| Geant4 floating point exception [message #4781] |
Wed, 01 August 2007 12:07  |
Sebastian Neubert
Messages: 282
Registered: March 2006
Location: Munich
|
first-grade participant |
From: *e18.physik.tu-muenchen.de
|
|
Hi!
I tried to simulate lambdas (generated with EvtGen) and I get the following error. Any idea?
Program received signal SIGFPE, Arithmetic exception.
[Switching to Thread -1235704640 (LWP 29419)]
0xb29d8b6f in G4UrbanMscModel::SampleCosineTheta ()
from /afs/e18/panda/SIM/FAIRRoot/cbmsoft/transport/geant4/lib/Linux-g++/libG4 emstandard.so
(gdb) bt
#0 0xb29d8b6f in G4UrbanMscModel::SampleCosineTheta ()
from /afs/e18/panda/SIM/FAIRRoot/cbmsoft/transport/geant4/lib/Linux-g++/libG4 emstandard.so
#1 0xb29d8732 in G4UrbanMscModel::SampleSecondaries ()
from /afs/e18/panda/SIM/FAIRRoot/cbmsoft/transport/geant4/lib/Linux-g++/libG4 emstandard.so
#2 0xb2a6dd86 in G4VMultipleScattering::PostStepDoIt ()
from /afs/e18/panda/SIM/FAIRRoot/cbmsoft/transport/geant4/lib/Linux-g++/libG4 emutils.so
#3 0xb173432b in G4SteppingManager::InvokePSDIP ()
from /afs/e18/panda/SIM/FAIRRoot/cbmsoft/transport/geant4/lib/Linux-g++/libG4 tracking.so
#4 0xb173429f in G4SteppingManager::InvokePostStepDoItProcs ()
from /afs/e18/panda/SIM/FAIRRoot/cbmsoft/transport/geant4/lib/Linux-g++/libG4 tracking.so
#5 0xb1731efa in G4SteppingManager::Stepping ()
Cheers! Sebastian.
Sebastian Neubert
Technische Universität München
Department Physik E18
sneubert@e18.physik.tu-muenchen.de
tel: +49-8928912592
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Re: Geant4 floating point exception [message #4789 is a reply to message #4785] |
Wed, 01 August 2007 13:50  |
StefanoSpataro
Messages: 2736
Registered: June 2005
Location: Torino
|
first-grade participant |
From: *physik.uni-giessen.de
|
|
The point of my answer (but not written) was:
Our VirtualMC has problems with G4 lists, the only persons that can change the code are from outside the collaboration, so we cannot operate in this sense, just report to them and wait (and this was done).
And in our external packages the data directories where completely missing, and we do not know how it affects our data.
Meanwhile there are two new versions of virtualMC, that require new external packages (such as geant4 9.0), so in some sense our code for g4 is obsolete. The update takes some time and it is not already planned, because it requires some tests.
So with our release it is better to use only G3, at the moment. This will changed hopefully in the next future, but it does not depend only from us. We are the only hep experiment which is using the virtualMC geant4, so we are the "beta tester".
|
|
|
|
|
|
|
|
|
|
| Hypernuclei simulation status [message #4795 is a reply to message #4787] |
Wed, 01 August 2007 14:58 |
asanchez
Messages: 350
Registered: March 2006
|
first-grade participant |
From: *gsi.de
|
|
hi, i'm still having problems to connect.
From my side,
i wanted to say that the tracking and identification of low kaons seems to be feasible for the tpc and the tof detector.
Unfortunally i can not extract the reconstructed information from the tpc
because Sebastian is still working on.
I have extracted the moementum and the length of the trayectory by myself(helix parametrization ,of course it is an ideal case).
But i will wait till the tpc is able to produce some information from the tracking, to study what is the expected resolution for the low kaon identification(tof).
Here attached i send you some moemnta distributions
in which kaons and the pions distributions (coming from antiXi- annihilation) are plotted. thetPK.eps(kaonon P vs polar angle) and thetPP.eps(pion P vs polar angle)
The only requirement which has to be filled by them, is that both particles have to have a transversal moemnta larger than 300 MeV in order to reach the tof detector + tpc region. The tranversal moemntum distribution of both particles are plotted also together.See ptKP.eps
I'm still working on the Secondary target(silicon strips) and on the HPGE cluster detector digitization.
I want also in the next weeks to finish the implementation
of the Hypernuclei generator, and maybe to start to work on the reconstruction.
I will commit as soon as possible the tof detector.
OK, i hope that can be helpful.
cheers ALicia.
-
Attachment: thetPK.eps
(Size: 24.46KB, Downloaded 386 times)
-
Attachment: ptKP.eps
(Size: 10.60KB, Downloaded 353 times)
-
Attachment: thetPP.eps
(Size: 28.71KB, Downloaded 386 times)
[Updated on: Wed, 01 August 2007 15:19] Report message to a moderator |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Re: Geant4 floating point exception [message #4813 is a reply to message #4801] |
Thu, 02 August 2007 18:15 |
Florian Uhlig
Messages: 424
Registered: May 2007
|
first-grade participant |
From: *gsi.de
|
|
Hi Sebastian
> I tried to just run G4 4.9 yesterday, but somehow the
> libraries seemed to have changed. I got some dynamic link
> problems.
This can be solved in changing the order of the libraries in gconfig/g4libs.
The real problem is that geant4 don't run with root 5.14.00 but
needs root 5.16.00 which envorces big changes in the framework.
So the step to geant4 9.0 is not easy to make.
> Then I looked a little bit closer to the UrbanMsc and I found > out that the error occurs when
> a) the steplength is smaller than stepmin (in
> ComputeGeomPathLength)
> and
> b) the steplength is larger than 0.05*currrentRange (in
> SampleCosineTheta)
I will try to ask the GEANT4 people for advice.
Ciao
Florian
|
|
|
|
GSI Forum
Search
Help
Members
Login
Home
