[Pw_forum] source of error

josepht at chips.ncsu.edu josepht at chips.ncsu.edu
Mon Jul 12 19:27:15 CEST 2010


Hello,

I don't see a positive termination of the thread.  Has this been solved?

In your nscf calculation prior to running pw4gww, what was nbnd?  And in your pw4gww
input file, what is num_nbnd_first (size of FCM, corresponding to an FCM energy
cutoff, which is the physical parameter of convergence)?  Reduce num_nbnd_first to
something **below** nbnd.  First try something small, like num_nbnds+10, just to see
if the error goes away.

If this works (you should be able to tell if the error goes away without having to
wait for the job to complete - you will see the line "Out of
matrix_wannier_gamma_big" if the offending grid hurdle is cleared), increase 
num_nbnd_first.  Check for convergence.  If you need so many FCM states (i.e.
approx=nbnd) that the same davcio error is again triggered during tests, you will
have to go back to the nscf calculation and increase nbnd.  Better to start with a
really large value of nbnd, IMO.

Joseph Turnbull
NC State University




>>
>> Dear all
>> Using GWW interface in wannier part(CVS version) I run PW4GWW on its input
>> file m but in the midway I see this error.
>> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>>      task #         9
>>      from davcio : error #        99
>>      error while reading from file
>>  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>>
>> Where can i find the reason of such error(in which f90 program)?
>>
>>
>> thanks a lot
>>
>>
>>
>> Ali Kazempour
>>
>>
>>
>> Fritz-Haber-Institut fax : ++49-30-8413 4701
>> der Max-Planck-Gesellschaft
>> Faradayweg 4-6 e-mail: kazempou at fhi-berlin.mpg.de
>> D-14 195 Berlin-Dahlem / German
>>
>>
>>       _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://www.democritos.it/mailman/listinfo/pw_forum
>>
>
>




More information about the Pw_forum mailing list