Re: langevin dynamics

From: Giovanni Bellesia (
Date: Sat Apr 12 2008 - 15:20:20 CDT

as far as I see Langevin is the only thermostat implemented in NAMD
which, in principle, can be associated with a canonical ensemble. Other
examples of canonical thermostats (not implemented in NAMD) are
Andersen, DPD and Nose'-Hoover thermostats.
Berendsen thermostat is NOT a "canonical" thermostat. Strictly speaking
it's an isokinetic thermostat.
However, in several cases the deviation from the canonical distribution
is quite small and one can neglect it.
> Hii all..
> Langevin dynamics is supposed to be much faster than MD with other
> thermostats due to the implicit nature of the solvent and the larger
> time steps that can be used. But as I got know from other posts on
> this list..langevin dynamics in namd essentially utilizes explicit
> force field for the solvent too..and utilizes the friction term merely
> as a method to maintain a constant average temperature. So is Langevin
> dynamics at all beneficial w.r.t. time of simulation...since the
> calculation of force which is supposed to take the majority of
> computational time is being calculated both for normal NVT with
> Berendsen thermostat as well as in Langevin dynamics.
> dibyadeep

Giovanni Bellesia
Department of Chemistry and Biochemistry
University of California, Santa Barbara
Santa Barbara, California 93106-9510, USA.
Phone: 805 893 2767

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:49:22 CST