[aspect-devel] Follow-up Question from Harry Levinson regarding reproducing the Solcx benchmark

[Wolfgang Bangerth]

Harry,

>           I think those are the parameters I was looking for.  But I’m having
> trouble figuring out which values to assign to them to setup the finite
> element types I listed in my previous email.  I haven’t been able to locate a
> table which indicates the value I should set each polynomial degree parameter
> to in order to setup a particular finite element.

For Q2 x Q1, you'd set the velocity degree to 2. For Q3 x Q2, you'd set it to 
3. For Q2 x P-1, you need to set the "Use locally conservative discretization" 
parameter.


>           I should have mentioned earlier that I’m trying to reproduce the
> Solcx benchmark results from a paper that you co-authored, “High accuracy
> mantle convection simulation through numerical methods” by Martin Kronbichler,
> Timo Heister, and Wolfgang Bangerth.  (I’m referring to the table at the
> bottom of page 10).

You probably also want to take a look at the benchmarks/solcx directory, which 
has all of the files necessary to set up this benchmark.

Best
  W.


-- 
------------------------------------------------------------------------
Wolfgang Bangerth               email:            bangerth at math.tamu.edu
                                 www: http://www.math.tamu.edu/~bangerth/