[Wien] matrix size issue

[Stefaan Cottenier]

Did you use exactly the same strucutre files? One possibility is that  
in the current case.struct you have one of the Rmt's set lower than  
before (misprint). The Rmt in Rmt*Kmax is the lowest one of your  
structure file, hence if in just one atom block you have a typo, the  
result could be a much larger matrix size.

Stefaan


Quoting Monique Body <monique.body at univ-lemans.fr>:

>  dear Wien2k Users,
>
> I have performed calculation on a structure in 2005, with
> RmtKmax = 8,
> Gmax = 14, and
> 28 calculated kpoints.
> The corresponding matrix size was 7990.
>
> I want to perform the same computation with the new version of
> Wien2k, using the sames conditions
> RmtKmax = 8,
> Gmax = 14, and
> 28 calculated kpoints.
> The corresponding matrix size is now equal to 15000 (maximum value
> allowed by NMATMAX) and the RmtKmax is reduced to 5.25 .
>
> How is it possible the matrix size change so much? Is there a
> parameter I forgot to check? Or is it a bug linked to the space
> group?
>
> I encounter this problem on 2 different parallel computers (Nantes
> and Le Mans) but only for this structure (space group n°13 : P2/n).
> In both cases, I start from the .struct file and perform the whole
> initialisation before starting SCF cycles.
>
> Thanks for your help,
>
> Monique
>
> ***********************************************************************
> Monique Body
> Laboratoire de Physique de l'Etat Condensé  33.2.43.83.31.40
> Université du Maine - Avenue Olivier Messiaen
> 72085 Le Mans - FRANCE
> monique.body at univ-lemans.fr
> ***********************************************************************
>



-- 
Stefaan Cottenier
Instituut voor Kern- en Stralingsfysica
K.U.Leuven
Celestijnenlaan 200 D
B-3001 Leuven (Belgium)

tel: + 32 16 32 71 45
fax: + 32 16 32 79 85
e-mail: stefaan.cottenier at fys.kuleuven.be


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