Re: Minimization Question

From: Jim Phillips (jim_at_ks.uiuc.edu)
Date: Thu Sep 18 2003 - 16:30:49 CDT

You'll wait a long time for minimization to converge for a large system,
and even then you're only guaranteed a local minimum. Minimization is
normally used before starting MD to eliminate bad contacts and initial
heating. I'd suggest plotting energy vs step and looking for diminishing
returns, or figuring out how many steps you're willing to wait. 1000
steps goes a long way and 10,000 should be sufficient for most systems.

-Jim

On Thu, 18 Sep 2003, Saladino, Alex wrote:

> Hello fellow NAMDers,
> I have a question about minimizations.
> How do you know when your system has fully minimized?
> Do people look at change in energies, absolute or relative? If not, what
> else should I be looking at?
>
>
> Thank you,
> Alex
>
>
>
> *************************************
> Alexander C. Saladino, PhD
> Department of Anesthesiology
> University of Pittsburgh
> *************************************
>
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:37:02 CST