AMBER Archive (2002)

Subject: turning off ewald (fwd)

From: amber_at_heimdal.compchem.ucsf.edu
Date: Mon Aug 26 2002 - 16:51:21 CDT

---------- Forwarded message ----------
Date: Fri, 23 Aug 2002 13:23:43 -0700
From: Matthew Lee <mrlee_at_cgl.ucsf.edu>
To: amber_at_heimdal.compchem.ucsf.edu
Subject: turning off ewald

I'm trying to run a non-periodic dynamics simulation with ntb=0,
cut=999, and a belly around my ligand. But, sander dies. The mdout
file says,

--------------------------------
Warning: Although EWALD will work with belly
(for equilibration), it is not strictly correct!
--------------------------------

and then at the bottom, it reportedly dies of an EWALD BOMB, because
my nonbond list exceeds the MAXPR.

The manual says PME "is always "on", unles ntb=0, or use_pme is set to
zero." So, after this, I tried explicitly setting use_pme=0 in the
&ewald namelist as well, but sander still behaves the same way. Any
ideas on what might be keeping me from turning PME off? Or is PME
actually off with the error message just trying to reflect the
pairlist exceeding my limit?

I don't understand how PME could be run on a non-periodic system and
the mdout file also says:

--------------------------------
| NONPERIODIC ntb=0 and igb=0: Setting up nonperiodic simulation
| *** cutoff > system size, list only builds once"
--------------------------------

--Matt

*******************************************************************
Matthew Randolph Lee, Ph.D.

Computational Scientist phone: (858)410-6534
Lion Bioscience Inc. fax: (858)410-6665
9880 Campus Point Drive e-mail: matthew.lee_at_lionbioscience.com
San Diego, CA 92121
*******************************************************************