GROMACS version: 2021.4
GROMACS modification: No
Here post your question
Hi, Groimacs users,
I simulate a polymer system comprising ~18000 atoms. In a system there are bonded (bonds, angles, dihedral and improper) interactions and only van-der-waals non-bonded interactions between atoms considered, and no electrostatic interactions (zero partial charges) taken into account.
When I used in mdp file pme to taken in account electrostatic interactions I got warning
" You are using full electrostatics treatment PME for a system without
charges.
This costs a lot of performance for just processing zeros, consider using
Cut-off instead."
However, the performance of simulation with gpu using pme when partial charges are zero is even higher (880 ns/day), as if I use cut-off (660 ns/day) istead pme.
How do you think, is it possible to use pme with zero partial charge to speed up the simulation? Or this simulation when I consider pme without zero partial charges is not the same when the cut-off scheme is used.
Best regards,
Yel21
Could you post the differences between the two mdp files you used?
Hi, hess!
The difference between two mdp files, please, see below
coulombtype = pme; cut-off
Just out of curiousity, is cut-off still slower if you use a small rcoulomb value?