[Jmol-users] Re: Jmol bug report - Gaussian 98 + NoSymm problem

[Miguel <mi...@jm...>]

Jamie wrote:
> NoSymm
> 	Here is the Gaussian developers definition
> 	http://www.gaussian.com/g_ur/k_symmetry.htm
>
> 	My version, being more of a chemist then a programmer, is this:
> 	In general, gaussian looks for symmetry within the molecule to
> 	simplify the calculations.
[snip]
> Using this command instructs Gaussian to
> optimize the structure in the Z-matrix format, and thus, no "standard
> orientation" output is defined, and the heading is missing.  Reverting
> back to the Water molecule, the O-H bond lengths would be treated
> separately, and optimized individually.
>
> 	I hope this explanation helps, although I am unsure as to how
> accurate it is.

Good enough for me.

> As for the coordinates::
>
> In the files sent to you, geometry optimizations where not necessary, as
> optimized geometries were used as input, simply to test the NoSymm
> operator.  Instead of standard orientation, it list the coordinates under
> "input orientation".  This may be the title that you are looking for.

OK, that is the data that I need.


> Aside from all this, I should point out that at present, JMol searches for
> the coordinates systems, and loads the first one that it finds.  This is
> fine for single point calculations, but in geometry optimizations, it can
> be misleading.  Opening an entire Gaussian output file displays the
> 'input' geometry, and not the 'output' geometry, which would be the last
> set of coordinates listed.

I believe that the GaussianReader should be selecting the *last* geometry
in the file, not the first.

If it is not picking the last set off coordinates, then it is a bug.



Miguel