On Thursday, July 8, 2004, at 11:44 AM, Anders Edin wrote:
// For me it would seem more natural if the engine class would be a // singleton, so that there can be only one object of any particular // random generator engine type.
// Our purpose for using random numbers is to simulate physical processes. // I can't think of a case when there would be a need for several random // engines, even if there can of course be many different random // distributions in an application.
// Having the same random number sequence in different parts of a simulation // can probably create very strange behaviour, since the numbers from the // different engines of the same type will be perfectly correlated.
// Perhaps other applications have use of several random engines?
One of the important characteristics of a pseudo random number generator is, oddly enough, determinancy. Given the same seed, you get the same sequence and can therefore reproduce the exact same results. This allows results to be checked and bugs to be tracked down reliably. If another thread, or just a library used by your application (for example, something that uses a pseudo-random numbers for dithering a graphic display), was using the same singleton engine as your application this characteristic would be lost. The number of values drawn by the other parts of your system might change or reseeding of the engine might happen without your knowledge. No serious statistical, scientific, or simulation results should ever be done using any kind of shared state RNG (including the C stdlib rand function). Topher Cooper