[Pw_forum] ph.x v3.2 on NEC SX-8

[wlyim at puccini.che.pitt.edu]

On Thu, 4 Jan 2007, Paolo Giannozzi wrote:

> 
> On Dec 28, 2006, at 23:00 , wlyim at puccini.che.pitt.edu wrote:
> 
> >  the NEC executables pass a larger "nrxx" value, 22200 in NEC vs 20736
> > in Intel, given that nr1=24,nr2=24,nr3=36. So in NEC, some zero "zeta"
> > were passed to dmxc_spin subroutine which led to "divide by zero"  
> > error
> > at line 1192 in Modules/functionals.f90.
> 
> this is very unlikely. At line 1192 there is a division by rho
> (= rho up + rho down), but rho > small = 1.d-30 (there is a check
> at the beginning of the function). There is also a division by another
> quantity (dz) that is never smaller than 1.d-6 by construction.
> You may try to increase the value of "small" to something less
> small, e.g. 1.d-10

I've checked, from ir=20737 to 22200, dz is zero. I've also tried to add a 
line in functionals.f90:
        zeta = (rhoup - rhodw) / rhotot
        if (zeta <= small) return      ! added line

ph.x can continue but give wrong result. 

> 
> Paolo
> ---
> Paolo Giannozzi, Democritos and University of Udine, Italy
> 
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
> 

-- 
Dr. Wai-Leung Yim
Institut fuer Reine und Angewandte Chemie,
Theoretische Chemie,
Carl von Ossiezky Universtaet Oldenburg,
26129 Oldenburg,
Germany
Email: wlyim at puccini.che.pitt.edu             
Phone:	+49-441-798-3950 (office)              
	+49-441-798-5102 (home)                
Fax:	+49-441-798-3964