AMBER Archive (2008)


From: FyD (
Date: Fri Dec 12 2008 - 09:16:01 CST

Hi Francesco,

> Surely we can learn much from these confabulations. However, any time
> I am somewhat surprised about such concentration of thoughts about
> Gaussian. Is anything in NWChem inferior to Gaussian as far as basic
> ab initio or DFT are concerned?

No this is just that historically AMBER developers only use Gaussian.
Nothing less, nothing more.

This is why we developed R.E.D.-I (some time ago) because we were not
able to get identical charge values using GAMESS-US or Gaussian - even
if the Connolly surface & the CHELPG algo. used in MEP computation
were available in these 2 QM soft.

Moreover, even using Gaussian, charge values differ if the orientation
of the optimized structure used in MEP computation is not
identical/controlled. The problem here is that the Gaussian "Standard
orientation" of the optimized structure that is supposed to be unique
is not unique in some cases.

Using the rigid body re-reorientation algorithm implemented in R.E.D.
we have achieved a charge reproducibility of .0001 independently of
the QM program or the initial structure choice.

> If so, I could understand, however I
> would like to see were computations with NWChem are at fault. NWChem
> is provided on academic free license for nearly any machine, without
> any restriction about the number of nodes/processors or whatsoever
> else, and with code that you are allowed to compile at your will. Is
> the same true for Gaussian?

Gaussian is a proprietary QM program (provided with source code). For
small academic laboratory the price to get it is limiting (although
not that high).

GAMESS-US (like NWChem) is an academic QM program which is not GNU -
it is copyrighted + provided at no cost. This is another reason why we
targeted GAMESS-US to derive RESP and ESP charges in our R.E.D.

PC/GAMESS - Firefly have more recently showed up (copyrighted +
provided at no cost as well). R.E.D. interfaces also Firefly because
the Connolly surface and CHELPG algo (respectively used in AMBER &
GLYCAM force field development) are also available in Firefly.

Concerning NWChem, and as I told you the problem is not to generate
charge values but to be able to generate charge RESP and ESP values
which are similar (or to characterize the differences) to those
obtained with Gaussian, GAMESS-US or PC-GAMESS/Firefly. Concerning
NWChem, I am not sure the Connolly surface algo is available. We are
working on it...

> Probably Gaussian offers ready-to-use
> plots, with (I suspect largely illusory) things as simulated IR
> spectra. However, I could use with NWChem the functional M05-2X at a
> time that Gaussian was at B3LYP (which failed to perform correctly on
> my molecules).

You might decide to generate your optimized structures using NWChem &
feed R.E.D. with those QM output(s) and compute MEP using GAMESS-US,
Firefly or Gaussian...

> But I know that Francois is sensible to all that and
> promised to extend to NWChem his important contributions. Hope 2009
> will be the right year.

Yes, beginning of 2009 ;-)

regards, Francois

The AMBER Mail Reflector
To post, send mail to
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)