[QE-users] Effect of esm_nfit and how to choose it.

[Evgeny Permyakov]

Hello!
I ran a few calculations and found that use of assume_isolated=esm messes
with bfgs somewhat fierce in my system. After skimming the docs, I found
the esm_nfit parameter that says something about polynomial interpolation
with 4 point grid by default. This is a rather small value, especially
considering I use a supercell 18x12x30 A with ~10 rows of atoms in
Z-directions.


Am I right that increasing this value should help? If so, is there any
suggestion for a good initial value? (It seems to me that it should be
something like several tens if not a few hundreds for my system....)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20210630/ee578694/attachment.html>