AMBER Archive (2003)

Subject: Re: problem about virtual box?

From: Youyi Peng (pengyo_at_UMDNJ.EDU)
Date: Thu May 01 2003 - 12:42:47 CDT

Dr. Cheatham:
   Thanks. But my system is non-periodic and in vaccuo. Does `nscm` still
affect the system so much?
  Youyi Peng

Thomas Cheatham wrote:

> As Prof. Simmerling pointed out, you want to see if the molecule is
> moving in the box. Your e-mails suggest, based on the first and
> last frames that this is indeed the case. This is in fact highly
> likely since:
>
> > nscm = 0,
>
> This implies you are not removing the center of mass motion (translation
> in a periodic system). Due to initial velocity assignment there is a high
> probability that there will be some initial motion (i.e. the sum of the
> velocity vectors likely will not sum to zero). Further, if energy is not
> rigorously conserved and you are coupling to a temperature bath, energy
> drains will lead to a slow growth in COM translational (and rotational)
> motion. Eventually all the temperature will be in translational motion.
> You can check this by setting NSCM = 500 or some such and re-starting. If
> the temperature drop is significant on the first step, you saw the flying
> block of ice problem... For more info, see papers in JCC by Harvey
> et al. (1998) and also by Chiu et al. (2000) and/or search for block of
> ice on the archive.
>
> When in a periodic system it is wise to set NSCM to be some finite value
> (like 500 in a periodic gas phase system) to minimize this motional
> artifact.
>
> Good luck,
>
> \ Thomas E. Cheatham, III (Assistant Professor) College of Pharmacy
> | Departments of Medicinal Chemistry and of University of Utah
> | Pharmaceutics and Pharmaceutical Chemistry 30 South 2000 East, Room 201
> | & Center for High Performance Computing Salt Lake City, Utah 84112
> |
> | e-mail: tec3_at_utah.edu phone: (801) 587-9652 FAX: (801) 585-9119
> \ http://www.chpc.utah.edu/~cheatham Offices: BPRP295A / INSCC 418