Hello Fuming,
I am Itay, student from Avital Shurki's lab.
We have encountered a problem with VBPT2 calculations:
When we use two different versions of XMVB we get very different results (energies). 70 kcal/mol difference.
When
using the old version of XMVB (2.0 from Jan 2015) we get different
energy than
the energy we get from the new version (2.1 from Sep 2015).
For example,
the difference for the system OH- + CH3F in the reactant state is 70(!)
kcal/mol. Although, xmi, gus and integrals files are identical.
example files are attached.
We will be happy to know which of the versions is correct.
Thank you,
Itay Karach,
from Avital Shurki's lab.