- Ifeffit - millenia.cars.aps.anl.gov

Re: Artemis Crashed too often
by Aria 29 Apr '04

29 Apr '04

After reading these days' discussion in 'minor bugs report' thread, it reminds me of keeping a track of memory usage in my case. Here is a detailed list. Artemis starts: 56,100 K Loads project file: 59,180 K Save: 65,256 K After 1st fit: 88,252 K 2nd fit: 88,824 K 3rd fit: 89,320 K Save: 98,160 K 4th fit: 97,840 K Save: 98,164 K 5th fit: 298,152 K Crash I have also noticed the CPU usage is normally 25% during the fit process. But in the 5th fit, it went up to 25% for first several seconds, then drops to 0% suddenly. The memory usage jumps to 298,152K at the same time, and stayed for a few seconds before it finally crashed. Meanwhile, another process called 'dwwin.exe' with memory usage of 248,016K appeared. After the error message showed up, and closed all the Artemis windows, both of those big memory consuming processes disappeared. This practice is reproducible. I hope it helps a little bit. Best regards, Aria

1 0

RE: [Ifeffit] Minor bugs in the new versions
by Jensen, Mark P. 29 Apr '04

29 Apr '04

I agree with Gerrit on this. This morning I took a look at the two "nearly identical" computers and found that 1. Athena and Artemis each consume about 230 MB of memory to run on my Windows 2000 or XP computers. 2. That the computer that could run both Athena and Artemis with 128 MB RAM had a considerably larger maximum virtual memory setting (600 MB) than the computer that would not (300 MB). Increasing the size of the virtual memory on the latter computer allows both to run. Bruce, as long as that is the amount of memory the programs should use, I guess that means this is not a minor bug in your software. In my case, it's a user that was being miserly with hard drive space. Mark Jensen ------------------------------ Mark P. Jensen Chemistry Division Argonne National Laboratory 9700 South Cass Avenue Argonne, Illinois 60439-4831 USA 630-252-3670 630-252-7501 FAX mjensen(a)anl.gov -----Original Message----- From: Schmithals [mailto:schmithals@hmi.de] Sent: Thursday, April 29, 2004 3:06 AM To: XAFS Analysis using Ifeffit Subject: Re: [Ifeffit] Minor bugs in the new versions Matt Newville schrieb: > 2 I can definitely run artemis and athena at the same on a Windows > machine, so I doubt there's a lock on Perl's Ifeffit.dll. I'd > assume that if other people could only run one of these at a > time there would be many complaints. > > 3 On my WinXP box, athena and artemis each use about 55Mb at > startup -- without reading in any data. That seems to be about > 50% Ifeffit / 50% Perl/Tk (the ifeffit shell takes ~24Mb). > I don't know enough about what changed to guess why the memory > usage would've changed between versions of artemis and athena, > but I think they should both be able to run in 256Mb. Anyway, > I'd suggest watching the memory usage in TaskManager as you > start athena and artemis to see if it really is a memory issue. I now think its a memory problem, too. My computer is running Win2000, 256 MB of RAM, virtual memory is fixed to 400 MB. I started the Windows Task manager to watch memory usage: - without any programs running (except virus scanner, ..) its 119 MB - After starting Artemis it goes up to 412 MB - Athena won't run Then I set the virtual memory to fixed 600 MB. - After starting Artemis it again takes ~400 MB - Athena now starts! Memory usage is ~700 MB. But shutting down one program before starting the other has been OK for me, too. Regards, Gerrit _______________________________________________ Ifeffit mailing list Ifeffit(a)millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

