Re: TFLUKA

Dear Andreas,

thanks a lot! I will check this now. Best regards, Dmitry

Andreas Morsch пишет:
> Dear Dmitry,
> the difference is that in FLUKA there is no way to "see" the secondaries
> before the first step. So PreTrack is not called. I informed Alfredo
> Ferrari about this and he promised to put a hook. Please have a look into
> Ex06MCApplication.cxx where Ivana has put a workaround.
>
> Best regards - Andreas.
>
>
>> Dear VMC team,
>>
>> recently I upgraded my installation of VMC software to test fluka interface.
>> Let me understand if Fluka is interfaced in exactly the same way as G3,G4?
>> I compiled my code without problems and it runs. However when I visualize
>> tracks after propagation I see a strange thing: within TFLUKA I have many
>> tracks with wrong names (like a muon gives me a lot of muons "radiated" from
>> my main muon. Obviously these were just photons with wrongly assigned names.
>> This might happen if TFLUKA has a different definition of the step compared to
>> G4.
>> I visualize tracks as follows:
>> In my application in PreTrack() method I initialize the track in MCStack as:
>> //_____________________________________________________________________________
>> void EMCApplication::PreTrack() {
>> // // Actions at beginning of each track.
>> //
>>
>> Int_t id = fStack->GetCurrentTrackNumber();
>> Int_t pdgcode = gMC->TrackPid();
>> TLorentzVector position;
>> gMC->TrackPosition(position);
>>
>> // Add track via TGeoTrack
>> fStack->PreTrack(id, pdgcode, position);
>> }
>> Then at each step I update this track:
>> //----------------------------------------------------------------------------
>> void EMCApplication::Stepping()
>> {
>> //
>> // Actions at each step.
>> //
>> fVerbose.Stepping();
>>
>> TLorentzVector position;
>>
>> // update track
>>
>> gMC->TrackPosition(position); Int_t id = fStack->GetCurrentTrackNumber();
>> TLorentzVector previous_position = fStack->GetTrackPosition(id);
>> fStack->UpdateTrack(id, position);
>>
>> ...
>>
>> }
>>
>> Thus if id if the track is wrong it would be assigned to another track. Could
>> you please give a hint if I understand correctly the problem? Apart of that
>> TFLUKA gives similar results which is good. Also related to that I remember
>> from time to time people were asking if VMC is going to provide native
>> visualization of the tracks with help of TGeoTrack. I think it should be
>> similar to what I did in the piece of code above - but done in a more general
>> and clever way :-)
>> What is the status of this?
>> Best regards, Dmitry
>>
>>
>>
>
>
Received on Tue Feb 12 2008 - 14:45:13 CET