[gmx-users] emtol in a water box

[David van der Spoel]

On Thu, 2002-10-31 at 20:29, Tanos wrote:
>     Hi users,
>     When peforming a energy minimization of a dimeric protein (formed by 02
> monomeric units of 442 residues) in a water's box, I noted that GROMACS
> never converges to emtol values under about 418 Kj/mol.nm. So, I decided to
> run the "speptide in a water's box" example of the tutorial to check this. I
> noticed that the em.mdp file of this example uses emtol of 2000 than, I
> tryed to test values under 2000. GROMACS converged until the value of emtol
> of about 205. When I tryed 100, the calculation stopped and output the
> message: "stepsize too small converged to machine precision". So, once the
> same problem happens with the tutorial example, I would like to ask if a
> emtol of 2000 or 1000 could be fine to a minimization of a protein in a
> water's box.
>     I am trying to perform the minimization until about 41,8 because in a
> energy minimization in vacum this is the ideal value, but I am not sure if
> it would be the same to a water's box. I am also looking for examples in
> literature but it would be very nice if someone answer this question

could be accuracy problems (i.e. that the local minimum is very small,
and that the minimizer cannot find it with the stepsize you are using.)
Could also be a bug in the minimizer routine (please debug it and send
me the patch...). For equilibration for MD this is usually good enough. 


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://xray.bmc.uu.se/~spoel
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++