Introduction

Abstract

Recently, the proliferation of personal computers and workstations has greatly broadened the applicability of technology for stochastic computer simulation, in which we need to model and analyze real or virtual systems in uncertain conditions. On a computer, the uncertainty in such systems is usually simulated by using random numbers. While, in the past, the purpose of simulation was to obtain qualitative understanding of the targeted system or the model being validated, the explosive growth of computing power in the last decade has made the objective much more quantitative. The so-called `large-scale’ simulations have made two kinds of estimation feasible: estimation of quantities with large variances, and highly accurate estimation of targeted variables. We now need random numbers, with high quality and long periods, that are portable and can be efficiently generated.

Article FootNote

The most important questions of life are, for the most part, really only problems of probability.—Pierre Simon, Marquis de Laplace(1812)