[CP2K:3837] Re: Printing MO eigenvalues

[Peter Tentscher]

Dear Matt,

thanks for that information.
Things are a bit more complicated than I thought.
Ultimately, what I want is the SOMO energy from
a ROKS calculation. That is, I do need OT in order
to converge to the right solution.
I tried using MO_CUBES in conjunction with OT/ROKS,
but what I get is
"  Unclear how we define MOs in the restricted case ... skipping"
so this doesn't solve my problem.
I also tried to use the *-RESTART.wfn from an OT calculation
as a guess for the corresponding ROKS calculation without
OT. But the subsequent SCF procedure then leads away from
the former OT solution, i.e. raising the energy.

Is there any way to get orbital energies from a ROKS calculation
using OT?

Best,

Peter

On 05/23/2012 10:39 PM, Matt W wrote:
> Hi,
>
> seems you're a victim of some not fully synchronized algorithms in the 
> code.
>
> "Controls the printing of the molecular orbitals.Note that this is 
> only functional with diagonalization based methods, in particular not 
> with OT (see MO_CUBES)"
>
> Turn OT off. Or try pdos or MO_CUBES in the print section of DFT.
>
> Matt
>