Re: [Pw_forum] dos with hybrid calculations

2016-12-16 Thread Mohammed Khalafalla
Few days back I successfully performed DOS after hybrid scf with occupations = "tetrahedra". This was the case with QE v. 5.4.0 but not with v. 5.0.2, I don't know why. You must carefully choose nqx mesh in view of k points mesh. Regards M Khalafalla Taibah University On Fri, Dec 16, 2016 at 3:17

Re: [Pw_forum] problem with tddfpt: This type of the supercell is not supported

2016-12-16 Thread Timrov Iurii
Dear Davide, > Dear all > When I try to calculate the charge response using turbo_lanczos.x. I obtain > this error: > Error in routine lr_read_wfc (1): This type of the supercell is not > supported Actually this message comes from the routine "shift_d0psi" in TDDFPT/src, and not

Re: [Pw_forum] compilation problem with intel parallel studio 2017

2016-12-16 Thread Stephen Weitzner
Uri, Check out this GitHub project put together by Hans Pabst, who is an Intel engineer I believe. https://github.com/hfp/xconfigure It contains a set of shell scripts for building quantum espresso and ELPA with Intel compilers / Intel MPI leading to high parallel efficiency. Hope that

Re: [Pw_forum] compilation problem with intel parallel studio 2017

2016-12-16 Thread Paolo Giannozzi
I heard that Intel changed once again the way to load mkl libraries ... Paolo On Fri, Dec 16, 2016 at 3:12 PM, Uri Argaman wrote: > I compile QE with: > ./configure compiler=intel MPIF90=mpiifort FC=ifort F77=ifort F90=ifort > CXX=icpc CC=icc MKLROOT=/opt/intel/mkl

[Pw_forum] compilation problem with intel parallel studio 2017

2016-12-16 Thread Uri Argaman
I compile QE with: ./configure compiler=intel MPIF90=mpiifort FC=ifort F77=ifort F90=ifort CXX=icpc CC=icc MKLROOT=/opt/intel/mkl LIBDIR=/opt/intel/mkl/ and it still do not find MKL libraries: checking for dummy main to link with Fortran 77 libraries... none checking for Fortran 77 name-mangling

[Pw_forum] dos with hybrid calculations

2016-12-16 Thread Tiana Davide
Dear all I know that to calculate band structure for hybrid calculation we need to make a scf calculation adding K points with weight 0 in our brillouen zone. I was wondering if the same procedure does make sense for DOS or not. If not is there a way to calculate DOS with hybrid functional?