Elsevier Science Home
Computer Physics Communications Program Library
Full text online from Science Direct
Programs in Physics & Physical Chemistry
CPC Home

[Licence| Download | New Version Template] adgw_v1_0.gz(37 Kbytes)
Manuscript Title: A family of enhanced Lehmer random number generators, and hyperplane suppression, and direct support for certain physical applications.
Authors: I.G. Dyadkin, K.G. Hamilton
Program title: RNLEHMER200
Catalogue identifier: ADGW_v1_0
Distribution format: gz
Journal reference: Comput. Phys. Commun. 107(1997)258
Programming language: Assembler.
Computer: IBM PC Intel 386/387.
Operating system: MSDOS, Windows 95, Windows NT.
RAM: 6.5K words
Word size: 32
Keywords: General purpose, Statistical methods, Monte Carlo, Random numbers, Random number generators, Pseudorandom, Klein-Nishina, Neutron scattering, Nuclear geophysics, Well logging, Petroleum exploration.
Classification: 4.13.

Nature of problem:
Any Monte Carlo calculation in physics, especially those related to nuclear geophysics.

Solution method:
Approximately 200 Lehmer sequences, each with a different (specially chosen) multiplier may be used, either singly or in a circulating manner. In the latter case, the familiar Marsaglia hyperplane pattern is disrupted, and a more chaotic distribution ensues.

Unusual features:
The user is provided with the tools necessary to avoid the well-known non-randomness of the multiplicative congruence generator. In particular, he can destroy the hyperplane regularity, by using any combination of the built-in sequences, and thus produce a more chaotic stream of numbers.

Running time:
Approximately 3-4 microseconds per pseudorandom number for the uniform case, using an Intel 486DX2-66 CPU.

References:
[1] Microsoft Corporation, Microsoft Macro Assembler, Programmer's Reference, version 6.10, (Redmond WA, 1992).