On Fri, Mar 05, 2010, steinbrt.rci.rutgers.edu wrote:
>
> I have a question about combining ntt=3 and ibelly (yes, I know ibelly is
> an option that should normally be avoided, but it has its uses...)
>
> The combination of these two options was prohibited by Ross citing
> 'issues' with MD. My question is, is there anything practically or
> conceptually wrong with using ibelly and ntt=3 when we add an additional
> call to zero velocities after the temperature adjustment step? It seems to
> run fine for me...
Well, it is a hack, but all the belly stuff is anyway. Be sure to check that
the number of degrees of freedom is computed correctly--with a belly, you
really don't allow any rotation or translation of the whole system. If you
get the right mean temperature (correctly computed) then I guess you could
argue that things are OK.
>
>
> Diff of runmd.f:
>
> ***************
> *** 1743,1748 ****
> --- 1751,1759 ----
> #endif
> end if ! ( gammai == 0.d0 )
>
> + ! Now zero out the non-moving velocities again if a belly is active:
> + if (belly) call bellyf(nr,ix(ibellygp),v)
> +
> ! --- consider vlimit
>
> if (vlim.and.ipimd==0) then
>
....dac
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Thu Mar 11 2010 - 06:00:04 PST
