[Ifeffit] Question on Glitch in FEFF for Ru EXAFS measurements

[Matt Newville]

Hi Dooshaye,

On Mon, Nov 26, 2012 at 5:46 PM, Dooshaye Moonshiram <dmoonshi at gmail.com> wrote:
> Regarding my first query, basically if you upload an experimental file on
> Artemis and run a Feff calculation page from the exact same XRD coordinates,
> a different fit is sometimes obtained from trial to trial when summing all
> paths. Also if you run FEFF on the XRD coordinates without presence of the
> experimental file, I again can get a different result,which leads me to the
> conclusion that this might be a glitch. My labmates and I have obtained
> these problems several times. We have to close the whole program before
> running another calculation. I just wanted to bring it to your attention.

You get different results from a sum of paths when you run the Feff
calculation with an experimental file in Artemis  versus when running
the Feff calculation in Artemis without an experimental file?   That
seems really strange to me, but maybe I'm not understanding.  I
thought you had said earlier that you get different fits, but if you
don't have experimental data, I don't see how you could do a fit.
That's just to emphasize that I'm confused by what you are comparing.

What is the nature of the difference you see?   Do you see differences
in numerical values of fitted parameters, in chi(k), in chi(R)?  I
hate to guess, but I wonder if what you're seeing could be due to
differences in k- and R-ranges.  If you're comparing chi(R), you have
to be sure all the FFT parameters are identical.  Of course, you also
have to be sure that all the values of all the Path Parameters are
identical too.

Providing a short, complete example would be really helpful.
Otherwise, it's really hard for us to guess what might be going on.

Thanks,

--Matt