[Gmsh] pesky extrusion command

Matthias K Gobbert gobbert at math.umbc.edu
Fri Aug 31 21:23:08 CEST 2001


Hi, Christophe:

Sorry for taking so long to get back to you. But yesterday was the first
day of the semester, and I was swamped.

Indeed, the new version 1.24-2 works for the extrusion (hexagonal mesh).
That is really a strange problem, if the optimization screws things up
like this. I am glad I sent you the file, we would not have found this
otherwise.

Your hint on manually swapping triangles is amazing: How did you identify
which Surface was causing the problem? There must be something I am not
understanding, because I was not even able to tell, where the issue is.

Thanks again for looking into these things for me,

Matthias

On Thu, 30 Aug 2001, Christophe Geuzaine wrote:

> Matthias K Gobbert wrote:
> >
> >
> > (I should say that I also have a triangular/tetragonal version of this
> > geometry, i.e., same surfaces and extrusions, but without Ruled and
> > Recombine. That one does not create a fatal error from gmsh, but there is
> > a warning that it cannot swap triangles properly.)
> >
>
> Yes, swapping should be more intelligent and also try to swap the 2D
> mesh... At the moment you can simply globally swap the triangles in the
> problematic surface, i.e. replace
>
> Transfinite Surface {36} = {  8,  9, 16, 15};
>
> with
>
> Transfinite Surface {36} = {  15, 16, 9, 8};
>
> and everything should be fine...
>
> Christophe
>
> --
> Christophe Geuzaine
>
> Tel: 32 (0) 4 366 37 10    http://geuz.org
> Fax: 32 (0) 4 366 29 10    mailto:Christophe.Geuzaine at ulg.ac.be
>


Matthias K. Gobbert                                Office: Math/Psyc 416
Department of Mathematics and Statistics          Hours: TTh 03:00-03:50
University of Maryland, Baltimore County           Phone: (410) 455-2404
1000 Hilltop Circle                                  Fax: (410) 455-1066
Baltimore, MD 21250                                gobbert at math.umbc.edu
U.S.A.                                 http://www.math.umbc.edu/~gobbert