Re: [Ifeffit] Bugs in new versions of Athena and Artemis
Bruce, thanks for the comprehensive response to my questions.
You can edit the titles on the data page to put your own information in the header written to the data chi(x) files. But I see your point and will include that in the next release.
Concerning Athena I see how to work around the problems. Putting parameter-information in the project title is indeed a help. But usually the parameters will change often so an automated writing to the exported files would be better. I would therefore appreciate this function in a future release. I will now concentrate on Artemis. I did again a new full installation of Ifeffit and set up the example files again, carefully looking at the error messages. (BTW: Initially the FEFF-executable is set to feff6, it is necessary to switch to feff6l by hand).
1. Did you do anything to the folders or files under C:\Program Files\Ifeffit\horae\stash\artemis.project.0\ while Artemis was still running?
I did never edit any file in the \horae\ .. folders but before deleting the old installation I had a look there and found around 20 folders called "artemis.project.0", "artemis.project.1" etc. After the new installation the "artemis.project.0" was removed each time not cluttering up the folder. Could there be a relation to the errors?
The basic flowchart for an Artemis 0.7 project is this: (1) import data, (2) import or create atoms data, (3) run atoms, (4) check the feff.inp data and run feff, (5) set up your fitting model, (6) hit the big green button.
I set up the project file in the order you described: and indeed everything works normal ! I therefore tried to reverse the order: first load in a feff-calculation, play with it and afterward load a data-file. The project file went corrupt again (error messages in ARTEMIS.TRAP). I tried it a few times in the "wrong order", some times it went OK. After a while I figured out what the problem might be: When I load the data file (for the first time) via the dialog Data | Change Data File problems will occur (which I usually did despite the fact the the dialog was built to CHANGE an existing data file - I was just confused about the variety of ways to load in a data file). When I first load in the data file via File | Open data file or project it will be OK. Does that make sense? I am sorry but the errors are not always reproducible and I did my best to figure out a possible scheme. I included a corrupted and a good file but they seemed pretty equal. When I load in the corrupted file a second time everything will be okay, probably only the copy in the stash-folder gets corrupt? I summarize: when I follow the correct order I could not find any errors so far. All the different error messages I had might be related to a wrong loading procedure.
(1) Completely delete the feff calculation from the project using the discard option in the Feff menu and then start over again
Do you think it is risky to delete the FEFF-calculation in the Zip-file manually? HTH, Gerrit
On Thursday 11 March 2004 02:48 pm, Schmithals wrote:
Bruce,
thanks for the comprehensive response to my questions.
You can edit the titles on the data page to put your own information in the header written to the data chi(x) files. But I see your point and will include that in the next release.
Concerning Athena I see how to work around the problems. Putting parameter-information in the project title is indeed a help. But usually the parameters will change often so an automated writing to the exported files would be better. I would therefore appreciate this function in a future release.
D'oh! When I wrote that about the titles, I has Artemis on the brain. I did not grok that you were suggesting that feature in Athena. I agree that in the context of Athena, my suggestion is not a very practical one.
I will now concentrate on Artemis. I did again a new full installation of Ifeffit and set up the example files again, carefully looking at the error messages. (BTW: Initially the FEFF-executable is set to feff6, it is necessary to switch to feff6l by hand).
1. Did you do anything to the folders or files under C:\Program Files\Ifeffit\horae\stash\artemis.project.0\ while Artemis was still running?
I did never edit any file in the \horae\ .. folders but before deleting the old installation I had a look there and found around 20 folders called "artemis.project.0", "artemis.project.1" etc. After the new installation the "artemis.project.0" was removed each time not cluttering up the folder. Could there be a relation to the errors?
Could be. I'd send them all to the recycle bin in any case. You do point out an interesting problem, even if not related to the topics at hand. I allow for multiple artemis.project.N folders to allow for multiple instances of Artemis. However, when something bad happens, Artemis tends to leave behind an abandoned project folder. This is at least ugly and messy and may be a problem. That gives me something to think about....
The basic flowchart for an Artemis 0.7 project is this: (1) import data, (2) import or create atoms data, (3) run atoms, (4) check the feff.inp data and run feff, (5) set up your fitting model, (6) hit the big green button.
I set up the project file in the order you described: and indeed everything works normal !
I therefore tried to reverse the order: first load in a feff-calculation, play with it and afterward load a data-file. The project file went corrupt again (error messages in ARTEMIS.TRAP). I tried it a few times in the "wrong order", some times it went OK. After a while I figured out what the problem might be: When I load the data file (for the first time) via the dialog Data | Change Data File problems will occur (which I usually did despite the fact the the dialog was built to CHANGE an existing data file - I was just confused about the variety of ways to load in a data file). When I first load in the data file via File | Open data file or project it will be OK.
That's a good point. Using the "Data -> Change data file" function should not lead to trouble.
Does that make sense? I am sorry but the errors are not always reproducible and I did my best to figure out a possible scheme. I included a corrupted and a good file but they seemed pretty equal. When I load in the corrupted file a second time everything will be okay, probably only the copy in the stash-folder gets corrupt?
