Hi Dave,
Thanks for the idea- I haven't actually tried any of this on the GPU code
so far, since it's all on CPU sander right now. I couldn't find any
equivalent of it in the pmemd portion of the manual, but I didn't even
think of rescaling charges in the topology proper to mimic that effect so
I'll keep it in mind.
Best,
Kenneth
On Wed, Dec 30, 2020 at 12:17 PM David Cerutti <dscerutti.gmail.com> wrote:
> I don't know all that is in the sander code, or perhaps even the pmemd
> Fortran layer, but I can say that there is no support for fiddling with the
> dielectric in the periodic GPU code, nor that I know of in the GB GPU
> code. That said, if all you want is a change in the overall dielectric,
> you can mimic this effect by scaling the charges in your simulation
> topology and running igb=6 or periodic simulations with a fixed volume
> (NVE, NVT, ntp=0). If you want a distance-dependent dielectric that's not
> something any of the fast codes will support--as DAC said this is not
> something we support because it is just not a preferred way of doing things
> any more.
>
> Dave
>
>
> On Wed, Dec 30, 2020 at 11:44 AM Kenneth Huang <
> kennethneltharion.gmail.com>
> wrote:
>
> > Hi David,
> >
> > Thanks for the response!
> >
> > An additional clarification- dielc=4 and eedmeth=5 would give a distance
> > dielectric of 4r, but if I wanted a constant dielectric with some solvent
> > box, would it instead be extdiel=4 with igb=6 instead? Or is there no
> > modern equivalent to the constant dielectric (idiel=1) with a solvent
> box?
> >
> > Best,
> >
> > Kenneth
> >
> > On Sat, Dec 26, 2020 at 8:59 AM David A Case <david.case.rutgers.edu>
> > wrote:
> >
> > > On Mon, Dec 21, 2020, Kenneth Huang wrote:
> > > >
> > > >The Amber5 manual mentions two dielectric functions (idiel=0 for
> > distance,
> > > >idiel=1 for constant) with the dielc controlling the constant.
> > > >
> > > >But I'm not clear on is how this would translate to amber16 sander? I
> > > think
> > > >if I want a gas phase with a dielectric of 4 I'd use
> > > >
> > > >#current min
> > > >> &cntrl
> > > >> ...
> > > >> igb=0, extdiel=4,
> > > >> ...
> > > >> &end
> > >
> > > Above shouldn't work: extdiel is for GB calculations, but you have
> turned
> > > GB
> > > off. I think the combination you want is dielc=4 and eedmeth=5.
> > >
> > > Usual caveats: no developer has used this combination in a long time,
> and
> > > we've removed it from the documentation. (It was already considered to
> > > be effectively obsolete when the eedmeth variable was first introduced
> > > 25 years ago.) So test carefully. And for sure, don't expect to get
> > > physically reasonable results from this model.
> > >
> > > ....dac
> > >
> > >
> > > _______________________________________________
> > > AMBER mailing list
> > > AMBER.ambermd.org
> > > http://lists.ambermd.org/mailman/listinfo/amber
> > >
> >
> >
> > --
> > Ask yourselves, all of you, what power would hell have if those
> imprisoned
> > here could not dream of heaven?
> > _______________________________________________
> > 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
>
--
Ask yourselves, all of you, what power would hell have if those imprisoned
here could not dream of heaven?
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Dec 30 2020 - 12:00:03 PST
