From: Ye Yang (knightyangpku_at_gmail.com)
Date: Wed Jun 15 2011 - 14:50:23 CDT
Hi, everyone:
I have tried to use namd to do coarse grain simulation (with rbcg.top
and rbcg.par), but I have got very different result with the coarse grain
simulation in Gromacs (orders of magnitude difference). For my
understanding, the problem seems to be coming from the different force
field. So I tried to use software on (
http://smog.ucsd.edu/cgi-bin/GenTopGro.pl#DEFAULThelp )to build Gromacs
parameter files and tried to run them in namd.
The problem comes when the simulation started the equilibrium process:
On one machine, the log file reads:
TCL: Minimizing for 5000 steps
Charmrun: error on request socket--
Socket closed before recv.
On the other machine (8 cores), there was no error, but the the
simulation seems to be stopped mysteriously, or running at very very low
speed. (it takes more than an hour to run 1 step). I am not familiar with
gromacs at all, so I do not understand what is happening here. Could anyone
help me with this?
# Input
#paraTypeCharmm on
#parameters rbcg-2007.par
gromacs on
grotopfile ni6cgro.3819.pdb.top
grocoorfile ni6cgro.3819.pdb.gro
temperature $temp
# Periodic Boundary conditions
if {1} {
cellBasisVector1 980 0. 0.
cellBasisVector2 0. 980 0.
cellBasisVector3 0. 0 980
cellOrigin 0.06 22.46 -8.67
}
wrapWater off
wrapAll off
# Force-Field Parameters
exclude scaled1-4
1-4scaling 1.0
cutoff 25.
switching on
switchdist 15.
pairlistdist 30
Thank you very much.
This archive was generated by hypermail 2.1.6 : Mon Dec 31 2012 - 23:20:27 CST