Thanks for your suggestion. Please, let me to give more information about the problem.
My system have 99 atoms This problem did not appear when the code calculated 162 k points. Now I need 452 k points and I run the QE. The nscf output file do not give any information about some error. The file indicated me that the 452 k points were calculated, but the last lines of nscf output file are: the Fermi energy is -0.599772 ev Writing output data file work_electron_ao_98C1F.save They are the last lines of the nscf file. I think that the directory work_electron_ao_98C1F.save did not write correctly and this is the problem to get projected density of states, And when I revised what was written in the terminal when the calculation was finished. # FROM IOTK LIBRARY, VERSION 1.2.0 # UNRECOVERABLE ERROR (ierr=1) # ERROR IN: iotk_write_dat (iotk_dat+COMPLEX2_0.f90:1013) # CVS Revision: 1.27 I am following your suggestion, I am running the same input of 452 k points in the newest version of the QE. If you have some advice to me, please let me to know. Thanks On 2015-01-25 05:47, Paolo Giannozzi wrote: > On Sat, 2015-01-24 at 19:22 -0300, robert.guzman wrote: > > A possible solution would be to use another compiler? On my case I am > using ifort. > > ifort shouldn't give miscompilation problems. Maybe there is a better > explanation. You should first of all try if the error persists in the > newest version (5.1.1). Then please provide a case that can be easily > run and displays the problem. > > Paolo > > I saw that inside espresso directory exist fortran compilers, maybe > these compilers do not repeat this problem, That would be true? > At the moment I am making calculation one more time with espresso-4-2, > would be espresso-4-2 more stable than espresso-5-1 in this situation? > > Thank for your attention. > > On 2015-01-24 12:20, Paolo Giannozzi wrote: > > On Sat, 2015-01-24 at 02:11 -0300, robert.guzman wrote: > > > > # FROM IOTK LIBRARY, VERSION 1.2.0 > > # UNRECOVERABLE ERROR (ierr=1) > > # ERROR IN: iotk_write_dat (iotk_dat+COMPLEX2_0.f90:1013) > > # CVS Revision: 1.27 > > > > almost invariably, iotk errors that do not have an obvious > > explanation are due to compilers that do not compile iotk > > properly. > > > > P. > > > > And when I try to calculate the projected density of states, inside of > > the output file was written: > > > > Error in routine pw_readfile (1): > > error reading header of xml data file > > > > Thus the projected density of states had not been calculated. > > > > Please, Where is my error? > > > > Thanks! > > > > R.M. Guzman Arellano > > Instituto Balseiro. > > _______________________________________________ > > Pw_forum mailing list > > [email protected] > > http://pwscf.org/mailman/listinfo/pw_forum > > > > > > _______________________________________________ > > Pw_forum mailing list > > [email protected] > > http://pwscf.org/mailman/listinfo/pw_forum > _______________________________________________ > Pw_forum mailing list > [email protected] > http://pwscf.org/mailman/listinfo/pw_forum > > > _______________________________________________ > Pw_forum mailing list > [email protected] > http://pwscf.org/mailman/listinfo/pw_forum _______________________________________________ Pw_forum mailing list [email protected] http://pwscf.org/mailman/listinfo/pw_forum
