Re: [Boost-users] Newbie question: Random numbers.

From: Rene Rivera (grafik.list_at_[hidden])
Date: 2006-05-19 16:19:27

admin_at_[hidden] wrote:
>>> How about Mersenne Twister by math profs at Hiroshima U.:
>>> http://www.math.sci.hiroshima-u.ac.jp/~m-mat/MT/emt.html
>> You have to seed the Mersenne Twister with something, which is what this
>> portion of the thread is about; a good source of entropy to obtain a seed
>> from.
> 
> Yeah I just went back through the thread and saw that.  Didn't Schneier
> give some examples of how to generate highly entropic seeds in his book? 
> If I recall correctly, he suggested hooking up hardware to measure various
> properties of subatomic particles to get some truly random values.  Of
> course it all depends on how much entropy you really need and how much
> you're willing to pay for it.
> 
> Entropy requires no maintenance.

:-)

A cheap method, that I've used, is to hook up a radio to your audio card 
and record a frequency which is sufficiently far away from real radio 
traffic. With some post processing, to clean up and to ensure crypto 
safe entropy levels, it works very well and gives gigs of entropy data 
quickly.

-- 
-- Grafik - Don't Assume Anything
-- Redshift Software, Inc. - http://redshift-software.com
-- rrivera/acm.org - grafik/redshift-software.com
-- 102708583/icq - grafikrobot/aim - grafikrobot/yahoo