GSI Forum
GSI Helmholtzzentrum für Schwerionenforschung

Home » PANDA » PandaRoot » Bugs, Fixes, Releases » Question about geometryversion for EMC
Question about geometryversion for EMC [message #11793] Fri, 13 May 2011 17:32 Go to next message
donghee is currently offline  donghee
Messages: 385
Registered: January 2009
Location: Germnay
first-grade participant
From: *kph.uni-mainz.de
Hi all,

What is different between SetGeometryVersion(15) and (19) at EMC?

In version 15, the file of "emc_module125.dat" is called.
and for 19 case used "emc_module12.dat".


And I'm wondering whether both case(ver.15 and 19) will be put also very forward calorimeter (Shashlyk-type calorimeter) automatically in the simulation or not.

Thank you for your teaching...
Donghee
Re: Question about geometryversion for EMC [message #11794 is a reply to message #11793] Fri, 13 May 2011 19:11 Go to previous messageGo to next message
StefanoSpataro is currently offline  StefanoSpataro
Messages: 2736
Registered: June 2005
Location: Torino
first-grade participant

From: *w92-140.abo.wanadoo.fr
The 19 has no forward calorimeter.considering that we are not going to use it,that is consuming memory and it occupies disk space it is taken out.
Re: Question about geometryversion for EMC [message #11795 is a reply to message #11794] Fri, 13 May 2011 23:07 Go to previous messageGo to next message
donghee is currently offline  donghee
Messages: 385
Registered: January 2009
Location: Germnay
first-grade participant
From: *dip.t-dialin.net
I see, therefore 19 is much faster than 15.
ps, Some peoples are interesting to use forward calorimeter too in some reason like me.

Thanks, stefano
Re: Question about geometryversion for EMC [message #11876 is a reply to message #11794] Wed, 25 May 2011 06:48 Go to previous messageGo to next message
Dmitry Morozov is currently offline  Dmitry Morozov
Messages: 34
Registered: May 2007
Location: Protvino, Russia
continuous participant
From: 194.190.166*
Stefano Spataro wrote on Fri, 13 May 2011 21:11

The 19 has no forward calorimeter.considering that we are not going to use it,that is consuming memory and it occupies disk space it is taken out.


There is version 17 with root version of fsc "emc_module5_fsc.root". Is it also consuming memory? I have not noticed that.

Moreover versions with emc_modulexx5.dat is obsolete.

Re: Question about geometryversion for EMC [message #11877 is a reply to message #11876] Wed, 25 May 2011 10:38 Go to previous messageGo to next message
StefanoSpataro is currently offline  StefanoSpataro
Messages: 2736
Registered: June 2005
Location: Torino
first-grade participant

From: *to.infn.it
Some time ago I have seen that running complete simulation with DPM (10k events), after a while the simulation exits without giving error messages at all. This should be a matter of memory, when it is completely filled (if I remember well).
I have seen that removing the fsc from simulation (emc geometry type 19), this does not happen anymore... My fear that this is connected to the higher number of secondaries, due to the separated physics lists (SetSpecialCuts gMC->SetMaxNStep((int)1E6)Wink, or something similar.

Maybe it could be good if a check on the stability for large number of events is performed for such geometry, for sure we want to reduce the risks of crashing macrocand I have taken it out for the TDR production.
Could you please check this point?
Thanks in advance.
Re: Question about geometryversion for EMC [message #11889 is a reply to message #11877] Thu, 26 May 2011 07:13 Go to previous messageGo to next message
Dmitry Morozov is currently offline  Dmitry Morozov
Messages: 34
Registered: May 2007
Location: Protvino, Russia
continuous participant
From: 194.190.166*
Ok, Stefano, I'll check this
Re: Question about geometryversion for EMC [message #11945 is a reply to message #11793] Thu, 02 June 2011 06:38 Go to previous messageGo to next message
Dmitry Morozov is currently offline  Dmitry Morozov
Messages: 34
Registered: May 2007
Location: Protvino, Russia
continuous participant
From: 194.190.166*
Hello,

I have tested the full sim with version 17 of EMC geometry (FSC included).

The macros I used tdrct/run_sim_tpccombi_dpm.C and run_digi_tpccombi.C. Number of event 10k, momentum 5 GeV.

sim macro run smoothly without any problem, and results looks reasonable. So, no problems from FSC side. I think you can return it to business.

digi macro have eaten 4GB of RAM plus ~4GB of swap, but PC survived and macro has finished. I believe this memory consumption is due to TPC digi procedure.

My PC is ubuntu 11.04 64 bit, "may11" externals and rev. 12075 of pandaroot.

Best regards,
Dmitry
Re: Question about geometryversion for EMC [message #11957 is a reply to message #11945] Thu, 02 June 2011 17:25 Go to previous message
StefanoSpataro is currently offline  StefanoSpataro
Messages: 2736
Registered: June 2005
Location: Torino
first-grade participant

From: 222.190.126*
Hi,
I have tried to launch the run_sim (g3) at GSI, with exactly the same machine and the same seed number. If I compare the computing time:

Geometry 17: 1999 seconds
Geometry 19: 361 seconds.

I have tried twice and I got almost the same results.
Are you able to reproduce it? Considering that there are also other detectors, it seems the presence of fsc increase a lot the computing time.

[Updated on: Thu, 02 June 2011 17:26]

Report message to a moderator

Previous Topic: Reconstruction macro crush - PndMCTrackAssociator
Next Topic: Dart.sh ncpu issue
Goto Forum:
  


Current Time: Fri Mar 29 01:02:40 CET 2024

Total time taken to generate the page: 0.00971 seconds