[Wien] NEW1

[Peter Blaha]

deltas have the same meaning, however, they will be internally automatically 
adjusted when the minimization proceeds, so their influence is rather small 
(determines only the first step). eta is not used.

As was mentioned also by L.Marks: When computational parameters are low or scf 
convergence is difficult, forces and total energies may not always be 
consistent. In such cases PORT will not work (one "bad" energy (eg. too low) 
will confuse PORT, since it wants to minimize E-tot, but had a lower (wrong) 
energy before ....

Metallic slabs are notoriously known for bad scf convergence (k-points, ...) 
and thus PORT may not work. NEW1 does not care too much about ONE bad step, 
because it discards older history than the previous one, so it should be 
superior in this case.

However, when force=0, but E-tot is NOT the minimum (or close to the minimum) 
you might have a serious problem and check scf-convergence, FERMI method, 
k-points, RKmax, ...

Are you using the latest WIEN2k version ?

John Appleton schrieb:
> Dear Prof. Blaha,
> I kindly want to know if delta and eta have the
> same meaning is NEW1 as in NEWT. Also if
> for example port fails to relax hygrogen mlecule on Pt slab and 
> one wants to use NEW1, what would be a good set of
> delta and eta since Pt and H have way different masses.  
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien

-- 

                                       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/
--------------------------------------------------------------------------