The ABINIT Users Mailing List ( CLOSED )

[Xavier Gonze <gonze@pcpm.ucl.ac.be>]

Dear Ron,

bzr is only for hard-core developers ... Please, follow the usual
http://www.abinit.org/Infos_v5.4/install_notes/install.html#get% 
20basic_compile
for your use of ABINITv5.4.2 . Then, implement the small  
modifications of vtorho.F90
mentioned in my mail, and then run the input file.

Xavier

On 19 Jun 2007, at 21:00, Ronald Cohen wrote:

> I have downloaded and installed bzr, but do not find working  
> directions on downloading 5.4.2 to try this test.
>
> Ron
>
> On Jun 18, 2007, at 10:28 AM, Xavier Gonze wrote:
>
> > Dear Ron,
> >
> > As far as I can judge, I have succeeded to output it by the  
> > modifications of the routine vtorho
> > attached below.
> > And I have attached also an input file, derived from the one you  
> > sent me
> > (ecut and some other input variables have been modified for  
> > speed ; I have two datasets, one for self-consistent, then
> > non-self-consistent run ; ixc has been set to 1 - also in the psps).
> > For dataset 2 , the kinetic energy is written in the log  file  
> > (not in the output file), see line 1209 :
> >
> > -5.9182E-01 -5.9182E-01 -5.3070E-01 -5.3070E-01
> >   vtorho : kinetic energy=    40.81174384327724
> >   afterscfloop : fred=    0.000000000000000         0.000000000000000
> >
> > Take the last printed value (it is incremented after each k point).
> > In the official ABINIT version v5.4.3, I will make the output nicer.
> >
> > Please, test carefully, to be sure everything is coherent ...
> >
> > Best regards,
> > Xavier
> >
> >
> >
> > On 13 Jun 2007, at 03:34, Ronald Cohen wrote:
> >
> > > Yes, exactly. That is what I want to do! So far I have not been  
> > > able to see how to do so.
> > >
> > > Sincerely,
> > >
> > > Ron
> >
> > Modifications as for v5.4.2 (should be OK with v5.3.x also,  
> > provided you find the
> > correct location).
> >
> > --- orig/src/18seqpar/vtorho.F90
> > +++ mod/src/18seqpar/vtorho.F90
> > @@ -1094,7 +1094,7 @@
> >      enlnk(1+bdtot_index : nband_k+bdtot_index) = enl_k(:)
> >     end if
> > -   if(iscf>0)then
> > +   if(iscf>0 .or. iscf==-3 )then
> > !   Accumulate sum over k points for band, nonlocal and kinetic  
> > energies,
> > !   also accumulate gradients of Enonlocal:
> >      do iband=1,nband_k
> > @@ -1110,6 +1110,9 @@
> >        end if
> >       end if
> >      end do
> > +!DEBUG
> > +    write(6,*)' vtorho : kinetic energy=',energies%e_kinetic
> > +!ENDDEBUG
> >     end if
> >     deallocate(eig_k,ek_k,grnl_k,occ_k,resid_k,ylm_k,zshift)
> >     if (psps%usepaw==1) then
> >
> > <input>
>
> Ronald Cohen
> Geophysical Laboratory
> Carnegie Institution of Washington
> http://www.gl.ciw.edu/~cohen
> rcohen@ciw.edu
>
>
> <Ronald Cohen.vcf>