[gmx-users] Warning with L-BFGS minimization

[Jochen Hub]

Sorry, I haven't read the thread for a couple of days...too many
projects in parallel, I guesss. :)

Thanks a lot for your answers!

It was indeed the LJ cutoff which caused the error. With the following
settings, the L-BFGS minimization went allright:

vdwtype                  = Shift
rlist                    = 1.2
rvdw                     = 1.0
rvdw-switch              = 0.8

coulombtype              = PME
rcoulomb                 = 1.2


Btw, I've had quit good experience with the l-bfgs minimizer. In some
cases, where cg or steep failed, l-bfgs was succeessful.

Thanks again,
jochen




Tsjerk Wassenaar wrote:
> Hi Jochen,
> 
> It's probably the rvdw cutoff (from readir.c):
> 
> if (ir->eI == eiLBFGS && (ir->coulombtype==eelCUT || ir->vdwtype==evdwCUT)
>      && ir->rvdw != 0) {
>     warning("For efficient BFGS minimization, use switch/shift/pme
> instead of cut-off.");
>   }
> 
> Bummer!
> 
> Tsjerk
> 
> On 10/24/08, Jochen Hub <jhub at gwdg.de> wrote:
>> Hi,
>>
>>  with l-bfgs minimzation, grompp always reports
>>
>>  WARNING 1 [file em.mdp, line unknown]:
>>   For efficient BFGS minimization, use switch/shift/pme instead of cut-off.
>>
>>  even if the mdp gives
>>  ; Method for doing electrostatics
>>  coulombtype              = PME
>>  rcoulomb_switch          = 0
>>  rcoulomb                 = 1.
>>
>>  I could't find anything on that in the archive. Is that a known issue?
>>
>>  Cheers,
>>  Jochen
>>
>>
>>  --
>>  ************************************************
>>  Dr. Jochen Hub
>>  Max Planck Institute for Biophysical Chemistry
>>  Computational biomolecular dynamics group
>>  Am Fassberg 11
>>  D-37077 Goettingen, Germany
>>  Email: jhub[at]gwdg.de
>>  Tel.: +49 (0)551 201-2312
>>  ************************************************
>>  _______________________________________________
>>  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
>>
> 
> 


-- 
************************************************
Dr. Jochen Hub
Max Planck Institute for Biophysical Chemistry
Computational biomolecular dynamics group
Am Fassberg 11
D-37077 Goettingen, Germany
Email: jhub[at]gwdg.de
Tel.: +49 (0)551 201-2312
************************************************