forum@abinit.org
Subject: The ABINIT Users Mailing List ( CLOSED )
List archive
RE: [abinit-forum] No *_DOS output file after a DOS computation w ith tetrahedron method.
Chronological Thread
- From: James Raynolds <JRaynolds@uamail.albany.edu>
- To: "'forum@abinit.org'" <forum@abinit.org>
- Subject: RE: [abinit-forum] No *_DOS output file after a DOS computation w ith tetrahedron method.
- Date: Thu, 3 Apr 2003 08:00:15 -0500
Raphael,
I had the same problem until I migrated to version_4.0.3. It works
correctly to give the density of states. Take a look at Test_v3/t46.in as
an example showing how to set up a job to calculate the DOS with the
tetrahedron method.
For the problem that I'm studying it is also usefull to set up the
optimal k-point grid using the prtkpt 1 option (see Test_v2/t61.in through
Test_v2/t73.in). I'm looking at Si vacancies and substitutional impurities.
I found an error with the tetrahedron method which occurs when you
break the symmetry by removing an atom. The error is in the subrouting
get_full_kgrid.f. The problem has been fixed but wont appear until the next
version however. The fixed subroutine is attached to this e-mail. So
substitute this file for the old get_full_kgrid.f and re-compile.
Beware, Microsoft e-mail often puts extra control characters in the
file which the compiler doesn't like (such as ^M). So you may have to
remove them. Good luck.
Sincerely,
Jim Raynolds
-----Original Message-----
From: raphael.petion@isen.fr [mailto:raphael.petion@isen.fr]
Sent: Thursday, April 03, 2003 5:01 AM
To: forum@abinit.org
Subject: [abinit-forum] No *_DOS output file after a DOS computation
with tetrahedron method.
Dear all,
I recently encountered a strange fact while using the tetrahedron method
(option prtdos 2). Everything was correct, computation had well converged
and I had no errors and even 0 warnings. Good. I look at my work directory,
read the output file, and when I want to get the results of the computation,
I see that... there is no *_DOS file !
With prtdos 1 (smearing), everything goes well. Nevertheless, I think that
tetrahedron method would be more interesting for me.
I'm using a sunblade 100 sparc machine, running under Solaris 8. This is
not due to a lack of disk space or quota. Does someone knows the solution ?
I would greatly appreciate.
Thanks,
Raphael Petion.
Attachment:
fixed_get_full_kgrid.f
Description: Binary data
- RE: [abinit-forum] No *_DOS output file after a DOS computation w ith tetrahedron method., James Raynolds, 04/03/2003
Archive powered by MHonArc 2.6.16.