[aspect-devel] Limiting dofs with AMR

[Timo Heister]

> It seems like the "Coarsening fraction" and "Refinement fraction"
> parameters don't have the effect I would expect. I want to stabilize
> the number of potential degrees of freedom by setting these parameters
> equal to one another. Intuitively, I would expect the number of cells
> to stay the same after a refinement in such a case. Instead, the
> number of cells seems to grow.

Sadly, this is more difficult and much harder to do in parallel. Even
if you manage to mark the "correct" number of cells that would keep
the number of DoFs constant, you can not guarantee that this will
indeed happen in a parallel computation. To keep the mesh consistent,
additional cells might be refined.

In ASPECT we are using refine_and_coarsen_fixed_fraction() which is
not at all doing what you want: it will try to refine the cells that
make up x% of the total error. You could experiment with
refine_and_coarsen_fixed_number.


-- 
Timo Heister
http://www.math.clemson.edu/~heister/