davidcortesortuno
commented
5 years ago Looks ok. Might be worth checking this functionality in the future to avoid re-calculating the demag tensor in large systems. I guess for now this option is not necessary as the tensor calc should not take long compared to the simulation time
Disable the saving of the demagnetising tensor.
This was causing an odd error where NaN's appeared in the yz component of the Demagnetising tensor. I haven't attempted to fix this yet; but disabling the saving/loading fixed the issue, so it likely needs rearchitecting. There's a test which checks for NaN values in these tensors now, so it shouldn't happen again!