There are compiler options that catch invalid operations (e.g. -fpe0 for Intel compiler). Unfortunately each compiler, and sometimes each compiler version, has its own set.
Paolo On Fri, Jan 26, 2018 at 6:09 AM, zhout2 <[email protected]> wrote: > Dear QE community, > > I was using pwcond.x to do conductance calculations for metals, which were > successful in many systems, but failed in a few others, in which cases the > calculated transmission coefficients are -NAN, as below: > > > to transmit > Band j to band i transmissions and reflections: > j i |T_ij|^2 |R_ij|^2 > > 1 --> 1 -NAN -NAN > 1 --> 2 -NAN -NAN > ... > 1 --> 7 -NAN -NAN > Total T_j, R_j = -NAN -NAN > > As the calculations were not interrupted and no error message was given, I > am struggling to find the cause of error. Could someone offer some insight? > > I have tried qe 5.1, 5.4 and 6.1, yet this error always appeared. The > crystal structure I used should be correct, because for some other elements > the calculations were successful. I'm attaching both scf and pwcond files, > though there was nothing wrong I could notice with the scf step. > > Thanks and best regards, > > Tianji > > > > Tianji Zhou > Department of Materials Science and Engineering > Rensselaer Polytechnic Institute > > _______________________________________________ > Pw_forum mailing list > [email protected] > http://pwscf.org/mailman/listinfo/pw_forum > -- Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche, Univ. Udine, via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222
_______________________________________________ Pw_forum mailing list [email protected] http://pwscf.org/mailman/listinfo/pw_forum
