Message: Re: Simulation improvements with replicated volumes Not Logged In (login)
 Next-in-Thread Next-in-Thread
 Next-in-Forum Next-in-Forum

More Re: Simulation improvements with replicated volumes 

Forum: Geometry
Re: Question Simulation improvements with replicated volumes (Nicola Mori)
Date: 06 Jul, 2012
From: Gabriele Cosmo <Gabriele Cosmo>

On Wed, 04 Jul 2012 15:27:50 GMT, Nicola Mori wrote:
> Hi, I am trying to optimize a simple geometry made by many adjacent
> scintillating fibers. As a first approach, I created a G4PVPlacement for
> each fiber. Then I looked for something more sophisticated and replaced
> the G4PVPlacements with a single G4PVParameterised. I was (naively)
> expecting that this would generate a speedup in simulation but my
> preliminary tests show that there is almost no difference in speed.
> Digging into the code, I had some hints that during tracking only the
> regular geometry made by voxels is used. This would mean that using many
> G4PVPlacement or a single G4PVParameterised or G4PVReplica to describe
> the same geometry will lead to the same simulation time (albeit an
> improvement in memory utilization can be achieved). Is this correct?
> Sorry for the naive question, thanks.

Yes, right. The main benefit is in saving memory, particularly in setups where
many physical volumes are concerned. Tracking performance is about the same.

Cheers, Gabriele

Inline Depth:
 1 1
 All All
Outline Depth:
 1 1
 2 2
 All All
Add message: (add)

1 Ok: Re: Simulation improvements with replicated volumes   (Nicola Mori - 06 Jul, 2012)
 Add Message Add Message
to: "Re: Simulation improvements with replicated volumes"

 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 ]