The ABINIT Users Mailing List ( CLOSED )

["Josef W. Zwanziger" <jzwanzig@dal.ca>]

You need to have ngkpt from the GS calculation consistent with the ngqpt
you're going to use in anaddb. In addition, changing ngkpt will generate a
different set of qpt's that you have to run. This is discussed in section 2
of the RF2 lesson.

Josef W. Zwanziger                                 
Prof. of Chemistry
Canada Research Chair in NMR Studies of Materials
Director, Atlantic Region Magnetic Resonance Centre
Dalhousie University
Halifax, NS B3H 4J3 
Canada

tel: 902.494.1960     net: jzwanzig@dal.ca
fax: 902.494.1867     web: http://jwz.chem.dal.ca


-----Original Message-----
From: hyello25@gmail.com [mailto:hyello25@gmail.com] 
Sent: August 30, 2007 12:52 PM
To: forum@abinit.org
Subject: [abinit-forum] How to make the phonon band structures more
accurate?


  In RF2 lesson, when running the example, the input file sets ndset=10
using ngkpt=4x4x4. But if i am not satisfied with the phonon band
structures, what can i do except set higher ecut and more stringent tol***
variables?
 
  I want to do more accurate GS calculatio, so I tried to set : 
    ngkpt1  6 6 6
    ngkpt2  6 6 6
    ngkpt3  6 6 6
    ngkpt   4 4 4
  but the program failed when the 4th process starting? Why? 
  Does that mean the GS calculation should have the same 'ngkpt', consistent
with phonon band structures calculation? 

  Can anyone get me some help??

----- Another problem troubled me a lot:
  When i follow the steps after 'tutorial on analysis tools', error appeared
in during installing the ElecStruc_OpenDX:

 cc  -L/export/home/qgr/program/dx/lib_linux -o AddSticks userAddSticks.o
AddSticks.o -fPIC -shared  
/usr/bin/ld: userAddSticks.o: relocation R_X86_64_32S against `a local
symbol' can not be used when making a shared object; recompile with -fPIC
userAddSticks.o: could not read symbols: Bad value
collect2: ld returned 1 exit status
make: *** [AddSticks] Error 1

Thanks in advance~!