Message: Re: BooleanProcessor - Too many edges problem Not Logged In (login)
 Next-in-Thread Next-in-Thread
 Next-in-Forum Next-in-Forum

None Re: BooleanProcessor - Too many edges problem 

Forum: Visualization
Re: None BooleanProcessor - Too many edges problem (Edward Mitchell)
Date: 22 Nov, 2011
From: John Allison <John Allison>

Dear Edward

Indeed, as the Release Notes say, we "improved" boolean processing in 9.4. Seems it improved it for our test cases but made things worse for you!!

First, let me say that this has nothing to do with tracking - tracking will be OK. You can check that by viewing your volume with the RayTracerX driver, which uses Geant4 tracking algorithms.

Boolean processing is used solely in creating polyhedral representations for the purposes of visualisation (for drivers other than RayTracer). It's a tricky computational problem, especially if components share or closely share some surfaces; computer rounding errors are tricky to deal with. One thing you can do is to make sure that subtracting solids "stick out" so avoiding shared surfaces. You can often improve things just by moving components around by a few micrometres.

The message comes after boolean processing and during the conversion to a HepPolyhedron (G4Polyhedron). Curiously, it doesn't flag an error, so presumably the HepPolyhedron will still be a valid one, as indeed you have discovered, possibly with some bits missing(!). If you're confident that your solid is OK, you could simply go into, ll.1949-1952 and comment out the write statement.

We need to try and fix this. Could you please package up a snippet of your code so that we can try to reproduce the problem. Quick results not guaranteed!

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

1 None: Re: BooleanProcessor - Too many edges problem   (Edward Mitchell - 28 Nov, 2011)
(_ None: Re: BooleanProcessor - Too many edges problem   (Joseph Perl - 30 Nov, 2011)
 Add Message Add Message
to: "Re: BooleanProcessor - Too many edges problem"

 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 ]