[Gmsh] gmsh 2.0

Christophe Geuzaine cgeuzaine at ulg.ac.be
Tue Feb 6 08:39:37 CET 2007

Dmitry Firsov wrote:
> I already tested that algorithm. You know... it doesn't work :'( . 
> Program crashes with Delaney mesher without any warnings :'( .
> I took a look on to your source codes, seems to be that you have still 
> isotropic mesher. That will be grate if you will recompile the gmsh with 
> that algorithm...

PS: looking at your file, the fact that this geometry could even get
meshed with the old algorithm is surprising, as we never intended for 
embedded surfaces to work...

Feel free to use the old version for now--we definitely plan to add
"real" support for embedded surfaces in the future. (Embedded curves are
already supported in Gmsh 2.0; they were not supported in 1.65.)



> Christophe Geuzaine wrote:
>> Dmitry Firsov wrote:
>>> Dear GMSH Developers,
>>> Probably that was not grate idea to remove isotropic algorithm from 
>>> gmsh 2.0. I was able to generate with isotropic algorithm meshes with 
>>> a lot higher quality then Netgen gives (the quality: smallest 
>>> distribution of  area/volume ratio along the mesh).  Also gmsh 2.0 is 
>>> unable to generate a mesh (it is crashing), when gmsh 1.65  with 
>>> isotropic mesh generator gives a good result. Attached geo file shows 
>>> exactly that problem... Of cause isotropic mesh generator sometimes 
>>> is crashing and takes to much CPU resources, but it gives good results.
>> You can test a "beta" of a brand new version of the "Isotropic" 
>> algorithm by choosing
>> Options->Mesh->General->3D algorithm->Delaunay
>> in the GUI. Let us know if that works...
>>> Best,
>>>    Dr. Dmitry Firsov
>>> ------------------------------------------------------------------------
>>> _______________________________________________
>>> gmsh mailing list
>>> gmsh at geuz.org
>>> http://www.geuz.org/mailman/listinfo/gmsh

Prof. Christophe Geuzaine
University of Liege, Electrical Engineering and Computer Science