[Wien] lapw2 for TELNES.2

2008-02-15 Thread oyama
Dear all:

I have a problem in in lapw2(c)para for TELNES.2.

I use WIEN2k08_3 installed in a cluster machine with 8-node(16-core).
WIEN2k08_3 is compiled with ifort9.1 cc9.1 and intel_mkl9.1.

lawp0para, lapw1para and lapw2para in SCF calculation can be
parallelized without any problem.
lapw1 is also parallelized in TELNES.2.

However, lapw2 can not be parallelized in TELNES.2 although lapw2para
runs without any errors.
I see same problem with using any methods such as k-point parallel,
mpi-parallel or both together.

If you have a solution about this problem, please tell me what I should
do.

Thanks in advance.

Best regards
Takashi

-
Takashi OYAMA
Murata Manufacturing Co., Ltd.
2288 Oshinohara, Yasu, Shiga, 520-2393 JAPAN
Phone: +81-77-586-8309
Fax: +81-77-587-6782
E-mail:oyama at murata.co.jp
-


[Wien] Optical properties

2008-02-15 Thread Rocquefelte
Dear Igor,

You should put the number of k-points related to the full-Brillouin zone 
(i.e. 1 in your case).

Regards

Xavier

Igor Djerdj a ?crit :
 Dear all,
 I am calculating optical properties, and in order to do this I ran scf 
 as well as lapw1 and lapw2-fermi with N(k)=1 or 841 points in 
 irreducibile wedge. In input file for optic inop I should write at the 
 very beginning the number of k points. Which number should I write 
 1, 841 or some third number independent from scf calculation?
 Many thanks,
 Igor
  
 

 ___
 Wien mailing list
 Wien at zeus.theochem.tuwien.ac.at
 http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
   



[Wien] Indium EFG

2008-02-15 Thread Sergio Yanuen Rodriguez
Dear Stefaan

Thank you very much for your help, now I got the case.output2 file, just
one more question do you know why is lapw2 giving an error message when
the flag -efg is used? Even though the case.output2 file seems to be
alright.

Sergio


 4.-Save calculation save_lapw In_scf.

 This you do too early: save_lapw deletes temporary files, which might
 be needed for postprocessing calculations. If you want to save, do
 then at least 1 iteration after saving.

 5.Construct the case.int file as specified.

 This step is not needed for EFG analysis (it's for DOS).

 6.-Modify the case.in2 file by changing on the first line TOT by EFG.
 7.-Run lapw2 (x lapw2 -efg).

 Here is when I get an error message, and I cannot get the case.output2
 file.

 If you change TOT to EFG, you do not need the -efg switch. Either
 leave TOT and use -efg, or change to EFG and use plain 'x lapw2'.

 Stefaan




 Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm

 ___
 Wien mailing list
 Wien at zeus.theochem.tuwien.ac.at
 http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien