On 01/17/2018 02:45 PM, Frohne, Joerg wrote:
obviously this is a mistake. I checked the source files which we have used for
results in corresponding paper.
There I have found the following coding:
resid_old = resid;
resid_vector = system_rhs_newton;
resid_vector.compress(VectorOperation::insert);
int is_my_set_changed = (active_set == active_set_old) ? 0 : 1;
int num_changed = Utilities::MPI::sum(is_my_set_changed,
MPI_COMM_WORLD);
if (num_changed == 0 && resid < 1e-8)
break;
active_set_old = active_set;
Unfortunately I have not been working with deal.ii for over two years now.
For this reason I would be glad if someone else could fix this bug.
Otherwise I would need a short manual how to do it by myself.
I can do that for you. Do I understand correctly that the only change
that needs to happen is to move
active_set_old = active_set;
to *after* the if-statement?
Hope you're doing alright! Best
Wolfgang
--
------------------------------------------------------------------------
Wolfgang Bangerth email: bange...@colostate.edu
www: http://www.math.colostate.edu/~bangerth/
--
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see
https://groups.google.com/d/forum/dealii?hl=en
---
You received this message because you are subscribed to the Google Groups "deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to dealii+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
