On Sat, Apr 14, 2007, David Cerutti wrote:
>
> In response to a message I recently fielded on the listserv, I wanted
> to inquire as to why the new TIP3P model from Charlie Brooks's group,
> designed for use with Ewald electrostatics, was not the default when
> specifying TIP3PBOX (the original TIP3P is). When I do simulations with
> TIP3P and Ewald, I find that the density goes off by about 1.5%, not to
> mention other thermodynamic properties which doubtless no longer fit.
Amber developers can test model "F" from Price and Brooks: in amber10
look for frcmod.tip3pf and TIP3PFBOX. If these turn out to be good, they will
be available in the next release, and I may post them on the web site for
current users.
I don't think we can change the default for what "TIP3P" means, since that
would break backwards compatibility with about 20 years of simulations.
...regards...dac
Received on Wed Apr 18 2007 - 06:07:22 PDT
