COS 302: Pseudo-Random Numbers

Generating from distributions.

[@Mutual_Information]

"You should always use an off-the-shelf, high quality, well tested implementation of random number generation" - that's useful to hear. I've never attempted to create one, but I would be quick to fool myself into think it's easy (from a distance, it sounds easy!). I will now.. never do that :)

[@oj0024]

You can still reasonably safely fool around with DIY PRNGs if you test them properly (run them through test suits like PractRand and BigCrush)

[@Mutual_Information]

@@oj0024 interesting. Still, I think it’ll be easy to avoid the temptation. Not sure what the use of making one of these would be, except to learn.