[Wien] Lapw1 for down spins crashed

[Laurence Marks]

I believe this means that the Hessian is not positive definite, i.e.
it is singular because two (or more) of the entries are the same.
Possible sources include:

a) Duplicate energies in case.in1, e.g.
 0   -0.0225    0.000 CONT 1
 0   -5.2390    0.000 CONT 1
 0   -5.2390    0.000 CONT 1

b) Accidental degeneracy between the LO components for a given atom,
same L. If you look in case.output1_?? at the lines which start
LO COEFFICIENT: l,A,B,C
you may find that the A,B,C values are identical (to numerical error),
and I think this also means that the corresponding U(R) (a couple of
lines above) will be close to zero. If you get this you have to shift
the linearization energy by about 0.05 and/or switch the energies of
the two.

c) Some other error in input files.

On 4/25/07, XU ZUO <xzuo at nankai.edu.cn> wrote:
> Dear Colleagues,
>
> I am using Wien2k_07.2 to calculate DOS of FeCr2S4. Lapw1 for down spins
> crashed in the 1st cycle. The error message is following:
>
> **  Error in Parallel LAPW1
> **  LAPW1 STOPPED at Wed Apr 25 05:02:56 EDT 2007
> **  check ERROR FILES!
>  Cholesky INFO =         1168
>  'SECLR4' - POTRF (Scalapack/LAPACK) failed.
>  Cholesky INFO =         1140
>  'SECLR4' - POTRF (Scalapack/LAPACK) failed.
>
> Any idea about this error? Thanks a lot.
>
> Xu Zuo
>
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>


-- 
Laurence Marks
Department of Materials Science and Engineering
MSE Rm 2036 Cook Hall
2220 N Campus Drive
Northwestern University
Evanston, IL 60208, USA
Tel: (847) 491-3996 Fax: (847) 491-7820
email: L-marks at northwestern dot edu
Web: www.numis.northwestern.edu
EMM2007 http://ns.crys.ras.ru/EMMM07/
Commission on Electron Diffraction of IUCR
www.numis.northwestern.edu/IUCR_CED