Kari Ruotsalainen

Hi Kay, That did the trick. Thanks for the quick reply. Best, Kari

Hi all, I compiled elk with gfortran from gcc-9.4. I linked it against libxc6 and openblas-0.3.2.1. elk and the libraries passed their built-in tests. The input file at the end of this post produces a segmentation fault at iteration #1 if I enable highq. The error message you find below. Without highq the calculation runs just fine and e.g. the magnetic moments on the Mn atoms are within 1% of values I obtained with pwscf. I also noticed that if I insist on rerunning the same input, sometimes the...

Hi Red, elk can be modified to calculate the matrix elements for emission. I can try to dig out the right file and part to modify for you, but this might take a while as I have the relevant source files on some old backup HD. The developers can probably help you faster. Peter Elliott helped me out with this some years back. Creating a RIXS part that allows you to identify DOS/band structure related features should be pretty straight forward to implement. The transition matrix elements can be calculated...

Hello, I'm having problems compiling elk-7.2.42 on our cluster. I'm using gnu-4.8.1 compilers. The error occurs with splined.f90 and the output is "splined.f90:18.6: !$OMP SIMD LASTPRIVATE(f1,f2,f3,f4) 1 Error: OpenMP directives at (1) may not appear in PURE or ELEMENTAL procedures" I don't recall seeing this error when compiling older versions of elk on the same cluster. My make.inc is essentially the same as in the example for AMD processors and gnu compilers posted above on September 2019. The...