hanabi

Hi Marcin Thanks, this argument works for me. Indeed, I'm just checking about an error when running with MPI. The previously worked input file doesn't work any more with the following promotion: Warning(rdirac): maximum iterations exceeded Error(rdirac): zero wavefunction I'm not sure why this error comes from when running with MPI processes more than 1. OpenMP works fine by either gfortran or intel fortran. My system is: System version: Mac OS 10.14.5 with Darwin Kernel Version 18.6.0 Compiler version:...

Dear Elk devlopers When I try to compile ELK (6.3.2), the following error message promoted (only when using GCC 10.0) mpif90 -O3 -march=native -mtune=native -fopenmp -g -m64 -I/opt/intel/mkl/include -c zfpack.f90 mpif90 -O3 -march=native -mtune=native -fopenmp -g -m64 -I/opt/intel/mkl/include -c addlorbcnd.f90 zfpack.f90:34:17: 23 | call dcopy(k,zfmt(:,ias),1,v(n+1),1) | 2 ...... 34 | call dcopy(k,v(n+1),1,zfmt(:,ias),1) | 1 Error: Type mismatch between actual argument at (1) and actual argument...

When I try to compile ELK (6.3.2), the following error message promoted (only when using GCC 10.0) mpif90 -O3 -march=native -mtune=native -fopenmp -g -m64 -I/opt/intel/mkl/include -c zfpack.f90 mpif90 -O3 -march=native -mtune=native -fopenmp -g -m64 -I/opt/intel/mkl/include -c addlorbcnd.f90 zfpack.f90:34:17: 23 | call dcopy(k,zfmt(:,ias),1,v(n+1),1) | 2 ...... 34 | call dcopy(k,v(n+1),1,zfmt(:,ias),1) | 1 Error: Type mismatch between actual argument at (1) and actual argument at (2) (REAL(8)/COMPLEX(8))....

Hi Mfechner Thanks for your suggestion, I'd like to use DFT+U as well, but before that I'd prefer to get a ferrimagnetic net moment. I'll try the 5 spins up and 3 down to see what's happen. Best regards.

Hi, I'm trying to calculate the Ferrimagnetism state of GaFeO3, so the first four iron atoms should have a different magnetude of spin moment to the rest of four iron atoms. My configuration is at below tasks 0 xctype 20 0 0 emd_dim 3 nwrite 1 nints 200 tempk 20.0 ngridk 5 5 4 spinpol .true. spinorb .true. cmagz .true. epsengy 0.5E-04 epspot 0.2E-05 mixtype 1 wplot 800 200 3 -0.4 0.2 sppath '/path_to/elk_species/' bfieldc 0.0 0.0 -0.3 reducebf 0.85 avec 0.000000000 0.000000000 9.575676221 16.47712561...

Hi, I'm trying to calculate the Ferrimagnetism state of GaFeO3, so the first four iron atoms should have a different magnetude of spin moment to the rest of four iron atoms. My configuration is at below tasks 0 xctype 20 0 0 emd_dim 3 nwrite 1 nints 200 tempk 20.0 ngridk 5 5 4 spinpol .true. spinorb .true. cmagz .true. epsengy 0.5E-04 epspot 0.2E-05 mixtype 1 wplot 800 200 3 -0.4 0.2 sppath '/path_to/elk_species/' bfieldc 0.0 0.0 -0.3 reducebf 0.85 avec 0.000000000 0.000000000 9.575676221 16.47712561...

Hi, I'm trying to calculate the Ferrimagnetism state of GaFeO3, so the first four iron atoms should have a different magnetude of spin moment to the rest of four iron atoms. My configuration is at below tasks 0 xctype 20 0 0 emd_dim 3 nwrite 1 nints 200 tempk 20.0 ngridk 5 5 4 spinpol .true. spinorb .true. cmagz .true. epsengy 0.5E-04 epspot 0.2E-05 mixtype 1 wplot 800 200 3 -0.4 0.2 sppath '/path_to/elk_species/' bfieldc 0.0 0.0 -0.3 reducebf 0.85 avec 0.000000000 0.000000000 9.575676221 16.47712561...

I think you can use Wannier project feature provided by ELK to get 'fatband'. There is an example with ELK already