Re: PRNG
- From: rossum <rossum48@xxxxxxxxxxxx>
- Date: Tue, 03 Mar 2009 22:21:35 +0000
On Tue, 03 Mar 2009 21:35:51 +0000, Guy Macon
<http://www.GuyMacon.com/> wrote:
No need to imagine, such an algorithm exists:
Phoenix wrote:
Guy Macon <http://www.GuyMacon.com/> wrote:
Phoenix wrote:
I would like to now the opinion experts about this algorithm created by
me, that is a non periodic Pseudo Random Number Generator.
No such thing exists. You can make the time between repeats
so large that for all practical purposes it never repeats,
but you cannot write an non periodic Pseudo-Random Number
Generator algorithm on a computer that lacks infinite memory.
That's impossible.
Non periodic, is matemathicly speeking.
We all now the computers limits.
Ex: X(n)=0.3526373.38474844...75858585..., or PI, or SQR(2)
is, or not, non periodic?
Pi is not periodic, but Pi cannot be calculated on any computer
that has finite memory. No non-periodic number or string can.
You can only calculate the first N digits of Pi, with N being
very large but not infinite.
One might imagine that an algorithm exists that will generate
digits of Pi on the fly without remembering all past digits,
http://www.lacim.uqam.ca/~plouffe/Simon/articlepi.html
However, the algorithm takes n as input and so is limited by the size
of n that can be stored in the machine. Very large but not infinite.
A similar algorithm exists for the hex digits of pi which will get
more bits for a given maximum n.
rossum
just the last billion of them, but any such algorithm will
eventually find two places in the infinite length of Pi that
are identical in the last billion digits with no way to know
where it is or which one to move forward from.
Again, we are talking really large numbers and really small
probabilities here; all the stars would burn out and your
computer would lose power long before reaching such limits,
but the limits do exist.
.
- Prev by Date: Re: PRNG
- Next by Date: Re: PRNG
- Previous by thread: Re: PRNG
- Next by thread: Re: PRNG
- Index(es):
Relevant Pages
|
