numbers. However, it can generate huge fields of pseudo-random numbers
that act very much like a sequence of random numbers if you pick a
different starting point each time (hence seeidng with the system clock).
*Real* random numbers have to come from a source that isn't predictable
in any way (that being the definition of random and all).
In practice, here's what you do:
1) Get a radiation source, preferably an alpha source (easy to sheild
and detect).
2) Hook up a sensor so you know when it emits an alpha particle.
3) Knowing what you're using, you know the half life. Knowing the half
life and the mass, you know the probability that it will emit a particle
at any point in time.
4) Check the detector at intervals such that, all else being equal, 50%
of the time, there will be a particle detection.
5) Convert this into a bytestream.
Now, if you need a random, say, integer, you just grab 32 (or 64 or
whatever) bits from the stream, and you have a really, *really* random
number.
--Ben
Kevin Graeme wrote:
>>From a thread in CF-Talk, Ben Doom said:
>
>
>>To nitpick: to get truly random numbers, use a quantum source, like
>>radioactive decay. Although I agree that that's a good way to get the
>>most out of the pseudorandom number generator.
>
>
> I understand that most computer systems use the system clock to
> generate random numbers and that it's not really terribly good at
> being random, but can you explain the above reference to using a
> quantum source in more detail?
>
> I have a layman's cursory knowledge of quantum physics soI think I
> know where you're going with it, but I'd be interested in hearing it
> explained.
>
> -Kevin
[Todays Threads] [This Message] [Subscription] [Fast Unsubscribe] [User Settings] [Donations and Support]
