Hi Jim,
The default for nscm is 1000 which removes center of mass rotation and
translation every 1000 steps. In principal it should be fine to leave it
in since if your system starts with zero angular momentum it should keep
zero angular momentum in an NVE simulation. However, you can get small
rounding differences that lead to small amounts of translation or rotation
and with nscm on these get removed - thus removing energy from the system.
In perfect NVE it shouldn't matter since these degrees of freedom are
uncoupled from the rest of the system and with an infinite cut off in gas
phase it probably won't make any major difference to the calculation.
Setting it to zero just removes one extra step in the calculation where
rounding issues can occur.
A key with NVE simulations is to make sure your system has zero
translational motion at the beginning of a run though so make sure you at
least equilibrated with nscm set > 0.
All the best
Ross
On 7/1/13 9:51 PM, "James W. Snyder, Jr." <jsnyder3.stanford.edu> wrote:
>Hey,
>
>In retrospect, that makes a lot of sense, and I feel a little stupid for
>missing that. Out of curiosity though, why is nscm=0 necessary?
>
>Thanks,
>
>Jim
>
>----- Original Message -----
>From: "Ross Walker" <ross.rosswalker.co.uk>
>To: "AMBER Mailing List" <amber.ambermd.org>
>Sent: Monday, July 1, 2013 9:43:06 PM
>Subject: Re: [AMBER] NVE Energy/Temperature Drift Issue
>
>Hi Jim,
>
>You need to set cut=999 and nscm=0 for NVE runs. It is not possible to
>conserve energy in a non-pme calculation in AMBER with a cut off.
>
>All the best
>Ross
>
>
>On 7/1/13 9:40 PM, "James W. Snyder, Jr." <jsnyder3.stanford.edu> wrote:
>
>>Hey,
>>
>>Yeah, so I am not totally sure how much the energy is drifting because
>>there is a lot of noise. The temperature drifts by 60K is 120ps roughly.
>> Also, I am using an explicit solvent model/solvation sphere. From what
>>I understood, this warranted using ntb=0 and igb=0. If that's not the
>>case, please let me know. Also, I don't really want to write the results
>>every step. Those comments are residual from when I was debugging
>>something. I think dsum_tol = 0.000001 is also from something else, but
>>I didn't bother removing it because it has no impact on the calculation.
>>
>>Thanks,
>>
>>Jim
>>
>>----- Original Message -----
>>From: "Daniel Roe" <daniel.r.roe.gmail.com>
>>To: "AMBER Mailing List" <amber.ambermd.org>
>>Sent: Monday, July 1, 2013 9:17:59 PM
>>Subject: Re: [AMBER] NVE Energy/Temperature Drift Issue
>>
>>Hi,
>>
>>Could you provide some more details? Could you quantify what you mean
>>by substantial drift? The options ntb = 0 and since it's not
>>explicitly set igb = 0 imply you're running your calculation in vacuum
>>- is that correct? If so, your cutoff (16.0) seems really low,
>>especially for a 25K atom system - also note that Amber force fields
>>were not designed with vacuum calculations in mind, so you should be
>>cautious in interpreting your results.
>>
>>On Mon, Jul 1, 2013 at 9:35 PM, James W. Snyder, Jr.
>><jsnyder3.stanford.edu> wrote:
>>> ntpr = 1000, !print detials to log every step
>>> ntwx = 1000, !write coordinates to mdcrd every step
>>> ntwr = 1000, !write restart file every step
>>
>>If you really want to be writing this stuff every step (as your
>>comments imply) these should be set to 1, although that's really only
>>useful for debugging purposes.
>>
>>> dsum_tol = 0.000001, ! Increase PME accuracy
>>
>>Since you're not using the PME code (ntb = 0) this option has no effect.
>>
>>-Dan
>>
>>--
>>-------------------------
>>Daniel R. Roe, PhD
>>Department of Medicinal Chemistry
>>University of Utah
>>30 South 2000 East, Room 201
>>Salt Lake City, UT 84112-5820
>>http://home.chpc.utah.edu/~cheatham/
>>(801) 587-9652
>>(801) 585-9119 (Fax)
>>
>>_______________________________________________
>>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
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Jul 01 2013 - 22:00:07 PDT
