Boost logo

Boost :

From: Beman Dawes (beman_at_[hidden])
Date: 1999-07-15 19:20:40


At 03:16 PM 7/15/99 +0100, Paul Baxter wrote:

>In my quest for random generators I've found an alternative that
whilst not
>yet suited to the STL idiom, contains an excellent range of derived
>generators.
>(Including my favourites the Mersenne twister and ISAAC)
>
>IMHO the current boost implementation of the minimum standard could
be
>improved to add some of these others.

I would like to add some others. Same interface, though. So send
one or two in. That will force the issue.

One thing I like to see with any random number generator is some kind
of statement of authority. Same for hash function generators. What
kind of peer review has it been through? What references? For hash
functions particularly, there are a lot of bad ones out there that
users think are OK because they never measured collisions. Don't get
me going about hash table implementations that fail to report
collisions....

--Beman

------------------------------------------------------------------------

eGroups.com home: http://www.egroups.com/group/boost
http://www.egroups.com - Simplifying group communications


Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk