I wanted to add some clarification:
I realised that what I called *n_vertices* is missleading, it is not the
number of vertices I want but the number of "nodes per cell". Numerically,
it is however correct if *FE_DGQ(1)*.
More generally, it should be something like *n_nodes =
(fe_eq1.degree+1)
Prof. Kronbichler
First, I would like to thank you for the development of matrix free
library.
Sir, I am extremely thankful for your response. I have modified the
procedure and it's working perfectly.
I am simulating the evolution of Zr-delatH under mechanical loads by
developing a phase field m
I did not check all code in detail, but what I observe is that you write
into the joint_solution vector without ghosts. Now this would likely
result in a run time error (for countinuous elements and in debug mode),
rather than wrong output. But either way, I would recommend you move the
"locall
Dear all
I am using dealii to solve a multi component system has its own fe,
dofHandler, and solution vector. I am using a matrix free implementation
for the entire program.
I am using the Postprocessor class to derive quantities that depend on all
of the components.
The Postprocessor wo
Hei,
I will do that as soon as my re-compilation is done. I re-compiled deal.II
with gcc 11.2, and in that case I did not get a segfault. Thus, I will now
try to narrow down the issue if it might be related to the intel compiler,
or not.
Regards
Matthias Maier schrieb am Donnerstag, 23. Septemb
