2.6807

ans = 1e-1 × 
 0.1463   4.0537   8.8335   9.2374   2.6807

ans = 1e-1 × 
 2.6321   9.6424   8.6923   0.1035   3.6630

Difference with MATLAB

• rng(seed, generator) and s = rng are not available in this app.
• This app's algorithm for producing the initial state using a given seed number may be different from MATLAB's.
• Since this app's algorithms for generating random numbers may be different from MATLAB's, users should not expect this app and MATLAB give same sequences of pseudo-random numbers for the same seed number.