From: Vermaas, Joshua (Joshua.Vermaas_at_nrel.gov)
Date: Wed Jul 18 2018 - 07:45:36 CDT
You need to build uglier systems. :p 1000-2000 steps is pretty normal for uglier crystal structures. But basically the typical source of these errors is described in: http://www.ks.uiuc.edu/Research/namd/wiki/index.cgi?NamdTroubleshooting
On 2018-07-17 15:38:01-06:00 owner-namd-l_at_ks.uiuc.edu wrote:
Oh, I missed the part where this is a minimization - that's not a good idea. NAMD minimization does not wrap along PBCs, so large minimizations will almost assuredly fail. I would recommend 200 steps at the most (or at least I've never built a system so ugly that this doesn't work).
My previous advice would probably only help for dynamics.
On Tue, Jul 17, 2018 at 5:18 PM, Shriya Bansal <bansals10_at_gmail.com<mailto:bansals10_at_gmail.com>> wrote:
Thank you so much for your response. I have increased the margin and pairlistdist, but these changes to the configuration file have unfortunately resulted in the same errors.
I have attached two images of the TIP4P-2005 water box from different angles; is there a reason that certain molecules are stretched after minimization?
On Jul 17, 2018, at 3:17 PM, Brian Radak <brian.radak_at_gmail.com<mailto:brian.radak_at_gmail.com>> wrote:
I frequently encounter similar problems with systems that size - most of NAMD's performance heuristics are simply not designed for something small.
You can try increasing the margin (maybe 1 or 2?) or pairlistdist (16? 18?). You will probably also have less trouble without temperature control.
In general, I've never seen any stability benefit from heating a system slowly versus all at once. There might be good reasons to heat a biopolymer from a crystal structure, but for water this is almost certainly a waste of time.
On Tue, Jul 17, 2018 at 3:45 PM, Shriya Bansal <bansals_at_uchicago.edu<mailto:bansals_at_uchicago.edu>> wrote:
This archive was generated by hypermail 2.1.6
: Tue Sep 17 2019 - 23:20:04 CDT
I have been attempting to minimize and equilibrate a 38A TIP4P-2005 water box model built through CHARMM using the July 2017 TIP4P-2005 topology from http://mackerell.umaryland.edu/charmm_ff.shtml
This archive was generated by hypermail 2.1.6 : Tue Sep 17 2019 - 23:20:04 CDT