[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 Jan 4, 2007, at 12:17 , wlyim at puccini.che.pitt.edu wrote:
> 
> >> There is also a division by another quantity (dz) that is never  
> >> smaller
> >> than 1.d-6 by construction.
> 
> actually I misinterpreted the definition of dz: dz is never LARGER then
> 1.d-6 by construction. You can simply set dz to a fixed value: 1.d-6, or
> 1.d-4 .
> 

Dear Prof. Giannozzi,

I tried your suggestion and set dz = 1.d-6 in functionals.f90:

     dz = min (1.d-6, 1.d-4 * abs (zeta) )
     if (dz <= small) dz = 1.d-6          ! added line
     call xc_spin (rho, zeta - dz, ex, ec, vxupm, vxdwm, vcupm, vcdwm)
     call xc_spin (rho, zeta + dz, ex, ec, vxupp, vxdwp, vcupp, vcdwp)

I hope I added it at the right point. Unfortunately, ph.x still gave wrong
results:
     q = (    0.000000000   0.000000000   0.000000000 ) 
**************************************************************************
     omega( 1) =     -40.084193 [THz] =   -1337.073655 [cm-1]
     omega( 2) =     -39.071262 [THz] =   -1303.285663 [cm-1]
     omega( 3) =     -39.071262 [THz] =   -1303.285663 [cm-1]
     omega( 4) =     -38.838931 [THz] =   -1295.535888 [cm-1]
     omega( 5) =     -38.070039 [THz] =   -1269.888241 [cm-1]
     omega( 6) =     -38.070039 [THz] =   -1269.888241 [cm-1]
**************************************************************************

ph-2.1.4, complied with INTEL and LINUX64 flags and internal FFTW, gave
the results below (which is reasonable):
     q = (    0.000000000   0.000000000   0.000000000 )
**************************************************************************
     omega( 1) =      -0.439781 [THz] =     -14.669624 [cm-1]
     omega( 2) =       0.344752 [THz] =      11.499781 [cm-1]
     omega( 3) =       0.344752 [THz] =      11.499781 [cm-1]
     omega( 4) =       3.590255 [THz] =     119.758805 [cm-1]
     omega( 5) =       3.590255 [THz] =     119.758805 [cm-1]
     omega( 6) =       7.870822 [THz] =     262.544104 [cm-1]
**************************************************************************

Is there any benchmark test on NEC machine for the latest version of 
espresso?

Best regards,
William

-- 
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