Skip to content

Possible l-bfgs improvements - Redmine #2641

Minimization of even small molecules to convergence is problematic in GROMACS. I am finding that energy minimization not always succeeds, even for very simple molecules such as 2-octanol or pentoxypentane. This leads to problems when you want to do a normal mode analysis based on that structure.

It would be useful to have a minimizer that does go all the way to zero force (steepest descents doesn’t do it either).

To reproduce the problem, unpack the archive, then run

gmx_d grompp
gmx_d mdrun

It will finish with:

“Energy minimization has stopped, but the forces have not converged to the
requested precision Fmax < 0.001”

(from redmine: issue id 2641, created on 2018-09-11 by dspoel, closed on 2019-04-05)

Fixes bug in l-bfgs.

l-bfgs should now work excatly as it was in 4.6.5,
up-to some numerical precision in forces and energies.

Fixes #2641

Change-Id: I11a85a4241cc933fef94e2095dd8c3cbbb28b01b