[gmx-users] Energy minimization problem

[David van der Spoel]

On Wed, 7 Nov 2001, Dmitry Kovalsky wrote:

>Hi All,
>
>Negative number for sqrt encountered (-48075003396096.000000)
>Terminating minimization

You are probably using conjugate gradients with single precision mdrun.
This can be problematic for algorithmic reasons. If you are just preparing
for a MD run, conjugate gradients is overkill, and you can use steep.

Conjugate gradients minimizes better, but you typically need this only
when preparing for normal mode analysis, in which case you need doube
precision anyway.

Groeten, David.
________________________________________________________________________
Dr. David van der Spoel, 	Biomedical center, Dept. of Biochemistry
Husargatan 3, Box 576,  	75123 Uppsala, Sweden
phone:	46 18 471 4205		fax: 46 18 511 755
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://zorn.bmc.uu.se/~spoel
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++