Fundamental Change in Random Number Generation�

Abstract

The current state of the art in computer random number generation uses a method developed over thirty-five years ago. Although much has been done to improve the sequences generated by these methods, they still have serious problems. As the results become more "random", the methodology becomes more complex. Perhaps this explains the number of poor generators in use today. The following study is an attempt to develop a fundamental change in the methodology of random number generation in an effort to both simplify and improve current methods.