Randomness and it’s creation

When I was young I knew a guy who played the lottery a lot and firmly believed that his odds were higher if he put his money on numbers that had not come up in a long time. In other word he thought random systems tend “catch up” to become balanced, but this is of course totally fails. Random systems are only balanced over an infinite sample.

Before I go any further I should note that I am not an expert on any of this stuff. I don’t know much about the theory of randomness or the methods of it’s creation. This is just my own thoughts.

A lot of people seem to have strange ideas of what “random” is. For instance, look at the Dice-O-Matic built for GamesByEmail.com. It is a mechanical dice roller and is a really fun project BTW. I don’t mean to put it down at all with this post, I just don’t really understand why people would like it is better than a computer based random number generator (true or pseudo). It was built because users of the site complained that the computer generated dice rolls where not random enough. The sources he used were standard pseudo-random number generators and also true random number generators like random.org. PRNGs are certainly not random but over the sample size used on board games it seems very unlikely to me that anyone could detect the non-randomness. And if you introduce something like random.org you have random numbers generated by such a huge system (the earths atmosphere) that they would almost certainly be more random than dice rolling on a table (especially the cheap dice that come with most games).

My suspicion is that people felt like they were getting too many low rolls. But that happens in random systems. Anything can happen; that’s what random means.

There are many ways to build table top true random number generator. Many of them much smaller than a dice rolling machine (and probably more random). For instance:

Here are instuctions for building a small device that detects the alpha emitions from a sample of americonium (from a smoke detector): Alpha Radiation Visualizer. The output of this will be very random (the radiation is a quantum process).

Here is a chaotic circuit that can be sampled to produce random numbers: Make your own True Random Number Generator 2. This TRNG is not quantum (I don’t think) but it is highly chaotic and should be a good source of random bits.

[EDIT] There is also LavaRnd. Which looks much better designed than any of the other ones I mentioned. Though I think either of the above hardware devices could be used as good input into the LavaRnd algorythms.

Posted in Hardware