os1

Hi, I already did so and chose very conservative parameters: #!/bin/bash export OMP_NUM_THREADS=1 export OMP_PROC_BIND=false export OMP_STACKSIZE=64M ulimit -Ss unlimited ~/elk-9.2.12/src/elk But the error persists: +---------------------------------+ | Elk code version 9.2.12 started | +---------------------------------+ Error(readinput): error opening elk.in

Hi all, I successfully updated from version 6 to version 9. However, all the make test examples fail and I cannot run the program. This is the error message: Running test in directory test_001... Program received signal SIGILL: Illegal instruction. Backtrace for this error: #0 0x7a0e28c5c70f in ??? #1 0x7a0e2a2d1765 in ??? #2 0x7a0e29b093be in ??? #3 0x61b01f54c06c in ??? #4 0x61b01f584792 in ??? #5 0x61b01f558cd7 in ??? #6 0x61b01f708e86 in ??? #7 0x61b01f71bbd8 in ??? #8 0x61b01f54dcc8 in ??? #9...

On my 23 inch screen all the chess pieces are pixelated besides the designs ascii, merida1, paul's. Personally, I would prefer alpha, burnett or merida3. Is there any chance that I could contribute with pieces in higher resolution? Why are the other designs not pixelated? Is it the same format or just exported at a higher dpi size?

Hi all, I use the following make.inc file: MAKE = make F90 = gfortran F90_OPTS = -O3 -ffast-math -funroll-loops -fopenmp -fallow-argument-mismatch F77 = gfortran F77_OPTS = -O3 -ffast-math -funroll-loops -fopenmp -fallow-argument-mismatch AR = ar LIB_SYS = # LAPACK and BLAS libraries LIB_LPK = lapack.a blas.a LIB_FFT = fftlib.a SRC_OMP = I did not change all the other parameters. I use Arch Linux with Gfortran 11.2.0-4. When I execute "make test". I get a couple of errors: make test cd tests; ./test.sh...

Hi everybody, Is possible to split up the Eliashberg function into its contributions from different phonon branches, for example into different transverse and longitudinal phonons? One example would be Fig. 4 of DOI: 10.1103/PhysRevX.6.021003 Thank you and best regards, O.S.

Hello everybody, I have a case where a smaller ssxc (spin-scaling parameter) leads to a (little) higher magnetic moment than a higher ssxc. Is this a convergence issue, or could that be physically possible? Best regards and thanks in advance!

Under Main authors and contributors (http://elk.sourceforge.net/) there is a person, named Pink Floyd, listed. I want to use this list as a reference in BibTeX. Is this a joke or a nickname or can I trust this list?

Hello everybody, I did a spin spiral calculation and everthing went fine. I know that spin spiral + SOC is not implented/doesn't work. I wanted to ask, if there is a possibility to add SOC to a spin spiral calculation non-self-consistently, i.e. as a perturbation/ single shot. I thought about using task 1 to continue a fully converged spin spiral calculation without SOC: my idea was to remove the spin spiral and the magnetic field from the elk.in file and to set spinorb.=true Could that work out?...