[Pw_forum] GIPAW, write tensor field, xsf

[Tone Kokalj]

On Thu, 2009-11-19 at 15:59 -0500, Davide Ceresoli wrote:
> Goranka Bilalbegovic wrote:
> > 
> > Hello,
> > 
> > I have tested two systems: my example of a cluster and the GIPAW CH4 
> > molecule example.
> > 
> > The error is:
> > ---
> > ERROR: while executing exec
> > .../bin/gengeom 0 1 11 1 1 1 1
> > XC_gengeom.1684
> > /home/.../xcrys_tmp/xc_1684/FigFillcurr.xsf_UP_X.xsf.raw
> The problem could be that xcrysden cannot handle such an
> amount of "atoms". In fact, on my machine, gengeom crashed
> with a "segmentation violation" error.

This is precisely the case! Due to some awkward reasons parts of
xcrysden are written in plain fortran-77, and therein the current limits
for the number of atoms in the unit cell are 50000 and the total number
of atoms of 200000 !!!

Goranka, this is the reason why it crashed when you loaded "crystal" XSF
and not for the "molecular" XSF.

It is very easy to fix this problem. You need to recompile the fortran
part of xcrysden, that is located in the F/ subdirectory. Since it is a
plain fortran-77 the compilation is very straightforward. Make sure to
increase the maximum number of atoms in file F/param.inc (variable
"nac").

It is also true that 100.000 field arrows (& dummy atoms X) is a 
huge number of objects. Make sure you open a file on a powerful computer
with good graphics card, or else xcrysden's behavior will be
stalled-like.

Regards, Tone
-- 
Anton Kokalj
J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia 
(tel: +386-1-477-3523 // fax:+386-1-477-3822)

Please, if possible, avoid sending me Word or PowerPoint attachments.
See:  http://www.gnu.org/philosophy/no-word-attachments.html