[GuB-42]

I actually how people do purposefully non-random randomness more interesting. Like in this article.

Video games are known to cheat a lot, usually in the player's favor. The Tetris randomizer for instance is well documented. Early games drew pieces truly randomly, but now, the standard is to draw randomly from a bag of all 7 pieces until it is empty, then repeat, limiting flood and draught. Along the way, other algorithms have been used with the same purpose.

But sometimes, even the numbers are fake, for example, a 95% success rate may be closer to 99% in reality because it matches more closely what players feel 95% should be like.

[saghm]

The new Baldur's Gate game has a setting called "karmic dice", where it apparently biases the dice rolls to not cause repeated outcomes as much (i.e. if you're passing checks a lot, it will start to bias towards lower results, and if you're failing checks a lot, it will start to bias towards higher results). They devs made the interesting decision to have that setting on by default, which I'm guessing is based on the experience they expect more players to want. Regardless of my own preference, I think it's pretty cool they made that an explicit setting that players can choose to keep on or turn off.

[dylan604]

I was once tasked with creating a DVD game that was meant to randomly pick questions from its available pool. Learned lots of things about how unrandom random can be. On the lower end of DVD players, there was a stored list of values between 0 and maxInt that was randomized when created. Each call of the random function would just move the pointer to the next item in the list. This meant that it would essentially play the questions in the same order every time. I can't remember the specifics if the list was reset when loading a disc or when the player was turned on.

Turns out, the company wanting the game got scarred of some patent that invoked the word random, so we had to reprogram the thing essentially do the same as that player. We had multiple sequences of randomized questions, and the game pick one of the sequences to play back in order when started.

[slfnflctd]

This reminds me of an mp3 CD player I had. I don't remember if the feature was called shuffle or random, but it played the same sequence of tracks every time. I could only fully enjoy it once per CD.

[mark-r]

I had a CD player in my SUV that had the same problem. The only saving grace is that it could play mp3s from the CD, so I had mixes that had 100 songs on them.

[Semaphor]

I once wrote a small script (in C#) to pick a few tens out of a few 1000s. I got weird repeats. I switched to a cryptographically secure RNG, and the repeats were gone. It was probably pure chance, but I stopped using the normal random function ever since ;)

[GuB-42]

Just for curiosity sake, I looked at the standard RNG in C# and it is indeed flawed, but probably not enough to be seen unless you are doing formal tests.

In [1] it says that numbers have a 0.5034 probability of being odd due to rounding errors in the algorithm that picks a number in a range, which is unacceptable for a simulation, but may actually be better than a real coin toss [2]. The raw RNG is also flawed but not that badly [3]

[1] https://fuglede.dk/en/blog/bias-in-net-rng/

[2] https://arxiv.org/abs/2310.04153

[3] https://gist.github.com/fuglede/772402ecc3997ada82a03ce65361...

[Semaphor]

Okay, now you made me want actually look into that ;) I should be able to recreate the old code (it was years ago, but I still have the script) and make a comment if I end up actually finding statistical anomalies.

[Semaphor]

It will have to stay a mystery, looking at distribution, both algorithms looked the same.

[physicles]

Do you still have your original code? Finding out if the weird repeats were real, and especially why, would make a super interesting blog post.

[Semaphor]

The old code was var r = new Random() ;)

I'm assuming I either hit an PRNG bug (after all, I encountered a compiler bug in university already), or more likely, that I simply imagined the issue or bad bad luck with the data. Never did a statistical analysis after all.