From: Francesco Cerutti (Francesco.Cerutti@cern.ch)
Date: Tue Mar 27 2007 - 11:37:53 CEST
Dear Sunil,
> While I get the expected result on my laptop( with Fedora core 6) , I
> get a cumulative yield of 0.7on my desktop (running FC2).
in fact you get 0.07, even worse.
Quoting Ercan:
> i heard that different platform could be result in pretty little
> diffences in the simulations since the floating number handlers. that
> could be the reason?
No, 0.07 and 1 for the cumulative fluence show a HUGE difference. Note
that the two attached output files are perfectly identical (including
the final energy balance) apart from the execution times.
> Can someone please tell what is happening?
What happened is that, in order to process the USRBDX unformatted output
file, once you used the rdbdx code and the other time the rdtrk code (?),
as you can realize just looking at the headlines of the two obtained txt
files.
By the way my suggestion (not mandatory) is to use USXSUW (and similar
tools for scoring cards different from USRBDX, all available with the
normal FLUKA code distribution in directory $FLUPRO/flutil), which allows
to compute also standard deviations over several runs.
Ciao
Francesco
**************************************************
Francesco Cerutti
CERN-AB
CH-1211 Geneva 23
Switzerland
tel. ++41 22 7678962
fax ++41 22 7668854
This archive was generated by hypermail 2.1.6 : Tue Mar 27 2007 - 14:00:56 CEST