Useful.<p>One thing to keep in mind: there’s been a bit of a feud in the PRNG space between the Vigna camp [1] with xoroshiro and newcomer O’Neill [2] with PCG.<p>This website seems stay on the traditional side and shy away from recommending PCG. I’m not really qualified to judge on that dispute [3], but I think both options are better, smaller, faster than the good old Mersenne Twister, so it’s time that we moved on from that [4] and pick one of the latest generation PRNGs.<p>FWIW, Julia now uses xoshiro256++ by default [5], and they investigated their choice quite thoroughly, I believe.<p>[1] <a href="https://prng.di.unimi.it/" rel="nofollow">https://prng.di.unimi.it/</a><p>[2] <a href="https://www.pcg-random.org/" rel="nofollow">https://www.pcg-random.org/</a><p>[3] <a href="https://pcg.di.unimi.it/pcg.php" rel="nofollow">https://pcg.di.unimi.it/pcg.php</a><p>And the reply
<a href="https://www.pcg-random.org/posts/on-vignas-pcg-critique.html" rel="nofollow">https://www.pcg-random.org/posts/on-vignas-pcg-critique.html</a><p>[4] <a href="https://arxiv.org/abs/1910.06437" rel="nofollow">https://arxiv.org/abs/1910.06437</a><p>[5] <a href="https://docs.julialang.org/en/v1/stdlib/Random/" rel="nofollow">https://docs.julialang.org/en/v1/stdlib/Random/</a>
I am trying and failing to find an article where a poker website published their source code, including how they shuffled cards and seeded their randomness. The seed was the current timestamp. Someone realized an efficient way to reduce the search space, assuming vaguely accurate clocks, and predict every card distributed at the table after viewing only a small number of cards.
This article fails to explain why someone should explicitly reach for an RNG over PRNG.<p>It also suggests os.urandom as a Python RNG which (as the name suggests) uses /dev/urandom or the getrandom() syscall, both of which are PRNGs
All this, and the nuclear business is still using variants of LCGs from 60 years ago [1] for Monte Carlo simulations.<p>In this case, their speed, repeatability and “good enough” statistical properties haven’t motivated much to change.<p>1. <a href="https://www.osti.gov/biblio/976209" rel="nofollow">https://www.osti.gov/biblio/976209</a>