I like random numbers. They are useful; whether it’s making a decision (“coffee or tea”), performing a Monte-Carlo Simulation (“if we have 5-10 critical incidents per week, with a workaround time of 4-8 hours each, how many people do we need to perform 24/7 services taking sick leave and vacation into account”) or if doing solo Role Playing Games (“Critical hit? Critical FAIL, lol!”).

I always have dice with me, at my desk I do have a Magic Eight Ball, and because I need a lot of random numbers with good quality really fast, my workstation has a “Infinite Noise TRNG” hardware random number generator.

But if you want to do that in your head, good luck. We humans suck at inventing random numbers. It’s really that bad.

But Marsaglia, the inventor of the Diehard suite, came up with a clever idea how to create good-enough random numbers in your head with very simple math, as John D. Cook reported 1 in his blog.

It’s really pretty simple, I just lift it from John’s site as it’s trivial:

state = 42 # Set the seed to any two digit number
def random_digit():
    tens = lambda n: n // 10
    ones = lambda n: n % 10
    global state
    state = tens(state) + 6*ones(state)
    return ones(state)

But… last night, when I tried to fall asleep, I was generating random numbers rather than counting sheep, and bam, I found something unexpected. The number 59 converged into itself, just like the trivial case 0. Meaning:

  • If you pick 59 as the random seed, it will only return a 9 and the next state will also be 59.

So much for getting sleep…

  1. A mental random number generator ↩