[gmx-users] TIP5P and Free Energy

[Berk Hess]

>From: "Ronen Zangi" <rzangi at hotmail.com>
>Reply-To: Discussion list for GROMACS users <gmx-users at gromacs.org>
>To: gmx-users at gromacs.org
>Subject: [gmx-users] TIP5P and Free Energy
>Date: Mon, 24 Jan 2005 17:27:49 -0500
>
>
>Dear All,
>
>I have a problem that somebody may had encountered and solved. Running 
>TIP5P with the Free energy option in the mdp file seems to go wrong 
>(density is getting lower and lower). This occurs with all versions that I 
>checked 3.2.1,  3.1.4 and 3.0.5.
>
>Running TIP5P without Free Energy is good. Running SPC (for exmaple) with 
>Free Energy is also good. I would guess that it is becuase of the 
>combination of the
>[ dummies3 ] entery in the top file of TIP5P and the Free Energy option.

This is surely not the problem.

My guess would be the problem is related to a bug with the MNO solvent
optimization.
Try running after doing:
setenv GMX_NO_SOLV_OPT

I have recently found out that the MNO solvent optimization is not very 
fast.
It was only faster in some cases because I had forgot one detail in the 
optimization
of the neighborsearch. In the CVS code this optimization has been added and
the MNO loops have been removed. The CVS code (a new release will be out 
soon)
runs tip5p much faster.

Berk.

_________________________________________________________________
MSN Search, for accurate results! http://search.msn.nl