With my up-to-date version, I still get the smooth result when the I let Gmsh 
to choose the mesh.
You can e.g. force it in your .geo file:

Mesh.Algorithm = 2; 
// 2D mesh algorithm (1=MeshAdapt, 2=Automatic, 5=Delaunay, 6=Frontal, 7=bamg, 
8=delquad)

I get the second mesh when forcing algorithm 1, MeshAdapt.
The result you are getting is probably normal.

Regards,
Ruth

On 15 Apr 2013, at 07:57, Omid Mahabadi <[email protected]> wrote:

> Dear Christophe and Gmsh team,
> I may have encountered a possible issue with the new Gmsh revisions (r15256 
> and r15299) or I may have to modify something in my geometry file for it to 
> mesh properly again: after thoroughly examining my geo file and how it's been 
> meshed in different versions of Gmsh, it appears that the characteristic 
> lengths associated with the internal points are not been used for meshing in 
> later revisions of Gmsh. I used to mesh the attached geometry file easily 
> with Gmsh 2.5 UI or Gmsh r15203 non-interactive, resulting in a very nice and 
> smooth graded mesh. However, in the later revisions (r15256 non-interactive 
> or r15299 with the UI), the characteristic lengths of the internal points is 
> ignored. I've attached 2 screenshots with the resultant meshes for your view.
> I have tried varying the characteristic lengths, but it seems that only those 
> of the boundary or external points are used. Do you know what may be going 
> wrong in my geometry file? Do I have to change something, for instance a 
> meshing option, to force the mesh to use the specified mesh lengths? Or can 
> this be an issue with the updates in Gmsh?
> 
> Thank you,
> Omid
> 
> -- 
> Omid Mahabadi
> Geomechanica, Inc.
> Tel: (647) 478-9767 x824
> 
> <open_pit_base_geom_no_internal_lines.geo><new_gmsh.png><old_gmsh.png>_______________________________________________
> gmsh mailing list
> [email protected]
> http://www.geuz.org/mailman/listinfo/gmsh

—
Dr. Ir. Ruth V. Sabariego
University of Liege, Electrical Engineering and Computer Science
http://ace.montefiore.ulg.ac.be/







_______________________________________________
gmsh mailing list
[email protected]
http://www.geuz.org/mailman/listinfo/gmsh

Reply via email to