[gmx-users] Stepsize too small ( what to do?)

[Kai Zhuang]

try use cg (conjugated gradient), might help you to minimize lower.

also, beside changing the constriant to none, check in your top file
which version of water you have.  you want to include "flexible water"
by using:

#ifdef FLEX_SPC
#include "flexspc.itp"
#else
#include "spc.itp"
#endif

these two changes generally helps me to avoid the converge to machine
precision thing, at least not until a few tens of thousands
calculations anyways.

On 8/3/05, li dawei <dli at clarku.edu> wrote:
> I guess there are sth wrong in your initial simu cell.
> 
> Energy  minimzation cannot guarantee a "good" initial structure at this
> case.
> 
> ----- Original Message -----
> From: "Anil Kumar" <anil at chem.iitb.ac.in>
> To: <gmx-users at gromacs.org>
> Sent: Tuesday, August 02, 2005 12:46 PM
> Subject: [gmx-users] Stepsize too small ( what to do?)
> 
> 
> Hi to all,
> 
> I am doing normal MD simulation:- vac followed by em: followed  by pr: and
> then finally full MD.
> 
> But i am getting a problem like this during energy minimization by
> steepest descents,
> 
> 
> "Stepsize too small, or no change in energy.
> Converged to machine precision,
> but not to the requested precision Fmax < 100
> 
> Double precision normally gives you higher accuracy.
> You might need to increase your constraint accuracy, or turn
> off constraints alltogether (set constraints = none in mdp file)
> 
> writing lowest energy coordinates."
> 
> Steepest Descents converged to machine precision in 34 steps,
> but did not reach the requested Fmax < 100.
> Potential Energy  = -8.0718016e+04
> Maximum force     =  1.9289487e+03 on atom 4826
> 
> 
> I have changed the em step in em.mdp file but still it is not working.
> Secondly i have started pr with the em conformation, but it crashes.
> 
> any suggestion !!
> thanks in advance
> anil
> _______________________________________________
> gmx-users mailing list
> gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
> 
> _______________________________________________
> gmx-users mailing list
> gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
> 


-- 
Kai Zhuang
BioMedical Engineering @ University of Toronto
kai dot zhuang at gmail dot com
6478314789