[Wien] a parallel error of lapw0 with MBJLDA potential (updated)

[Peter Blaha]

> 1. I do not fully understand what you mean that "It is probably
> completely uncritical for the gap,...".
> After the temporary fixed in brj.f, can the code deal with correctly
> the system which have very heavy elements?
>  and their bandgaps are reasonable?

Yes.

> 2. We doubt that there are still some bugs in lapw0_mpi, because the
> bandgap is different,
> 
> Ge:
> 0.85eV  (lapw0)
> 0.71eV  (lapw0_mpi)

There was still a bug in the interstitial region in case you have more
processors than atoms.
It has been fixed and the new version is on the web.

PS: This new version includes also improved W2kutil and W2kinit subroutines,
which also compile under Sun Solaris.
-- 

                                       P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------