The ABINIT Users Mailing List ( CLOSED )

[Michel Cote <Michel.Cote@umontreal.ca>]

Dear Alex,

Sorry for the late response. I had a quick look at your input file and  
here are my comments:

-Dataset #1 is set to kptopt=1 which use symmetry to reduce the k- 
point calculated.

-Dataset #2 is set to kptopt=2 which only use time-reversal to reduce  
the number of k-points. This means in your case that you have around  
(6x6x12)/2=216 k-points. Then, you need to set MPI_GROUP_MAX and  
MPI_COMM_MAX to 216 or twice that, I don't remember. Set them to 450  
and that should work for dataset #2.

-Dataset #3 is set to kptopt=3 which means that no symmetry is used to  
reduce the number of k-points and you have to full grid, hence 432 k- 
points. Again, MPI_GROUP_MAX and MPI_COMM_MAX should be set to 432 (or  
twice that).

For your convergence problem, you may try to increase nline to 10 to  
see if it helps.

Let me know how it works.

Michel



Le 07-11-05 à 22:12, a.cote@ucl.ac.uk a écrit :

> Dear Michel,
>
> Here's my input file, I tried to make it as short as possible.
> Datasets after number 3, also have trouble converging, I just found  
> out.
> This is primarily an MPI communicators problem, (I can only do one  
> dataset
> at a time) but if anyone sees any mistakes in the input file that  
> cause
> trouble converging, please comment.
> ("Potential-based CG line minimization not converged after 13  
> restarts.
>  Action : read the eventual warnings about lack of convergence.
>  Some might be relevant. Otherwise, raise nband.")
>
> I tried to follow the tutorial example, and I have already used this
> format to successfully calculate phonon dispersion relations on
> a different system.
>
> ndtset 22
>
> #Ground state calculation
> getwfk1   0
> kptopt1   1             # Automatic generation of k points, taking
>                          # into account the symmetry
> nqpt1     0
> toldfe1   1.0d-10       # SCF stopping criterion
> rfphon1   0
>
> #Q vectors for all datasets
>
>     nqpt   1            # One qpt for each dataset (only 0 or 1  
> allowed)
>                         # This is the default for all datasets and  
> must
>                         #  be explicitly turned off for dataset 1.
>     qpt2        0.00000000E+00  0.00000000E+00  0.00000000E+00
>     qpt3        0.00000000E+00  0.00000000E+00  0.00000000E+00
>     qpt4        3.33333333E-01  0.00000000E+00  0.00000000E+00
> ...............................
>     qpt22       -3.33333333E-01  3.33333333E-01  5.00000000E-01
>
> #Response Function calculation : d/dk
> rfelfd2   2             # Activate the calculation of the d/dk  
> perturbation
> rfphon2    0            # Cancel default
> kptopt2   2             # Automatic generation of k points,
>                          # using only the time-reversal symmetry to  
> decrease
>                          # the size of the k point set.
>   iscf2  -3             # The d/dk perturbation must be treated
>                          # in a non-self-consistent way
> toldfe2   0.0           # Cancel default for d/dk
> tolwfr2   1.0d-22       # Must use tolwfr for non-self-consistent  
> calculations
>                          # Here, the value of tolwfr is very low.
>
> #Response Function calculation : electric field perturbation and  
> phonons
> rfelfd3   3             # Activate the calculation of the electric  
> field perturbation
>
> getddk3   2             # Uses as input ddk wfs the output of the  
> dataset 2
>
> kptopt3   2             # Automatic generation of k points,
>                          # using only the time-reversal symmetry to  
> decrease
>                          # the size of the k point set
> #Sets 4-10 : Finite-wave-vector phonon calculations (defaults for  
> all datasets)
> getwfk   1
> kptopt   3
> rfphon   1
> rfatpol  1 8
> rfdir    1 1 1
> toldfe   5.0d-8
>
> # Common variables
> iscf   5
> #spin related quantities (only second dataset)
> spinat   0.0 0.0 0.0
>         ........
>          0.0 0.0 2.0
> nsppol   2
> #generate the total density of states in both cases
>
> acell  9.08 9.08 4.54
> ecut   65.0
>  natom   8
>  nband   50
>  ngkpt   6 6 12
> nshiftk   1
>  nstep   100
> ntypat   2
> occopt   3
> shiftk   0.5 0.5 0.5
> tsmear   0.01
>  typat   1 7*2
>   xred   0.0  0.0  0.0
>           ......
>          3/4  1/4  1/4
> znucl    14 26 26 26 26 26 26 26  #Si and Fe
>
> Thank you for your comments!!
>
> Alex
>
>
> > Hello Alex,
>
> > Can you post your input file?
>
> > Michel

_____________________________________
Michel Cote
Departement de physique
Universite de Montreal
emai: Michel.Cote@umontreal.ca
web: http://www.phys.umontreal.ca/~michel_cote