The ABINIT Users Mailing List ( CLOSED )

[<xd0533@163.com>]

Dear Abinit users and developers,
  I used Methfessel-Paxton method (occopt=6) to do the response caluculation
for a transition metal. I chose the width of smearing (tsmear) carefully to
ensure that the difference between the free energy and the total energy, i.e.
the term 'entropy TS' in 'out' file,  is smaller than 1meV/at. But whenever I
use a large k-point mesh, for example 12*12*12 or 10*10*10, the calculation
stops with various errors. My question is that if there are any other special
criteria in choosing a proper k-point mesh in this  kind of calculation or it
is only a matter of convergence.

Thanks a lot!

                                                                              
 
                       Sincerely,
                                                                              
 
                       Tom