Hacker News new | past | comments | ask | show | jobs | submit login

> I have used Math.random, Random.org and other sources, but have always received numerous complaints that the dice are not random enough. Some players have put more effort into statistical analysis of the rolls than they put into their doctoral dissertation.

I seriously doubt that high-quality electronic randomness is non-random enough to have a noticeable effect on the outcome of board games. It's nice that the guy is accommodating enough to go to all this effort, but it seems unnecessary. Cool project, though.




I think you can assume he's in on that joke. The goal of this was to build a fun machine, not to actually increase entropy.


The fact that he offers to melt and mail a die to any player who complains of non-randomness is a giveaway he's having fun.


humans have an odd relationship with randomness. iTunes had to de-randomize their shuffle system, because it was too random (i.e. you could get the same song back to back etc, purely randomly). We tend to see patterns that just aren't there. I'm not surprised his player complained.

[1] similar article about spotify: https://www.quora.com/Is-Spotifys-Shuffle-feature-truly-rand...


A similar story I recall was that shuffling a deck (by hand) vs randomly sorting cards. It's more likely that the values were TOO random and resulted in a less enjoyable experience (or the player wanted to blame someone other than themselves).

I went looking for a video I recalled about a craps player able to choose what he'd roll. I didn't fint it but I did find: https://www.insidescience.org/news/dice-rolls-are-not-comple...


Bridge is particularly prone to being affected by poor shuffling, for two reasons. Firstly, cards of the same suit are clumped together as part of each trick, and if they are not separated during the shuffle, get evenly distributed back to the players. Secondly, suit distribution is important in bridge - it's normal to describe a hand by the amount of each suit it has.

Players noticed that voids and singletons were more common when playing online or with computer-generated hands. Unlike the board gamers of the article and everyone else who complains about bad RNGs, they had noticed a real effect, but the effect was backwards - the computer hands were better shuffled and more random.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: