Message: Navigation stuck in an G4IntersectionSolid of G4Polycone's Not Logged In (login)
 Next-in-Thread Next-in-Thread
 Next-in-Forum Next-in-Forum

Question Navigation stuck in an G4IntersectionSolid of G4Polycone's 

Forum: Geometry
Date: 21 Dec, 2011
From: Jan Pipek <Jan Pipek>

Dear experts on boolean solids and polycones,

I have the following problem: Sometimes - with no determining conditions I know of (e.g. 287 events done in several seconds, the 288th running for ever) - the computation of step (normal navigation) gets stuck somewhere in G4IntersectionSolid::DistanceToIn( const G4ThreeVector& p, const G4ThreeVector& v) - there is a while(true) loop which probably doesn't return (callgrind detection shows that 300 calls to this function results in millions of calls to DistanceToOut of the daughters) but the code seems a bit too complicated for me to deduce if this is a problem of this method or of the daughter solids.

My geometry is a bit, but not excessively complicated: I have a model of Gamma Knife Perfexion which is simplified to a G4Polycone that has as daughters ~500 channels ( G4IntersectionSolid of another G4Polycone with the mother volume ), all polycones have < 8 sides. I use just standard EM physics, no parallel geometry, ...

Is there a possibility that the calculation of distances in these solids just gets lost? What can I do about it? What should I be aware of? Or is it a bug in Geant4?

Thank you very much for any help or ideas

Jan Pipek

 Add Message Add Message
to: "Navigation stuck in an G4IntersectionSolid of G4Polycone's"

 Subscribe Subscribe

This site runs SLAC HyperNews version 1.11-slac-98, derived from the original HyperNews

[ Geant 4 Home | Geant 4 HyperNews | Search | Request New Forum | Feedback ]