That depends entirely on what you mean by "serious scientific work." Different PRNGs have different properties that make them better (or worse) for specific purposes. Most practical applications require neither astronomical periods nor cryptographic security, so I would be careful about blanket statements.
Cryptography is a whole different field and most PRNG's generally used in science are not fit for that. Not any of the proposed methods does match that criteria anyway. The fact that it is GENERALLY advised against usage in science does not imply that you should never use it, and was not meant as a blanket statement. The reason that I do not use it is that I an not have it when one of my larger simulation or randomnisation models is cracked down in a later stage as a artifact from the PRNG I used.
Kim -- http://www.kimvdlinde.com
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
