You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Robert Q. Topper edited this page Oct 20, 2026
·
1 revision
As helpful hints occur to us, we'll share them here!
Don't use the default search parameters for serious work. The code is shipped with a very quick test run that demonstrates its usage. The parameters in the config file MUST be improved for reliability. We have described parameters that we use in our research papers, cited on the man page.
It is almost always a good idea to use the "finale" option. This will achieve an inefficient, but effective, final geometry optimization.
A single run may not find the minimum-energy structure, and this becomes more and more likely as the system gets bigger. When we have search parameters that we think are reliable, we run 4-6 sets of 10-20 independent annealing runs, and look for certain patterns.
First of all; are multiple runs finding the same structure? If not, there may be an issue with insufficient sampling.
Second of all; if i. is satisfied, are new lowest-energy structures appearing? If so, run some more sets. If not, run one more set just to be sure.