Re: [cclib-devel] unit consistency issue

[Noel O'B. <bao...@gm...>]

On 06/11/2007, Karol Langner <kar...@kn...> wrote:
> On Tuesday 06 November 2007 04:28, Noel O'Boyle wrote:
> > On 06/11/2007, Karol Langner <kar...@kn...> wrote:
> > > I want to raise a consistency issue,
> > >
> > >  Most energy-related attributes are parsed in eV (scfenergies,
> > > mpenergies, etc.). The numbers in scftargets and scfvalues, however, are
> > > parsed without conversion. Is this intentional or unnoticed?
> >
> > For me, the only requirement is that the same units are used for
> > scftargets and scfvalues - in fact, the units are different for
> > different elements of scftargets. As far as I know, this attribute is
> > only useful for monitoring progress. Different programs have different
> > ways of measuring convergence; typically delta energy and delta
> > matrix. Although we can change the units of delta energy if you wish,
> > I don't know who is going to be accessing them apart from GaussSum.
>
> I was just thinking that if someone took the energy change criteria of
> scftargets to be in eV they could be very surprised. I have no reason to add
> the conversion.

OK - I'll look into it. I guess you're right.

> Karol
>
> --
> written by Karol Langner
> Tue Nov  6 10:21:47 EST 2007
>