Hi Maciek
Not being a particular expert in Gmsh but we have looked at the issue of
scaling in some detail for our calculations. Without knowing exactly
what happened during your meshing it is difficult to say what went wrong,
however we have found that as mesh sizes get larger, problems do arise
due to stability during meshing and optimisation.
We have found that using the latest version of Gmsh from the svn
repository helps improve Netgen stability in addition to breaking the
domain into sections/volumes helps greatly to reduce memory errors in
the optimisation step. As a way of example we have managed to produce meshes
of over 500 million elements however this is a simple box and uses a lot of
memory: 256GB and
a run time of 12 days or so.
Gmsh output:
Info : Done optimizing 3D mesh with Netgen (1.06872e+06 s)
Info : 93150259 vertices 552715749 elements
Regards
Andre
On 03/04/2012 00:35, M Klemm wrote:
Dear All
I'm trying to create tetrahedral mesh of my structure, but it seems
that I can't exceed certain size. Is there a limit on the maximum
number of tetrahedrons which GMSH can generate?
Many thanks for help,
Maciek
----------------------
Maciej Klemm
Senior Research Fellow, PhD (ETH)
EPSRC Career Acceleration Fellow 2010
Centre for Communications Research
Department of Electrical and Electric Engineering
University of Bristol
Bristol, BS8 1UB
Tel: +44 117 331 5160
E-mail: [email protected]
_______________________________________________
gmsh mailing list
[email protected]
http://www.geuz.org/mailman/listinfo/gmsh
_______________________________________________
gmsh mailing list
[email protected]
http://www.geuz.org/mailman/listinfo/gmsh
