Mathias Davids

Hi @mlstowell , a quick follow-up to your suggestions above: I tried to implement what you wrote in your point 1) but I think I may be missing something. `j_dual`...

Ok, I actually found the implementation in `ex34p.cpp` and realized I may have taken one confusing step that introduced some errors. I think it might be working now. I will...

Hi @mlstowell , just out of curiosity: is this feature mostly usable at the moment? I think it's a really really cool and important addition. For example, in our cases,...

Awesome, thanks for the clarification. I will definitely give this a try and share some results on performance here!

Hi all, I started to modify the Darcy example (`ex5p`) which I believe deals with a similar type of problem (I also realize there is a post on a very...

Hi @mlstowell , thanks a lot for the quick response! Yes, at the moment, the diffusion problem itself is solved on `H1`, and then I am computing the gradient in...

Thanks @mlstowell , that's very helpful! Will give this a try! Also thanks @najlkin for the additional remarks. I will definitely try to catch your talk at the MFEM workshop....

Hi @mlstowell and @najlkin , I took a stab at this, and I think I made a step in the right direction. Using the approach outlined by @mlstowell I get...

Hi @mlstowell and others, Thanks a lot for the input! I will try to give this a try at some point. I realize I should've provided some more info on...

Hi @mlstowell , I guess the wording was bit misleading: with the most recent runs, the solver always converged, although of course much slower for the anisotropic case: - isotropic...