Open GPMueller opened 5 years ago
This would definitely be a good idea, as I can see now.
The API would grow a little, but it would make a lot of code unnecessary, such as the split Calculate_Force
and Calculate_Force_Virtual
methods.
Note that these changes would also propagate into the GUI, making it less confusing (LLG+VP does not simulate LLG...).
While the velocity projection (VP) solver technically solves an equation of motion, the nonlinear conjugate gradients method is only good for minimisations. Both can only minimise the energy
It might be better to split a
Method_Optimizer
from theMethod_Solver
, and distinguishLLG
andEnergy Minimisation
as separate methods, at least in the GUI. It would remain to be seen whether this makes sense for the code structure, but it might make the LLG code less confusing in terms of the different forces.Question: How to still be able to use the solvers for energy minimisation in this model?