[gmx-users] energy minimization using double precision gromacs failed

[SeungPyo Hong]

Hi,

How about trying md with very small step size such as
  dt = 0.0001
?
Once I had similar problem like you and it is solved by performing
additional energy minimization step with very small step size.
I can't guarentee this method, but I think it might be OK.

On 8/13/07, Q733 <sansanqin00 at mails.tsinghua.edu.cn> wrote:
>
> Hello, gmx-users,I tried to energy minimize a lipidbilayer consisting of
> 32 lipids(16 up and 16 down) with two water layers(428 up and 428 down).
> my em.mdp is like this:
>
> title                    =
> cpp                      = /lib/cpp
> include                  =
> define                  = -DFLEXIBLE
> constraints              = none ;h-bonds
> lincs_iter               = 4
>
> ; RUN CONTROL PARAMETERS =
> integrator               = l-bfgs
>
> tinit                    = 0
> dt                       = 0.2  ; ps
> nsteps                   = 6000
>
> nstxtcout                               = 100     ; pos to xtc file
>
> nstlist                                 = 5
> ns_type                                 = grid
> rlist                                           = 0.7
> coulombtype                             = PME
> rcoulomb                                        = 0.7
> vdw-type                                        = Cut-off
> rvdw                                            = 0.7
> fourierspacing                  = 0.095 ; default to 0.12
> fourier_nx                              = 0
> fourier_ny                              = 0
> fourier_nz                              = 0
> pme_order                               = 4
> ewald_rtol                              = 1e-5
> optimize_fft                    = yes
>
> ; ENERGY MINIMIZATION OPTIONS =
> emtol                                           = 0.00001       ; default
> to 100
> emstep                                  = 0.5           ; default to 0.01
> nstcgsteep                              = 1000
>
> The em is done using double-precision gromacs program,and the out put is
> like this:
>
> Stepsize too small, or no change in energy.
> Converged to machine precision,
> but not to the requested precision Fmax < 1e-05
>
> writing lowest energy coordinates.
>
> Low-Memory BFGS Minimizer converged to machine precision in 7 steps,
> but did not reach the requested Fmax < 1e-05.
> Potential Energy  =  1.25811048952284e+15
> Maximum force     =  5.23956606456297e+17 on atom 483
> Norm of force     =  6.77707038974525e+15
>
> How can I prevent it from converging to machine precision? Is there any
> method to change the standard of machine precision?
> Any suggestion will be appreciated , thanks a lot.
>
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before posting!
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>



-- 
--------------------------------------------------
'God used beautiful mathematics in creating the world.'
-Paul Dirac
'But he created too many object.'
-Seungpyo Hong

Seungpyo Hong
Master student at PBIL(Protein BioInformatics Lab.) in KAIST, Daejeon Korea
Tel. (82)-18-372-2468
sphong_ at kaist.ac.kr
sp1020 at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20070813/e8d68d70/attachment.html>