[Pw_forum] (no subject)

Xijun Wang xijunw at gmail.com
Wed Sep 7 18:12:03 CEST 2011


Hi, Winfred,

I just tried you input file and I found you have three errors in you input file:

(1) You coordinates must not be {crystal}, see manual:

          crystal : atomic positions are in crystal coordinates, i.e.
          in relative coordinates of the primitive lattice vectors

This means crystal coordinates must be between 0 and 1.

(2) Your K_POINTS card should be:  K_POINTS {automatic}
But you use:  K_POINTS (automatic)

(3) Please remove the following lines:
pot_extrapolation = 'second order'
wfc_extrapolation = 'second order'

These parameters are for 'damp' algorithm of relaxation. But you are
using 'bfgs'. Please refer to the manual/

What's more, there are several extra blank lines somewhere in you
input file, I do not know how these will be handled, but remove them
if they are not necessary.

I suggest,

(1) Correct these errors and try it again;
(2) If it still crashes, provide the output whatever it come for
further discussion;

PS: I highly suggest you start from small K_POINTS, 4*4*4 is too
large, so you need quite a lot of memories. It may crash if memory
could not be allocated to the calculation. What's your hardware
configuration (Linux desktop or cluster? How much memory available? )

Xijun


On Wed, Sep 7, 2011 at 7:57 AM, Winfred Mulwa <mulwawinfred at gmail.com> wrote:
>   Dear all,
>
>         I have tried doing a vc-relax on the attached TiO2 rutile supercell,
> but the calculations stop after a few seconds without
> giving any error.  What might be the problem?
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
>



-- 
Department of Chemistry and Biochemistry, Concordia University
7141 Sherbrooke Ouest, Montreal, QC H4B 1R6
Tel: 514-848-2424-#5835 (Lab) TA office: SP175.23



More information about the users mailing list