More PEATHRES problems

From: Joseph Comfort <Joseph.Comfort_at_asu.edu>
Date: Thu, 1 Oct 2009 08:38:16 -0700 (MST)

I was running 40 parallel jobs of particle production by 21.5-GeV protons
on a 6-cm-long platinum target. 39 completed 7 cycles, before running out
of time; 1 crashed in the 3rd cycle.

Previously, I had run the case successfully withOUT the PHYSICS/PEATHRES
feature. I have also run successfully WITH this option for 30-GeV protons
on ~6 cm of nickel. I know there is a problem at high beam energies, but
I was surprised to encounter it here.

Another problem is execution time. When I ran the nickel case, the
PEATHRES option added about 10% in time. Now, on platinum but with lower
beam energy, the execution time has increased by about a factor of 8!

I attach a .tar.gz file of the relevant files.

We are needing the results for an experiment proposal. Overall, Fluka is
in remarkably good agreement with the results of an entirely independent
code, and I think the PEATHRES option will help the proposal. My
colleagues are refusing to accept results from a code with a known bug,
even though I think the results I can get are basically OK.

Time is very short, and I would like to get a corrected code very soon.

Thank you,
Joe Comfort

Received on Thu Oct 01 2009 - 18:56:34 CEST

This archive was generated by hypermail 2.2.0 : Thu Oct 01 2009 - 18:56:39 CEST