Message: Re: Yet another random-seed question -- reproducing event #N Not Logged In (login)
 Next-in-Thread Next-in-Thread
 Next-in-Forum Next-in-Forum

None Re: Yet another random-seed question -- reproducing event #N 

Forum: Event and Track Management
Re: Question Yet another random-seed question -- reproducing event #N (Michael H. Kelsey)
Date: 24 Feb, 2010
From: Tom Roberts <Tom Roberts>

The approach I have used for years won't solve your immediate problem, but will permit you to avoid it in the future.

Before the start of each event I seed the random number generator with the event number. This means I can always re-run any given event without any information other than the event number (assuming no changes to the simulation). It also makes it trivial to manage multiple jobs to combine their outputs -- I just need to ensure they have disjoint sets of event numbers.

The default random number generator is documented to ensure independent streams of random numbers for seeds from 0 to 900,000,000.

 Add Message Add Message
to: "Re: Yet another random-seed question -- reproducing event #N"

 Subscribe Subscribe

This site runs SLAC HyperNews version 1.11-slac-98, derived from the original HyperNews


[ Geant 4 Home | Geant 4 HyperNews | Search | Request New Forum | Feedback ]