home *** CD-ROM | disk | FTP | other *** search
- Path: hubcap.clemson.edu!hubcap!mjs
- From: mjs@hubcap.clemson.edu (M. J. Saltzman)
- Newsgroups: comp.lang.c
- Subject: Re: Statistically Random Number algorithm
- Date: 1 Apr 96 17:18:43 GMT
- Organization: Clemson University
- Message-ID: <mjs.828379123@hubcap>
- NNTP-Posting-Host: hubcap.clemson.edu
- X-Newsreader: NN version 6.5.0 #1
-
-
- schnitzi@longwood.cs.ucf.edu (Mark Schnitzius) wrote:
-
- |mjs@hubcap.clemson.edu (M. J. Saltzman) writes:
- |
- |>In a seminar that I saw, Percy Diaconis (the Harvard
- |>mathematician/magician known for some results about card shuffling)
- |>pointed out that some experiments with using atomic particle decay to
- |>produce random bits produced notoriously bad results, based on even
- |>simple tests for randomness.
- |
- |
- |Interesting. What were the problems with it? My understanding is
- |that particle decay is the only true source of randomness there is.
- |Was the problem with calibration or something? If particle decay
- |was proven to be non-random it would significantly affect our
- |understanding of the universe...
-
- I don't have the details, but my guess is that they failed at least
- some of the tests for random number sequences, such as those at DIEHARD
- (http://stat.fsu.edu/~geo/diehard.html). Particle decay is a good source
- of uncertainty, but maybe not so good a source of statistically nice
- "radomness".
-
- --
- Matthew Saltzman
- Clemson University Math Sciences
- mjs@clemson.edu
-