Hi Bill
Thanks for the reply. With 'identical' I meant the same general parameters
(simulation length, number of replicas, anneal protocol etc). Varying
random seed is from my point of view a necessity, so i consider this to be
obvious. Varying other parameters might also be interesting however.
The question was more about the possibility to visualize the force between
images since this is in many implementations (ASE, orca, ...) the way to
'determine' convergence. (I also know their method differs from Amber but
was just curious if we could get the forces connecting each image.)
Kind regards
Charles-Alexandre Mattelaer
Op ma 11 nov. 2019 12:41 schreef Bill Ross <ross.cgl.ucsf.edu>:
> I think the general rule in our species' evolutionary phase is you huff
> and puff and blow all the resources you can, but as wisely as possible.
> E.g. varying the random seed would be better than 'running identical
> simulations different times', and the direction and scope of blowing
> generally would depend on the kind of question being asked.
>
> Bill
>
> On 11/11/19 12:33 AM, Charles-Alexandre Mattelaer wrote:
> > Hi all
> >
> > Just wanted to bump this message since I'm also curious. The manual says
> "
> > Statistically relevant number of simulations
> > must be performed to ensure reproducibility (and convergence) of the
> > results." so it seems that up to this point the only way of checking
> > convergence is running identical simulations different times and checking
> > wether or not they end up with same results?
> >
> > Kind regards
> >
> > Charles-Alexandre Mattelaer
> >
> > Op za 26 okt. 2019 00:19 schreef Yang JIANG <jy1990217.126.com>:
> >
> >> Dear Amber Users,
> >>
> >>
> >> I am now running NEB calculations with QM/MM.
> >> I am wondering how to check the convergence of NEB calculations. As I
> >> know, people usually check the convergence of NEB by checking the
> >> convergence of the maximal force on NEB images. However, I don't know
> how
> >> to print out the max force on the NEB atoms. Does anyone know how to do
> >> that? Otherwise, is there an alternative way to check the convergency?
> >>
> >> Thanks a lot!
> >>
> >> Yang
> >>
> >> --
> >>
> >> Yang JIANG, Post-doc in O'Brien Lab
> >> Department of Chemistry, Pennsylvania State University
> >> 104 Chemistry building, Room #404, University Park, PA, 16802, USA
> >> Phone: +1 (814) 880-9918
> >> Email: jy1990217.126.com; yuj179.psu.edu
> >> _______________________________________________
> >> AMBER mailing list
> >> AMBER.ambermd.org
> >> http://lists.ambermd.org/mailman/listinfo/amber
> >>
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Nov 11 2019 - 04:30:02 PST
