[Octopus-devel] Curvilinear coordinates Hartree problem

[Angel Rubio Secades-]

Yes, we should start getting the Au paper, as well as the azobencene.
Concering the curvilinear coordinates, i do agree with the solution of 
using the rotated vectors, still i am not sure how this affects other 
parts of the code (for sure a lot the non-collinearand LS parts).
Cheers
Angel
  On Tue, 
4 Jan 2005, Miguel Marques wrote:

>  Hi all,
>
>  Now I finally understand why Gigy used the basis transformation in this
> paper!!!! I believe that the nicest solution of the problem is to use the
> changed vectors, i.e. the solution you suggested. I don't really know what
> kind of changes this would involve, besides changing every place where
> m%vol_pp(i) appears. And, of course you are right, also the cg to
> diagonalize the hamiltonian has to be changed...
>
>  cheers,
>  m.
>
> P.S. Au6 is not converging... I will still give it another try. In any
> case, we have to start writing the paper.
>
> _______________________________________________
> Octopus-devel mailing list
> Octopus-devel at tddft.org
> http://www.tddft.org/mailman/listinfo/octopus-devel
>

                           San Sebastian/Donostia, Tuesday January 04, 2005


===============================================================================
Dr. Angel Rubio
Dpto. Fisica de Materiales, Facultad de Quimicas   Phone: +34-943018292
U. Pais Vasco,  Centro Mixto CSIC-UPV/EHU and      Fax  : +34-943015600
Donostia International Physics Center (DIPC)       Mail : arubio at sc.ehu.es
Apdo. 1072, 20018 San Sebastian/Donostia. Spain    http://dipc.ehu.es/arubio
===============================================================================