Thank Agner. My confusion is I am not sure what a "most significant bit" means when it comes to two seeds, does it mean that two numbers differ in their binary representation in only one bit? Like 00 and 01 differ by one bit? So should I expect that two similar sequences if I use sgenrand(0) and sgenrand(1)? (But this doesn't happen I think). On the MT homepage, it says that the most significant bit is not well reflected to the state vector, and that confuses me too. Any help would be greatly appreciated. Thanks, FC |