Hi Emil and Susan,
After more investigation, unfortunately, I cannot
promise you to support this shape in VGM in a short
time.
The solution via general trapezoids does not
work in a general case.
Other possible solution would be to define TGeoXtru
solid as a G4BREP (boundary representation) solid.
However the implementation via BREP has pointed to
some problems in the BREP solid base class which are
now under investigation by Geant4 geometry experts.
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 Mon, 27 Mar 2006, Susan Kasahara wrote:Received on Thu Apr 27 2006 - 12:09:37 MEST
> Hi Ivana & Andrei,
> We (MINOS) would also be interested in having the TGeoXtru
> class supported in the G4 vmc implementation. We use TGeoXtru to
> model our steel plane geometry which has an irregular shape and
> also to model some other detector modular components.
> We are able to use only the G3 vmc implementation for now because
> of this use of TGeoXtru. We have been waiting for TGeo navigation
> support in G4 , but if it is quicker to implement TGeoXtru support in
> G4 than I think we would be happy to use that for now.
> Regards,
> -Sue
>
> Ivana Hrivnacova wrote:
>
> > On Mon, 27 Mar 2006, Emil Frlez wrote:
> >
> >
> > > Thanks Ivana.
> > >
> > > If VMC can handle that shape, I think that many
> > > experiments would be interested in using that
> > > capability. There are number of crystal ball
> > > detectors that are geodesic break-downs resulting
> > > in irregular hexagons.
> > >
> > > Can I be pushy and ask you what would be a time
> > > scale (several months or ...?) for inclusion
> > > of TGeoXtru class into vgm?
> > >
> > >
> > Not months :) I will have a look in a detail and let you know
> > in 1-2 weeks if it is feasible to support this
> > shape on the level of VGM only.
> >
> >
> > > Also, one more question. From your and a few
> > > other emails I received I realize that I don't
> > > understand the inner workings of VMC with
> > > G4/G3/other MCs. In my concrete case, TGeoXtru
> > > class - extruded polygon, is defined in ROOT,
> > > but not in VMC and Geant4. In order to use that
> > > class in Geant4 via root->Geant4 translation,
> > > is is enough to define that shape in VMC, or
> > > this class also has to be part of Geant4?
> > > Looking at your example vgm.2.07/examples/E02
> > > I thought that my geant4 code would run if
> > > TGeoXtru class is defined in ROOT and
> > > vgm.2.07/packages/Geant4GM/source/solids/
> > > and that something like G4Xtru class is no necessary?
> > >
> >
> > Yes, I hope so.
> > If we manage to define the same shape using
> > existing Geant4 solids, we don't need to define
> > G4Xtru solid.
> > Best regards,
> >
> > Ivana
> >
> >
>
>
This archive was generated by hypermail 2.2.0 : Tue Jan 02 2007 - 14:55:09 MET