Re: Extended particle list in TGeant3

From: Ivana Hrivnacova <ihrivnac_at_mail.cern.ch>
Date: Fri, 7 Jan 2005 13:21:25 +0100 (CET)

On Fri, 7 Jan 2005, Yuri Kharlov wrote:

> Hi Ivana,
>
> Agree, it is simple enough to a user to define its own decay mode via
> Gsdk, as I did myself.

 This is not what I said.
 Gsdk is Geant3 specific and cannot be called from a VMC application  (via TVirtualMC) and applied to other MCs than Geant3.  That's why we need to define a corresponding function  in TVirtualMC, so that your setting of the decay channels  can work also with eg. Geant4 or Fluka.

 Best regards,

 Ivana

---

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

---

> But to do so, life time of a particle cannot be
> set to zero, otherwise a transport code in geant3 gets fpe. As soon as
> no decay channels were defined for J/psi, this transport code were never
> called and nobody noticed such a problem.
>
> So, my proposal just to change definition of the particles via Gspart as
> I did in my patch sent in the previous mail, but remove all decay
> channel definitions. So,
> Gspart(fNG3Particles++, "J/Psi", 3, 3.09688, 0., 7.6e-21);
> instead of
> Gspart(fNG3Particles++, "J/Psi", 3, 3.09688, 0., 0.);
> and so on.
>
> Best regards,
> Yuri.
>
> Ivana Hrivnacova wrote:
>
> > Hi all,
> >
> > I think that this part of code in TGeant3 should
> > better go in a user application.
> >
> > Some time ago we have interfaced the
> > Gspart() function as TVirtualMC::DefineParticle()
> > and added the function TVirtualMCApplication::AddParticles()
> > where use can define own particles via this function.
> > However we have not interfaced the Gsdk() function yet
> > which will be necessary here.
> >
> > It does not look to be difficult,
> > user will have to pass the particles PDG codes instead of
> > G3 particles codes. The implementation for G3 should be then
> > trivial, for Geant4 it shouldn't be a problem, too.
> > Let me know what do you think about such solution.
> >
> > I haven't find anywhere a call to function
> > TVirtualMC::SetUserDecay(). I searched for it in AliRoot and
> > in geant3.
> > Andreas, could you, please, clarify, if this function is
> > still used (and if so, where)? Thanks.
> >
> > Best regards,
> >
> > Ivana
> >
> >
>
>

-- 

Received on Fri Jan 07 2005 - 13:21:31 MET