[fluka-discuss]: access cycle number in a source routine

From: Fasso, Alberto <fasso_at_slac.stanford.edu>
Date: Sat, 12 Apr 2014 10:37:25 -0700

Hi Moritz,

in a first moment, I thought that yours was an interesting question.
But thinking about it a little more, I realize that it makes no sense.
Why do we have several cycles? To obtain several independent results by sampling from identical distributions, and therefore to estimate the error due to the fact that each cycle is based on a different subset of the real distribution.
But if each cycle loads a different file, the distributions are not identical. You can do separate runs, each referring to a different distribution, by using a
different input file for each run. You can use for each run the same source routine, changing only one parameter in the SOURCE command of the relevant input file. If you don't like editing n input files, it is easy to write a simple script that would modify n times that one SOURCE input parameter.

But I warn you: you cannot calculate any statistics based on the results of the n runs: each run, sampling from a different distribution, gives the result of a
different problem.

Alberto

> This message: [ Message body ] [ More options (top, bottom) ]
> Related messages: [ Previous message ]
> Contemporary messages sorted: [ by date ] [ by thread ] [ by subject ] [ by author ] [ by messages with attachments ]
> From: Moritz Guthoff <Moritz.Guthoff_at_cern.ch>
> Date: Fri, 11 Apr 2014 09:28:55 +0000

> Dear FLUKA experts,

> I have several files containing properties of particles, which I am loading to FLUKA with a source routine.
> Different cycles should load different files. I would like to access the number of the current cycle in the source routine to achieve this.

> Is this possible somehow? Without the cycle number I cannot think of a way, how to tell the source routine to load a different file each cycle.

> Thanks a lot for your help,
> Moritz
Received on Sat Apr 12 2014 - 20:57:48 CEST