[Ifeffit] Problem with Demeter 0.9.18 pluck tool

Denis Testemale denis.testemale at grenoble.cnrs.fr
Fri Sep 27 07:24:46 CDT 2013


Dear members of the list,
I finally got the opportunity to start using Dathena. I don't know if 
I've found a bug or if I'm doing something stupid (chances are not too 
low, actually...).

- I am using Ubuntu 13.04
- Demeter 0.9.18 (I did the upgrade to 0.9.18 but ran into some error 
messages as reported on September 10th on this list, could they be 
related to the behaviour I am reporting here?).
- I cannot say if it worked before the upgrade to 0.9.18 since I never 
took the time to use Demeter properly until now.

I'm trying to pick/pluck a value from the graph.
- When I first launch Dathena, and I press a pluck button for the first 
time I can see the expected message at the bottom of the GUI ("Double 
click on a point to pluck its value ..."): as hard as I tried (I tried 
many different combinations) I could never get the thing to work 
(parameter value in the 'box' never records the value where I double 
click). After 10s I see the 'Pluck timed out' message.
- Then, when I click again on the pluck button, a value is recorded 
instantly in the 'box' without any double-clicking from me, and this 
value is crazy. An example: for a low k parameter of the FT I get 
-100000. The example in the attached screenshot shows a value for the 
normalization range low limit at -107112.52eV (note that the E0 value is 
'integrated' in this number: the iron edge at 7112.52, see screenshot. I 
tried another set of data, silver K edge and the same thing occured: a 
weird number, with the E0 value inside, got recorded in the 'box).
- Same problem with Dartemis.

Tell me what I can do to help... or what I'm doing stupid.
Thanks.
denis
-------------- next part --------------
A non-text attachment was scrubbed...
Name: athena_pluck_problem.png
Type: image/png
Size: 163596 bytes
Desc: not available
URL: <http://millenia.cars.aps.anl.gov/pipermail/ifeffit/attachments/20130927/0838a0bc/attachment.png>


More information about the Ifeffit mailing list