[Gate-users] Random generation mechanism in Gate 5
ycui at bnl.gov
Wed Sep 30 15:27:25 CEST 2009
When I have "/gate/output/root/setSaveRndmFlag 0" and run one of the
split task (ie. Gate SPECTbench_cluster4.mac), I get the error saying,
!!! From Gate v5.0 - This command is obsolete
!!! See the new Random mechanism in the Users Guide
*** G4Exception: Aborting execution ***
Without this line, it looks okay.
From: Simon Stute [mailto:gate.stute at gmail.com]
Sent: Wednesday, September 30, 2009 6:49 AM
To: Cui, Yonggang
Cc: gate-users at lists.healthgrid.org
Subject: Re: [Gate-users] Random generation mechanism in Gate 5
It is strange that the first line causes abortion of the simulation ...
but as it is the case, you have to delete it, manually (very annoying)
or in the source code of the gjs program. To do so, open the
Cluster_Tools/jobsplitter/src/GateMacFileParser.cc file and comment line
number 289 (the line that says:
'output<<"/gate/output/root/setSaveRndmFlag 0"<<endl;'), and recompile
the gjs program.
The second line is necessary because it will set a different seed for
The third line, I think, is added when you have not clearly disabled the
ascii output in your original macro file. So add the command
Tell us if something goes wrong again.
On Wed, Sep 30, 2009 at 5:06 AM, Cui, Yonggang <ycui at bnl.gov> wrote:
I am trying to run GATE 5 on a cluster machine. "gjs" splits the job
and always adds the following lines to each "mac" file.
The first line causes abortion of the simulation. It seems that GATE 5
uses different random generation mechanism.
Can anybody explain it in detail? How do we control "gjs" and get rid
of this line?
Gate-users mailing list
Gate-users at lists.healthgrid.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Gate-users