[Columbus] (no subject)

[Ron Shepard]

On Sep 28, 2005, at 10:09 AM, Andy Young wrote:

> I was wondering why the "FCORE" option in the mcscf
> has been disabled.  The mcscf.txt file in docs/fulldoc
> says that it was disabled but it does not indicate why
> or if it can be re-enabled.

I have forgotten the entire history of this, but at one time the MCSCF 
program allowed core orbitals to be frozen (meaning they were not 
optimized during the orbital optimization process).   I don't remember 
why, but the option was dropped.  Now if you want to freeze core 
orbitals, run the integrals through the TRAN program to transform away 
the frozen core orbitals, and use the output integrals as input to the 
MCSCF program.  The frozen core contributions to the hamiltonian and to 
any other property arrays that are on the integral file should be 
accounted for correctly during the transformation step.  That is, the 
total energy, total dipole moment, etc. should be the same with or 
without the frozen core step.

-Ron