Hi Martin
The probability of getting any feedback on this forum would increase, if you would provide the input files and the timings in the output files of the CELL_OPT and MD runs.
Best
Matthias
--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
cp2k+uns...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/cp2k/3964b36a-2b55-4013-9573-e05ad5f58268n%40googlegroups.com.
Hi Martin
The timings for the core Hamiltonian forces, especially the nonlocal PP part (pnnl), are quite large in the MD run. This is most likely due to the tight EPS_PGF_ORB parameter (1.0E-18). Remove that parameter in the MD input and just rely on EPS_DEFAULT like in the CEL_OPT run. The *_CUTOFF parameters in the &XC section are also dispensable.
HTH
Matthias
--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
cp2k+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/e9f76caa-dbce-49a3-a778-17912e6c2915n%40googlegroups.com.