[overture] on multigridding
- From: Joel Guerrero <joegi.geo@xxxxxxxxx>
- To: overture@xxxxxxxxxxxxx
- Date: Thu, 10 Jun 2010 08:55:55 -0700 (PDT)
Hi I have some questions regarding the multigrid solver,
I have been using the multigrid solver for a while and so far it works fine
(for moving and fixed bodies). Now I am trying to use it with the rigid body
motion solver, but I am getting totally different results to those that I get
when using petsc. I am getting these results with a modified version of the
rigid body motion solver, which I already validated with other results and I
obtained a quite good agreement using petsc, but when using the multigrid
solver the solution is totally different. Any clue of what could be the
problem?. I also ran the twodrops case using petsc and multigrid and I got
the same solution for both solvers, so I guess the error could be something
related to the modified version.
The second question is how do I change the solver for the coarser grids?. I am
always getting the same solver, I already tried to changed it by setting the
desired solver in oges parameters but I always get the YALE solver, which does
not perform well for the grid I am using (something about 5 millions points in
the finest mg level)
Regards,
Joel
Other related posts:
