Re: (no subject)

 Hi Dmitry,

 I suppose, that you have a new geant3 version;  since the version 1.0 you have to replace   new TGeant3("...");
 with
   new TGeant3TGeo("...");
 in your Config.C.
 This is already demonstrated in the VMC examples,  unfortunately it is not yet documented in  geant3/README.
 With older version of geant3, this choice was  handled with compiler flags WITHG3, WITHROOT, ...  which are now supressed.

> My variables like USE_VGM is set 1.

 This variable has impact only to Geant4 VMC.

 Best regards,

 Ivana

-- 
==============================================================
e-mail:  Ivana.Hrivnacova_at_cern.ch
address: Institut Physique Nucleaire, 91406 Orsay, France
phone:   (33) 01 69 15 65 94
==============================================================


On Thu, 12 May 2005, Dmitry Naumov wrote:



> Hi Andrei, and VMC developers

> 

> I think I understand why I have problems, it seems I compiled geant3_vmc

> without

> support for  Root to VMC convertions. At least I see that

> gMC->IsRootGeometrySupported() returnes me 0 instead of expected 1.

> My variables like USE_VGM is set 1.

> To include that support I modified in geant3_vmc/Makefile the following line:

> 

> CXXFLAGS := $(CXXOPTS) $(CLIBCXXOPTS) -I. -I$(ROOT_INCDIR) -ITGeant3

> -DWITHBOTH

> 

> Which I assumed to follow README advice. What wrong I am doing?

> 

> Best regards, Dmitry

> 

> Andrei Gheata wrote:

> 

> > Hi Dmitry,

> > 

> > > Top volume is SiliconPlatePadX0. Master volume is TOP

> > 

> > 

> > This is quite strange: you should get TOP instead of SiliconPlatePadX0. Make

> > sure TOP is the first volume that you create in your ConstructGeometry. Make

> > also sure that you do not draw some volume meanwhile because this will

> > change the current top volume.

> > 

> > > Warning in <TGeoManager::CloseGeometry>: geometry already closed 

> > 

> > 

> > If your geometry is built/closed directly via TGeo (not VMC) you should call

> > gMC->SetRootGeometry() after gGeoManager->CloseGeometry()

> > 

> > > Also a not clear thing for me is that I assumed to build the geometry in

> > > MCApplication::ConstructGeometry() 

> > 

> > 

> > Your assumption is correct: here is the place to build your geometry or to

> > load it from a root file.

> > 

> > Hope this helps,

> > Cheers,

> > Andrei

> > 

> > Dmitry Naumov wrote:

> > 

> > > Dear VMC developers,

> > > 

> > > Let me ask you a further question. Just briefly let me remind -

> > > I adapted Ex03 from VMC for my application which builds

> > > geometry with TGeo, converts it in memory to G3,4 geometries.

> > > 

> > > Up to now I was plying mainly with G4 engine, now I decided

> > > to check what happens if I turn on G3 instead of G4. What I find

> > > is that my original geometry defined as:

> > > TOP

> > > TOP/Plates

> > > TOP/Plates/tubes

> > > etc

> > > is seen with G3 in a strange way:

> > > 

> > > <skip>

> > > Top volume is SiliconPlatePadX0. Master volume is TOP

> > > Warning in <TGeoManager::CloseGeometry>: geometry already closed

> > > <skip>

> > > 

> > > Also a not clear thing for me is that I assumed to build the geometry in

> > > MCApplication::ConstructGeometry()

> > > 

> > > This works perfectly in G4 interface, while it crashes with G3 interface.

> > > In order not have this crash I have to build the geometry before InitMC()

> > > method.

> > > Is it normal or I am confusing something?

> > > 

> > > 

> > > Thank you

> > > Dmitry

> > > 

> 

>