I summarize: when I follow the correct order I could not find any errors so far. All the different error messages I had might be related to a wrong loading procedure.
That's useful information for both of us. It gives you a way to at least get Artemis to do something useful in the current release. And it gives me some hints for things to look at for the next release. Thanks!
(1) Completely delete the feff calculation from the project using the discard option in the Feff menu and then start over again
Do you think it is risky to delete the FEFF-calculation in the Zip-file manually?
Yes, I do, for now. There will still be information about the calculation that you delete that was in the description file. That will almost certainly confuse Artemis the next time she starts up. I'll try to put some logic in the next version that checks for that situation and does the right thing. These have all been the sorts of problems that are hardest for me to find. Since I know how I intend the program to be used, I tend not to see these problems. It's great that you look at it in a different way and it's great that you tell me about what actually happens ;-) B -- Bruce Ravel ----------------------------------- ravel@phys.washington.edu Code 6134, Building 3, Room 405 Naval Research Laboratory phone: (1) 202 767 2268 Washington DC 20375, USA fax: (1) 202 767 4642 NRL Synchrotron Radiation Consortium (NRL-SRC) Beamlines X11a, X11b, X23b National Synchrotron Light Source Brookhaven National Laboratory, Upton, NY 11973 My homepage: http://feff.phys.washington.edu/~ravel EXAFS software: http://feff.phys.washington.edu/~ravel/software/exafs/
On Thursday 11 March 2004 02:48 pm, Schmithals wrote:
(BTW: Initially the FEFF-executable is set to feff6, it is necessary to switch to feff6l by hand).
Has anyone else had this problem? I thought I had writtent things such that the name of the feff executable got set to "feff6" on unix and "feff6l" on Windows, so I am surprised that you had this problem. B -- Bruce Ravel ----------------------------------- ravel@phys.washington.edu Code 6134, Building 3, Room 405 Naval Research Laboratory phone: (1) 202 767 2268 Washington DC 20375, USA fax: (1) 202 767 4642 NRL Synchrotron Radiation Consortium (NRL-SRC) Beamlines X11a, X11b, X23b National Synchrotron Light Source Brookhaven National Laboratory, Upton, NY 11973 My homepage: http://feff.phys.washington.edu/~ravel EXAFS software: http://feff.phys.washington.edu/~ravel/software/exafs/
Hi Bruce, At 03:48 PM 3/11/2004 -0500, you wrote:
On Thursday 11 March 2004 02:48 pm, Schmithals wrote:
(BTW: Initially the FEFF-executable is set to feff6, it is necessary to switch to feff6l by hand).
Has anyone else had this problem? I thought I had writtent things such that the name of the feff executable got set to "feff6" on unix and "feff6l" on Windows, so I am surprised that you had this problem.
My student had this problem on a Windows XP machine. We changed it by hand as well. --Scott Calvin Sarah Lawrence College
On Thursday 11 March 2004 04:13 pm, Scott Calvin wrote:
Hi Bruce,
At 03:48 PM 3/11/2004 -0500, you wrote:
On Thursday 11 March 2004 02:48 pm, Schmithals wrote:
(BTW: Initially the FEFF-executable is set to feff6, it is necessary to switch to feff6l by hand).
Has anyone else had this problem? I thought I had writtent things such that the name of the feff executable got set to "feff6" on unix and "feff6l" on Windows, so I am surprised that you had this problem.
My student had this problem on a Windows XP machine. We changed it by hand as well.
OK. I looked into this issue and realized that I made a mistake months ago in the configuration system that effects windows users but not unix users. I have fixed it, after a fashion. I don't see a way to fix the problem retroactively. That is, if you have already installed artemis on your windows computer, you will have to go to the preferences dialog and reset the name of the Feff executable. After the next release, future installations of artemis will not suffer from this problem. Just to clarify -- if you press the "Run feff" button and feff doesn't run or something even worse happens, you will need to configure artemis to use the correct name of the feff executable. You can do this by choosing "Edit preferences" from the Settings menu, then setting "Feff->feff_exectuable" to the correct program name. The version of feff6 that comes with Ifeffit is called "feff6L". This is the same procedure you use to run feff7 or feff8 with Artemis. Thanks to Scott and Gerrit for making me aware of this problem. Boy, I am so looking forward to going home today! B -- Bruce Ravel ----------------------------------- ravel@phys.washington.edu Code 6134, Building 3, Room 405 Naval Research Laboratory phone: (1) 202 767 2268 Washington DC 20375, USA fax: (1) 202 767 4642 NRL Synchrotron Radiation Consortium (NRL-SRC) Beamlines X11a, X11b, X23b National Synchrotron Light Source Brookhaven National Laboratory, Upton, NY 11973 My homepage: http://feff.phys.washington.edu/~ravel EXAFS software: http://feff.phys.washington.edu/~ravel/software/exafs/
participants (3)
-
Bruce Ravel -
Schmithals -
Scott Calvin