🚀 go-pugleaf

#include <stdio.h>

/* This file implements the "minimal standard" RNG
   from the paper "RNGs: Good Ones are Hard to Find" by Park and
   Miller, CACM, Volume 31, Number 10, October 1988. */

static long int seed = 1;
/* static so programs will be more "orderly" in modifying the
   seed.  The setseed function is the prescribed way in which the seed
   is altered.  A good default to use is your SSN.
   Any seed from 1 to 2147483646 is equally ok. */

int setseed(long int newseed)
{
    if ((newseed < 1) || (newseed > 2147483646)) return 1;
    seed = newseed; return 0;
}

double ranpm()
     /* page 1195-right, "Integer Version 2", works iff ints are 32 bits.*/
{
    long int a=16807, m=2147483647, q=127773, r=2836;
    int lo, hi, test;

    hi = seed / q;
    lo = seed % q;
    test = a * lo - r * hi;
    seed = (test > 0) ? test : test + m;
    return (seed/ (double) m);
}

#ifdef TESTING
/* They give us one test (on page 1195-left): if seed = 1 then
   the 10001'th random number should be 1043618065.
   Apart from this, I do some superficial testing aimed at checking
   the code above is not buggy.  I'm not trying to test the
   statistical properties of the RNG. */

int main()
{
    int n; double u, sum=0.0, sum2=0.0;
    
    for (n=1; n<=10000; n++) {
	u = ranpm();
	if ((u < 0.0) || (u > 1.0)) {
	    printf("Garbage at the %dth number!\n");
	    return 1;
	}
	sum += u;
	sum2 += (u*u);
    }
    printf("The current value of seed is : %10d\n", seed);
    printf("The Truth is %17s 1043618065\n\n", " ");

    printf("Mean of 1st 10k nums is %f, variance is %f\n",
	   sum/10000, sum2/9999);
    printf("Expect .5018 and .3355\n");
    return 0;
}
#endif /* TESTING */
-- 
Ajay Shah, (213)749-8133, ajayshah@usc.edu

In article <l73ggjINNh52@almaak.usc.edu>, ajayshah@almaak.usc.edu (Ajay
Shah) writes:
|> #include <stdio.h>
|> 
|> /* This file implements the "minimal standard" RNG
|>    from the paper "RNGs: Good Ones are Hard to Find" by Park and
|>    Miller, CACM, Volume 31, Number 10, October 1988. */
|>
|> ...

PLEASE, PLEASE don't use this generator for serious work (i.e. any project
which uses more than a million numbers or so IN ALL) without checking!

Park and Miller's paper is quite good, but is rather optimistic in their
ideas of what is a minimal reliable generator.  Unfortunately, the same is
true of Marsaglia's (with co-authors) papers "Towards a Universal Random
Number Generator" and "A Random Number Generator for PCs", and Marsaglia
is one of the people who lead this field!  Good RNGs are very, very hard to
find.

Any generator which EITHER has a 32-bit seed OR is based on conventional
single-precision starts to give seriously misleading results for sequences
of a few million numbers.  See N.M. Maclaren "A Limit of the Usable Length
of a Pseudo-random Sequence" J.Statist.Comput.Simul. 42 pp 47-54 for the
proof of the first statement - the proof of the second is trivial to derive
using the same techniques, because of the 24-bit precision limit.

Nick Maclaren
University of Cambridge Computer Laboratory,
New Museums Site, Pembroke Street,
Cambridge CB2 3QG, England.
Email:  nmm@cl.cam.ac.uk
Tel.:   +44 223 334761
Fax:    +44 223 334679

In article <1992Jul25.191638.16109@samba.oit.unc.edu>,
Jim.Buddenhagen@bbs.oit.unc.edu (Jim Buddenhagen) writes:
 
|> Portable pseudo-random number generator desired, for uniform and/or normal
|> deviates.  What is desired is code (C preferred) or algorithm that is
|> machine independent (i.e. same seed, same sequence regardless of computing
|> platform).  Purpose will be to generate random datasets (or pointers into
|> databases) for benchmark tests accross several platforms.  Must be able to
|> generate several independent sequences 'simultaneously' (ok to have several
|> copies of routine with different names).
|> 
|> I am aware of routines in Numerical Recipies, and in Knuth, but unsure
|> of quality.  Would welcome any comments, suggestions, pointers to code,
|> or references.

There are NO generators currently available that can produce multiple
sequences with any reasonable degree of independence, and almost every
expert (including Knuth) warns against misusing single generators in this
way.  Almost the best that you can do is to get non-overlapping sequences,
which is NOT the same thing!  Most of the papers that purport to describe
generators for independent sequences use definitions of independence that
are completely unheard of by statisticians.

The main exceptions that I know of are various papers by A. De Matteis and
others (which are well worth reading) and N.M. Maclaren "The Generation of
Multiple Independent Sequences of Pseudorandom Numbers" Appl.Statist. (1989)
38 pp 351-359.  But PLEASE note the provisos in the latter about the
criterion used for independence - there is more justification than there
was room for in the paper, but little is certain in this area.


Nick Maclaren
University of Cambridge Computer Laboratory,
New Museums Site, Pembroke Street,
Cambridge CB2 3QG, England.
Email:  nmm@cl.cam.ac.uk
Tel.:   +44 223 334761
Fax:    +44 223 334679