Message: Re: Interaction length comparison Not Logged In (login)
 Next-in-Thread Next-in-Thread
 Next-in-Forum Next-in-Forum

None Re: Interaction length comparison 

Forum: Fast Simulation, Transportation & Others
Re: Question Interaction length comparison (George Dedes)
Date: 07 Feb, 2014
From: Marc Verderi <Marc Verderi>

Hi Georges,

Indeed this forum is not much for this generic question, but anyway.

Processes return a proposed step length (and not a cross-section) and 
are themselves responsible for making the related sampling of the 
exponential law, using their own mean free path. This sampling is 
actually done in the process base class G4VProcess. You can get the mean 
free path of a process calling its "G4double 
GetCurrentInteractionLength() const (method from G4VProcess)". If "-1" 
is returned, the process is not a true physical one (eg: transportation).
I would believe that if you call these methods at the end of a step (in 
a user stepping action) you would still get the mean free paths for the 
step that just occurred.
However, there is no easy way to get the proposed step lengths 
themselves. If this just for "curiosity" purpose, a /tracking/verbose 10 
would dump everything about what processes do, but if you do not that 
information for your application, it is more complicated.

The comparison between step length proposed by processes is made in the 
G4SteppingManager : this is its responsibility of selecting the process 
that returns the smallest step length.


PS: message re-sent: because of web mail server change, not sure the 
first attempt will succeed...

On 02/07/2014 02:47 PM, George Dedes wrote:
> *** Discussion title: Fast Simulation, Transportation & Others
> Hi,
> I put this question here, as it is generic and doesn't seem to fit
> anywhere else than a category that has "& Others":
> For the decision of each step size, each process proposes a step and the
> lowest one decides the invoked interaction and the actual step size. I
> guess that each process returns either a macroscopic cross section or
> directly a mean free path, from which then a step length is sampled
> according to the exponential decay law. Then all step lengths should be
> compared and take the lowest. I would like to know in which class(es),
> the sampling and the comparison of the proposed steps is done.
> Thanks in advance, George
> PS: I see that printing the proposed step from
> G4Navigator::ComputeStep(), gives already the final proposed step, which
> coincides with the step length that I get from my SteppingAction
> -------------------------------------------------------------
> Visit this GEANT4 at message (to reply or unsubscribe) at:

 Add Message Add Message
to: "Re: Interaction length comparison"

 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 ]