UT theoretical chemistry code forum

“We recommend using CG or LBFGS when accurate forces are available. This is essential for evaluating curvatures. For high forces (far from the minimum) or inaccurate forces (close to the minimum) the quick-min or FIRE methods are recommended. These two methods do not rely on curvatures, and tend to be less aggressive, better behaved, but also less efficient than CG/LBFGS. ”

The "minimum" in sentence above does not mean the saddle point? When a CNEB calculation is too close to saddle point, CG/LBFGS is improper, and should be switched to QM?

personally my favorite one is FIRE. use quick min with a small displacement 'maxmove=0.02' for 30 steps until the force is down to ~1ev/A; and then move to more powerful alogrithem with large displacement.