BUG: ReplicaUniformPatchGrids and "output" command namd-2.13NB

From: Norman Geist (norman.geist_at_uni-greifswald.de)
Date: Thu Jun 07 2018 - 01:56:47 CDT

It seems something is going wrong when writing restart files using the
"output" command when ReplicaUniformPatchGrids is active. I've observed this
with my own variant of

H-REMD where I used CheckpointStore and CheckpointLoad to swap the
coordinates between the Hamiltonians. This runs ok for the first "job" but
the restart files that are written in-between (as in original REMD)

are broken as they contain overlapping water molecules. I've fixed it using
just "output" and "reinitatoms" for coordinate swapping, but the in-memory
solution would't of course be the cleaner solution, although

there's no real performance penalty.

 

For clarity, it was only a dihedral scaling H-REMD, so a scaling of VDW
interactions is not the reason for the overlapping waters, which rather seem
to be a problem with collecting the patches that are probably mixed between
replicas.

 

Norman Geist

This archive was generated by hypermail 2.1.6 : Mon Dec 31 2018 - 23:21:09 CST