Hi Yuri,
You are right - at some point we changed the calling sequence to
FindNormalFast() to optimize the process, but unfortunatelly at some
point later an error slipped in... The only change needed to correct it is:
- file TGeant3.cxx, line 6386:
gGeoManager->FindNextBoundary(step);
has to be changed to :
gGeoManager->FindNextBoundary(-step);
You will find the correction in geant3 CVS during the day. Thanks for
reporting the problem.
Andrei
Yuri Kharlov wrote:
> Dear rooters,
>
> I am working now with the optical photon reflection in VMC with TGeant3.
> For the reflection, the normal to the reflecting surface is calculated
> by TGeant3::ggperp() which delegates this calculation to
> TGeoManager::FindNormalFast(). Apparently, this function returns the
> normal always (0,1,0) whatever the surface it is. In the comments to
> TGeant3.cxx it is written that before 2003/12/10 the normal was
> calculated via TGeoManager::FindNormal(). Returning to this old version
> gives more correct normal (I cannot say that is is absolutely correct,
> but it is better that (0,1,0)). Any clue?
>
> Thanks,
> Yuri Kharlov.
>
>
This archive was generated by hypermail 2b29 : Sun Jan 02 2005 - 05:50:08 MET