The ABINIT Users Mailing List ( CLOSED )

["matthieu verstraete" <matthieu.jean.verstraete@gmail.com>]

Hi everyone: here's my 2 cents

I have seen this as well, and it's certainly not a PAW effect.

The effect of ecutsm is clear: when it is finite, the real cutoff energy is higher than ecut. Even though the last pw are damped they are contributing (for the lowest ecut the effect is strongest). So to keep the same accuracy you need to either have such a large ecut that ecutsm is negligible (any indication that for higher ecuts the effect disappears?) or keep ecutsm as is. Does ecutsm affect the calculation of the stresses or only the energy? If it does this implies that the last pw have a large contribution to the stress... which in turn implies we are not really converged wrt ecut as far as the stress calculation is concerned

 At any rate I find the calculation of small pressures in DFT fairly dubious. The intrinsic error wrt experiment, eg in P(V), is larger than 1GPa in many cases, even if the numerical error is 10^-6.

More surprising is the effect of dilatmx (setting it to 1 still makes a difference). Could there be other arrays which are re-dimensioned due to dilatmx?

bye

Matthieu


--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Dr. Matthieu Verstraete

European Theoretical Spectroscopy Facility (ETSF) 
Dpto. Fisica de Materiales, 
U. del Pais Vasco,            
Centro Joxe Mari Korta, Av. de Tolosa, 72,   Phone: +34-943018393
E-20018 Donostia-San Sebastian, Spain        Fax  : +34-943018390

Mail : matthieu.jean.verstraete@gmail.com
http://www-users.york.ac.uk/~mjv500