[gmx-users] stochastic thermostat

[Berk Hess]

>From: "Dmitry Nerukh" <dn232 at cam.ac.uk>
>Reply-To: Discussion list for GROMACS users <gmx-users at gromacs.org>
>To: <gmx-users at gromacs.org>
>Subject: [gmx-users] stochastic thermostat
>Date: Tue, 7 Feb 2006 11:56:17 -0000
>
>Dear gromacs developers,
>
>
>
>Can you, please, describe the steps for running a normal molecular dynamics
>with a stochastic thermostat?  I have a very simple system - just 2 waters
>in vacuum.  I've tried setting the integrator to 'sd' and 'tau_t' to 0.  
>The
>integration failed.  Is it possible to have a stochastic dynamics with the
>velocity friction term set to 0?

No.
The random and friction forces are linked by the fluctuation-dissipation 
theorem.

Berk.