On Dec 8, 2012, at 11:17 AM, Thomas Evangelidis <tevang3.gmail.com> wrote:
> Dear Amber community,
>
> I have prepared a system for implicit solvent simulation with an infinite
> cutoff (cut=999). Is there any way to control how often to construct the
> list with the pairs of atoms for which non-bonded interactions should be
> calculated? According to the manual the default behaviour (nbflag=1)
> updates the list whenever any atom has moved more than 1/2 skinnb since the
> last list update. Is there any way to deactivate that redundant check? For
> instance in NAMD there is a parameter named "staticAtomAssignment" which
> does exactly that thing (deactivates pairlist construction). The other way
> (nbflag=0) is not possible with implicit solvent (igb>0).
The GB code in sander and pmemd does not use a pair list. It always loops over every atom pair (and simply doesn't calculate the energy and force if they are not within the cutoff.
>
> Below is my parameter file. I would be grateful if anyone tells what
> changes I could do to speed up my simulation.
The fastest you can get is using pmemd.cuda. Or using a GB model without the neck correction (e.g., 2, 5, or 1).
HTH,
Jason
P.S. I don't think the vdw flag in the ewald namelist does anything in GB calculations.
--
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Candidate
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sat Dec 08 2012 - 15:00:02 PST
