Re: [VMC] geometry ok in G4 and invisible for G3...

From: Andrei Gheata (Andrei.Gheata@cern.ch)
Date: Thu Sep 16 2004 - 20:58:26 MEST


Hi,

By default, TGeant3 compilation uses G3 native geometry modeller. If you 
want to be able to use a ROOT geometry instead (your case) you have to 
add -DWITHROOT to the compilation flags (in 
geant3/config/Makefile.yoursystem). Recompile and try again.

Andrei

Y. Shitov wrote:
> Hi, VMC-users,
> 
> I started to create VMC-application on the base of example01. So, I
> simply imported geometry from ROOT file (which was created by g2root)
> as:
> 
> void NEMOTestMCApplication::ConstructGeometry()
> {    
>   TGeoManager::Import("mygeom.root");
>   gMC->SetRootGeometry();
> }
> 
> Compilation is OK. But, run_g3.C scripts crashes with the message:
> 
> ...
> Geant3 has been created.
> ===> ,  created
> ### Geometry loaded from file...
> Top volume is FRAM. Master volume is FRAM
> Voxelizing...
> --- node ID tracking disabled
> ### 5 nodes/ 6 volume UID's in det.C
> ----------------modeler ready----------------
>  GGCLOS : NVOLUM =    0 *****
> 
>  Calculating cross section tables, see gphysi.dat for more information
> 
>  *** Break *** segmentation violation
>  Generating stack trace...
> 
>  *** Break *** keyboard interrupt FILE:run_g3.C LINE:23
>  0x4225dffa in g3media_ at ggeom/gmedia.F:92 from
> /data/nemo2a/shitov/vmc/geant3-0.6/lib/tgt_Linux/libgeant321.so
>  0x42426cfd in gmedia_ at TGeant3/TGeant3.cxx:6207 from
> /data/nemo2a/shitov/vmc/geant3-0.6/lib/tgt_Linux/libgeant321.so
> ...
> 
> It seems that geant3 don't see the geometry although it was created
> successively. With run_g4.C  the geometry is built (the only
> question why number of node/volumes in TGeoManager (5/6) doesn't
> correspond the number of phys/log volumes (4/5) in G4. But I suppose
> that mother volume is not counted by G4 Is it so?):
> 
> Geant4 has been created.
> ...
> ===> ,  created
> ### Geometry loaded from file...
> Top volume is FRAM. Master volume is FRAM
> ### Voxelization retrieved from file
> --- node ID tracking disabled
> ### 5 nodes/ 6 volume UID's in det.C
> Warning in <TGeoManager::CloseGeometry>: geometry already closed
> Nitrogen : WARNING from G4Element::G4Element ! Trying to define an
> element as a mixture directly via effective Z.
> FRAM is not found in /
> Sensitive detector /FRAM(1) has been created.
> FOIL is not found in /
> Sensitive detector /FOIL(2) has been created.
> SIDT is not found in /
> Sensitive detector /SIDT(3) has been created.
> DSLT is not found in /
> Sensitive detector /DSLT(4) has been created.
> ACZN is not found in /
> Sensitive detector /ACZN(5) has been created.
> 
> ### No special controls in user limits are set.
> 
> **********************************************
>     GEANT4 Geometry statistics: 
>               5 logical volumes
>               4 physical volumes
>               4 materials
>               5 user limits
>               5 sensitive detectors
> **********************************************
> ...
> 
> 
> Well, what could the origin of problem I'm wondering? Is some
> addinional initialization or G4->G3 conversion required if geometry
> imported from ROOT file?
> 
> I use ROOT 4.00/03
> VMC-3 v 0.6
> VMC-4 v 1.3
> GEANT4 - 6.1
> gcc (GCC) 3.2.2 20030222 (Red Hat Linux 3.2.2-5)
> System: RH-9.0:
> $ uname -a
> Linux 2.4.20-28.9smp #1 SMP Thu Dec 18 13:37:36 EST 2003 i686 i686
> i386 GNU/Linux
> 
> Best,
> Yuri
> 



This archive was generated by hypermail 2b29 : Sun Jan 02 2005 - 11:25:00 MET