On 12/28/2017 08:40 AM, Jianan Zhang wrote:
To illustrate, I tried to solve the problem with Ross's exact solution (but
with a larger value of omega) on the mesh generated by Gmsh, Then the results
obtained are obviously wrong. But when I try the same exact solution on the
mesh created by dealii, the results are correct. This demonstrates that dealii
of the current version truly has not so good support for non-rectangular meshes.
It's not deal.II itself, and it's not about non-rectangular meshes either --
it's a bug somewhere in the implementation of the Nedelec element. We've had
similar reports over the years before where something goes wrong with solving
on meshes that have non-standard face orientations. But we've never been able
to find exactly where in the ~4,000 lines of code that makes up the Nedelec
element the problem lies :-(
As Bruno noted, a small testcase is always helpful. I don't think we can
promise that anyone with the skills to find the bug has the time and
inclination to do so (it's a *volunteer* project, after all). But having a
small testcase is always a first step if anyone ever wanted to try to find the
problem.
Best
W.
--
------------------------------------------------------------------------
Wolfgang Bangerth email: [email protected]
www: http://www.math.colostate.edu/~bangerth/
--
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see
https://groups.google.com/d/forum/dealii?hl=en
---
You received this message because you are subscribed to the Google Groups "deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
