This may be related to #149 and needs more testing! But I'm concerned that things may get dicey if one reduces a k-point dimension to 1, i.e. 2x2x1.
Convergence testing will "fix" the problem, but it would be better (if this is actually an issue) to allow the fewest k-points possible for large systems.
This may be related to #149 and needs more testing! But I'm concerned that things may get dicey if one reduces a k-point dimension to 1, i.e. 2x2x1.
Convergence testing will "fix" the problem, but it would be better (if this is actually an issue) to allow the fewest k-points possible for large systems.