Hi Francois-Xavier, I see that we have more users at this stage of the geometry package that we were expecting. We decided in the development version to introduce the new class TGeoMedium to facilitate the interface with existing simulation packages like Geant3, Fluka or Geant4. This new class handles all the properties like magnetic field or other mechanisms that are specific to the simulation engines and are not material specific. A TGeoMedium references a TGeoMaterial. We are not yet completly ready to make a general announcement of the geometry package (really a great package developed with Andrei & Mihaela Gheata) We are currently comparing the results of the new package in complex detector geometries like the 4 LHC experiments with the Geant3 geometry package. During this work, we have developed an extension to Geant3 that uses the ROOT geometry package instead of Geant3. For example, the complex ALICE detector simulation is already running with it. It is our intention to make this interface available in the coming weeks. The system will be presented at CHEP end of March. As already mentioned a few times, we will provide the Virtual MC (TVirtualMC/TVirtualMCApplication) classes working with the geometry package. The first implementation will be with Geant3, but interfaces with Fluka are in the pipeline, and we also hope with Geant4. We are currently certifying the system in as many configurations as possible using real and complex detector setups. We are also collecting feedback from some "guinea-pigs" users, hoping that the user interface will become stable as soon as we announce it. Before announcing the availability, we also have to update the corresponding documentation. Rene Brun GENTIT Francois-Xavier DAPNIA wrote: > > Dear authors of the new geometry package, > I see that going from ROOT 3.04/02 to ROOT 3.05.02, a non compatible change > has been done in the method TGeoManager::MakeBox(). The arguments are no > more the same. I see that in the new version of the package, it seems that > each time one defines a TGeoMaterial, one has to define a TGeoMedium. These > 2 classes being totally undocumented, it is difficult to understand the > reason why TGeoMaterial needs its twin brother TGeoMedium and the difference > between the two ? > > Looking at the example provided in the class description of TGeoManager, I > also find that there are certainly 2 lines which are wrong > > TGeoMedium *Vacuum = new TGeoMedium(1, matVacuum); > TGeoMedium *Al = newTGeoMedium (2, matAl); > there is no constructor for TGeoMedium beginning with an integer. It must > begin with a name. > I am very interested by the new geometry package and I want to check whether > to use it in a new version of my program Litrani > (http://gentit.home.cern.ch/gentit/). It would be very useful to me to see > more documentations within the classes, and also to have more examples of > use, specially an example showing how to transport particles within the > setup. I know that it is in development and that I must be patient. > Thanks a lot for your help. > > François-Xavier Gentit > DAPNIA/SPP CEA Saclay > http://gentit.home.cern.ch/gentit/
This archive was generated by hypermail 2b29 : Thu Jan 01 2004 - 17:50:09 MET