home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!elroy.jpl.nasa.gov!ncar!noao!amethyst!organpipe.uug.arizona.edu!helium!uurtamos
- From: uurtamos@helium.gas.uug.arizona.edu (stephen j uurtamo )
- Newsgroups: sci.math.stat
- Subject: pseudo random numbers
- Message-ID: <1993Jan21.153350.1143@organpipe.uug.arizona.edu>
- Date: 21 Jan 93 15:33:50 GMT
- Sender: uurtamo@math.arizona.edu
- Distribution: usa
- Organization: UofA Mathematics Department
- Lines: 19
-
-
- Howdy. Could any of you give me good references to some good pseudo random
- number generators? They should have the following characteristics:
-
- (1) Be easy to implement in C (Or already exists fast C code).
- (2) The period should be scalable to fairly large values (10^20 or so).
- (3) The motivation behind the prng should be well documented and rigorous.
- (i.e. linear congruential prng would be documented in papers on finite
- fields. I actually need the straight skinny on why people think that
- these are any good.)
-
- If any of you know of some pretty good prng with these characteristics (and
- can send me references) I would really appreciate it. I don't recognize most
- of the prng mentioned by name in this group, so I'm a little lost. (If this is
- in a FAQ then you can shoot me.)
-
- Thanks in advance.
-
- steve
-