On Mon, Oct 12, 2015 at 9:36 AM, Roy Stogner <royst...@ices.utexas.edu> wrote: > > > From mesh_refinement.h: >> > > /** > * If \p edge_level_mismatch_limit is set to a nonzero value, then > * refinement and coarsening will produce meshes in which the > * refinement level of two edge neighbors will not differ by more than > * that limit. If \p edge_level_mismatch_limit is 0, then level > * differences will be unlimited. > * > * \p edge_level_mismatch_limit is 0 by default. > */ > (and likewise for node_level_mismatch_limit) >
Derp. Sorry, and thanks. > So yes, here "0" means "infinity". OK, so we think this is the problem with the over/underrefined_boundary_limit. In that case, we want "0" to really mean no mismatch and, IMHO, we should support that case for over/underrefined_boundary_limit. We can/will cook up a PR, but any comments before we do so? > Seems counter-intuitive, but it > works since there's no other sensible meaning for "0" to have - anyone > who wants to enforce "no mismatch at all" should be using > refine_uniformly(). > As an aside, I'm not sure I agree with the idea of no mismatch ==> you should do uniform refinement. In fact, we had 1 user ask for this some time ago (don't ask me how long, I don't have any sense of time anymore). Sure, it will force refinements of lots of nonlocal neighbors, but it may not be so bad as you're better off just doing uniform refinement (although I'm sure there are cases where that's true). Nevertheless, I'm not as familiar with these code paths, so if it requires lots of changes, I don't think it's worth our trouble (but I wouldn't turn someone away that was dead set on implementing it for us). Best, Paul
------------------------------------------------------------------------------
_______________________________________________ Libmesh-devel mailing list Libmesh-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/libmesh-devel
