AMBER Archive (2003)

Subject: Re: initial velocity

From: Natasja Brooijmans (nbrooij_at_itsa.ucsf.edu)
Date: Fri May 02 2003 - 09:18:26 CDT


Usually when you start running MD you use random velocities since you
don't have velocities yet, but when you
restart a MD run (e.g. you have "cut up" your MD run in several runs of
100 ps rather than running a single long run, or your run has crashes) you
use the velocities present in the restart file to obtain a single,
continuous MD run.

Natasja Brooijmans, Ph.D.
Visiting Post-Doctoral Scholar
Kuntz Laboratory
Department of Pharmaceutical Chemistry
University of California, San Francisco
San Francisco, CA 94143-2240
phone: 415-476 3986
fax: 415-502 1411
e-mail: nbrooij_at_itsa.ucsf.edu

On Fri, 2 May 2003, Youyi Peng wrote:

> Dear AMBER fans:
> I have a question about the MD initial velocity. There are two ways to
> generate the initial velocity: one to use random velocity seed, the
> other to read from the restart file. I am not sure when to use the
> velocity information from the restart file. Does the initial velocity
> affect the MD very much? Personally I think the initial velocity
> shouldn't impact the MD very much for s table system. Would you please
> give me some advice how to choose the initial velocity?
> Thank you very much.
> Youyi Peng
>
>