[gmx-users] NS in PME

[Janne Hirvi]

Hello David!

I read your interesting new article "The Origin of Layer Structure Artifacts in
Simulations of Liquid Water" and I would like to know especially what kind of
NS parameters you have used in the simulations with PME and are the following
parameters correct otherwise?

dt=0.002
nstlist=5
rlist=?
coulombtype=PME
rcoulomb=0.9
vdwtype=cut-off
rvdw=?    

I am interested especially about the rlist parameter because there seems to be
some energy conservation problems in my water simulations. At least I couldn't
achieve energy conservation with single precision complilation and with
double precision compilation energy is conserved only when switch(shift?)
function is used for vdw interactions (rlist>rvdw>rvdw_switch) and PME for
coulombic interactions (rlist>rcoulomb). If rlist is equal to rcoulomb I need
to update NS list at every time step but with preceding parameters energy is
conserved even when nstlist=10. 

So should I use rlist>rcoulomb with PME when nstlist>1 to take into account
movement of atoms/molecules(charge groups) near the real space cut-off limit
(rcoulomb) or is there something I don't understand? Atleast from the manual
someone could get the picture that rlist should be equal to rcoulomb and there
are also different kind of opinions on mailing list?


Thanks for any comments in advance! 

Janne

------------------------------------------------------------------------------
Janne Hirvi, MSc(Physical Chemistry), Researcher
University of Joensuu, Department of Chemistry, P.O.Box 111 80101 Joensuu, FI
Tel: +358 13 2514544 & +358 50 3474223
E-mail: Janne.Hirvi at joensuu.fi & hirvi at cc.joensuu.fi
------------------------------------------------------------------------------