[Haskell-cafe] Re: Splittable random numbers

2010-11-04 Thread Bryan O'Sullivan
On Thu, Nov 4, 2010 at 11:39 AM, Thomas DuBuisson wrote: > Before we bother to do that I think it would be worth deciding what > level of performance we are trying to achieve.  On my laptop (Core2 > 2.5Ghz) I generate 4MB of random values in less than 900ms (HashDRBG). >  What is StdGen getting, w

[Haskell-cafe] Re: Splittable random numbers

2010-11-04 Thread Tyson Whitehead
On November 4, 2010 13:38:12 Simon Peyton-Jones wrote: > From: Gideon Yuval (Gideon Yuval) > Sent: Wednesday, November 03, 2010 7:15 AM > To: Simon Peyton-Jones; Burton Smith > Cc: Tolga Acar > Subject: RE: Random number generation > > As long as the key, and the non-counting part of the counter,

[Haskell-cafe] Re: Splittable random numbers

2010-11-04 Thread Ian Lynagh
On Thu, Nov 04, 2010 at 05:38:12PM +, Simon Peyton-Jones wrote: > > The generator uses crypto functions, Does that mean it couldn't be used in some countries? > so it's probably more computationally expensive than common linear-sequence > generators, but in exchange you get robust splitting