possible bug: same Artemis project file, different results on 2 PCs
Hi Bruce, I was putting together another post for the list (which is currently awaiting moderation since apparently my attached project files were too large) when I encountered some strange behaviour in Artemis. I tried opening my project file on my second PC to check that it would open fine for other members of the list. But by simply opening the project and clicking Fit, I was getting different results for my guess parameters on the two PCs. Both PCs are Windows 7 and updated to Demeter 0.9.13. Here is a link to the project file (trying to avoid size limitations for the list this time) http://dl.dropbox.com/u/10663174/ZnO.fpj and I will attach a screenshot of Artemis demonstrating what I am talking about from my two PCs. Since I am working to determine S02 and it is very different on each PC, I am hesitant to trust any of my results right now. Thanks Jeremy --
Hi Jeremy, I got the answer on the right (amp=0.746, reduced chi-square=536) to all displayed digits on the linux and WinXP computers here in my office. I am willing to pursue this, but I'm not really certain how. How do these two computers differ? Are they both 32bit or 64bit? How was Demeter installed on each computer? Perhpas other scientific software one on that is somehow causing a conflict? I recently tried to make Demeter insensitive to the presence of another version of perl on the computer -- perhaps that's not working and one of your computers has this? I just need more to go on. B On Tuesday, October 09, 2012 04:42:07 PM Jeremy Thorbahn wrote:
I was putting together another post for the list (which is currently awaiting moderation since apparently my attached project files were too large) when I encountered some strange behaviour in Artemis.
I tried opening my project file on my second PC to check that it would open fine for other members of the list. But by simply opening the project and clicking Fit, I was getting different results for my guess parameters on the two PCs. Both PCs are Windows 7 and updated to Demeter 0.9.13.
Here is a link to the project file (trying to avoid size limitations for the list this time) http://dl.dropbox.com/u/10663174/ZnO.fpj
and I will attach a screenshot of Artemis demonstrating what I am talking about from my two PCs. Since I am working to determine S02 and it is very different on each PC, I am hesitant to trust any of my results right now.
-- Bruce Ravel ------------------------------------ bravel@bnl.gov National Institute of Standards and Technology Synchrotron Methods Group at NSLS --- Beamlines U7A, X24A, X23A2 Building 535A Upton NY, 11973 Homepage: http://xafs.org/BruceRavel Software: https://github.com/bruceravel
I tried it on a third PC and got the same result that you found so I am
confident that is the "correct" one. So it seems to be an isolated issue
with my one PC and I am fine just not using that one for now. I will look
into it more and try to isolate the cause. For what it's worth, both PCs
are 64 bit and I don't believe there are any other perl installations on
the offending machine. If I solve the problem or uncover any useful details
I will report back. Thanks for checking this out.
Jeremy
On Wed, Oct 10, 2012 at 11:32 AM, Bruce Ravel
Hi Jeremy,
I got the answer on the right (amp=0.746, reduced chi-square=536) to all displayed digits on the linux and WinXP computers here in my office.
I am willing to pursue this, but I'm not really certain how.
How do these two computers differ? Are they both 32bit or 64bit? How was Demeter installed on each computer? Perhpas other scientific software one on that is somehow causing a conflict? I recently tried to make Demeter insensitive to the presence of another version of perl on the computer -- perhaps that's not working and one of your computers has this?
I just need more to go on.
B
On Tuesday, October 09, 2012 04:42:07 PM Jeremy Thorbahn wrote:
I was putting together another post for the list (which is currently awaiting moderation since apparently my attached project files were too large) when I encountered some strange behaviour in Artemis.
I tried opening my project file on my second PC to check that it would open fine for other members of the list. But by simply opening the project and clicking Fit, I was getting different results for my guess parameters on the two PCs. Both PCs are Windows 7 and updated to Demeter 0.9.13.
Here is a link to the project file (trying to avoid size limitations for the list this time) http://dl.dropbox.com/u/10663174/ZnO.fpj
and I will attach a screenshot of Artemis demonstrating what I am talking about from my two PCs. Since I am working to determine S02 and it is very different on each PC, I am hesitant to trust any of my results right now.
--
Bruce Ravel ------------------------------------ bravel@bnl.gov
National Institute of Standards and Technology Synchrotron Methods Group at NSLS --- Beamlines U7A, X24A, X23A2 Building 535A Upton NY, 11973
Homepage: http://xafs.org/BruceRavel Software: https://github.com/bruceravel _______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
-- ------------------------- Jeremy Thorbahn thorbahn@gmail.com
participants (2)
-
Bruce Ravel -
Jeremy Thorbahn