Stephen DeWitt
Stephen DeWitt
By default, deal.II outputs only the nodal values. However, for higher order elements this makes the solution appear choppier than it is in reality. In PRISMS-PF we compensate for this...
A few ideas (not all will make v2.2): - Give a warning if the grains cannot all be reassigned without overlap - Store the bounds of the grain more efficiently...
Currently, if one has in the parameters file to read the initial conditions from a vtk file, but that file doesn't exist, the simulation hangs. It should throw an exception...
This method is getting too large. There should be separate functions for the explicit update, nonlinear solver, and fully time-independent solves.
We've had a non-functional implementation of non-zero Neumann BCs for a while. What I didn't realize until now is that some coefficients (and the time step size for a time-dependent...