Re: velocities

From: Peter Freddolino (
Date: Sat Oct 11 2008 - 12:46:05 CDT

Hi Magnus,
a single usage of reinitvels only reinitializes (not rescales) the
velocity once, at the very beginning of the simulation. It thus provides
initial conditions that are more reasonable for a system at 310K than
starting with zero velocities (as you normally would have after
minimization). Some people instead choose to gradually increase the
thermostat temperature over a few hundred ps, in which case the
reinitvels is unnecessary. So long as you have a thermostat and barostat
turned on, as soon as you start the run the thermostat/barostat will
appropriately act on the system to yield the NPT ensemble. Initializing
your velocities may shorten your equilibration time (and, if done in
multiple different runs on the same system, will yield different
trajectories each time) but should otherwise have little effect.

Magnus Heltstrom wrote:
> Hi all,
> I have become a little bit puzzled over the use of
> reinitvels $TEMP
> if I use it like this
> minimize $min
> reinitvels $temperature
> run $md
> does this make any sense to my MD simulation or is the rescaling of
> velocities to temperature(310 K) not representing any ensemble. I
> would like to be in the NPT ensemble using the thermostats for this
> but is the reinitvels destroying that?
> all the best and thanks in advance
> M

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:49:57 CST