2 1

Minor bugs in the new versions
by Schmithals 29 Apr '04

29 Apr '04

Dear Bruce, I installed the new versions of the programs. Many thanks for your ongoing work and especially for the new feature of exporting parameter values in Athena. While working with the new versions I listed the problems that I still ran over (all just minor problems that are not critical or for which I figured out a workaround). Some of the bugs have already been mentioned, but I was not sure if they should already be fixed with this release - if not please just ignore these points. Hope it helps to improve the software. My System is Win2000. Artemis 0.7.003 - After the update the path to the stash-folder is set to the default path: "c:\Program Files\Ifeffit\horae\stash" although my Ifeffit is installed at "c:\Programme\Ifeffit" and previously the stash folder was "c:\Programme\Ifeffit\horae\stash". Is it possible to set the stash folder automatically to the subfolder in the installation directory? Obviously, this should be corrected via: Options | General | Workspace. But Artemis will not let me do that. I included the messages that Artemis gives when I change the path and press "Save changes to future sessions" and "return to main window". In the latter case Artemis will ask if I want to keep changes for future sessions (obviously it recognizes that changes have not been saved) but will not return to the main program (and not change the path). However, it does affect working with Artemis in any other way (as far as I can tell). - Atoms potential cards (in feff.inp) is sometimes wrong. (work-around: enlarge r_max in atoms.inp) - Artemis will stumple if I try to import 30 paths (my default) but feff generated less than that (no paths will be loaded and and error message generated, see ARTEMIS Import.TRAP). - If I run feff, import 30 paths and then do that a second time, Artemis does not delete the old entries in the "data&paths"-window but just append the new entries with names "feff0001.dat:2" etc., is that the intended behaviour? The problem is that if I decrease r_max in feff.inp before doing the second run (to a value so that feff produces less than 30 path-files) Artemis will generate error messages when trying to plot the first path entries (obviously because they dont exist anymore). - Function "discard this dataset" does not delete the main folder "data0.feff.0". Therefore, after deleting a few FEFF-datasets I end up with a lot of folders "data0.feff.x", because the empty ones are not reused. The function also does not delete the corresponding parameter values (which is probably better than deleting them). - Set-up of Atoms/Feff-page before loading a datafile will lead to errors (already discussed) Athena 0.8.027 - Athena does not start when Artemis is running. Two new Windows will show up for a moment in the taskbar and disappear again, no error logs are written to the stash-folder. No problems occur if Artemis is not running. Also Artemis will not start when Athena is running. - When using the export function via "File | Save marked groups as data" Athena will state that I may not export more than 15 datafiles. But it also produces errors with less files: it seems that the maximum line width is limited to 259 characters. The line with the names of the datagroups is usually longer and therefore gets truncated. - Access to last opened files is not released: it is not possible to rename/delete project files that just have been closed or data files that just have been read in. This is possible after a new project file is opened or a new data file imported. - Sometimes the selection of groups gets faulty: when clicking on one of the groups it will not be highlighted. Instead the previously selected group will be redisplayed (and remains highlighted). Sometimes a completly different group will be highlighted and displayed. Closing and reopening will not solve the problem, but restarting of Athena does. I could not figure out how to reproduce this error and nothing is written to the stash-folder. - Athena sometimes resizes the width of the right panel (Data groups) by itself. E.g. when exporting data via "File | Save chi(R)" the window becomes wider and gets small again when clicking on a datagroup afterwards (but not always ?!). Atoms ? - I am not sure about the version number because "Help"|"About Atoms" displays "left" (?!) instead of version number. - The "Browse Space Groups" will still invoke the browser only once. Hephaestus 0.01 - Multiple choosing of Settings | Energies will TOGGLE between display in energies/wavelengths (instead of just keeping displaying in terms of energy). Moreover, the checkmark in the Settings-Dialog will stay at "Energies", even when displaying in wavelengts. Despite of this I would propose to display the unit on the main page anyway. - On my machine the checkmarks are hard to recognize as they only shown up as two tiny red dots. But they will show up as checkmarks when I hover over them with the mouse. (This behaviour is common to all the programs, also in previous versions). I also have some general question: - Is it possible to do an offline installation on a computer not connected to the internet? (e.g. by manually loading the updates-files and changing "update_site" in updater.ini) - Is there a way in Athena to display the raw data of a group as a text file? Athena will not allow to edit this data via "Edit | Edit data as text". I only had the idea of saving the group and looking at the resulting text file. - This raised another point: if I compare the file I read into Athena with the file I exported then I notice that they have a different scaling (abszissae values) - does Athena automatically change the scaling when importing data files? I also noticed that when exporting multiple groups then all groups are scaled to the same abszissae values which is not the case when I export them separately. Regards, Gerrit

3 9

RE: [Ifeffit] Minor bugs in the new versions
by Jensen, Mark P. 28 Apr '04

28 Apr '04

Ifeffit friends, I have done several test with Athena and Artemis this afternoon, and the problem of running Athena and Artemis at the same time does not seem to be solely memory related. On two similar Win2000 machines, my old one with 128 Mb behaved as usual, only Artemis or Athena would run at one time. The first "program box" for Aretmis representing the text window would appear in the task bar, the updater program box would appear and then both would disappear instantly. I do not observe the Aretmis splash screen and she never really starts. It happens so fast that I can't read anything in the text window. This was it's "usual" behavior on the 3 computers I mentioned in the previous e-mail. This happens right from the get-go. Nothing else, other than the manditory windows services and the lab mandated anti-virus software was running. On the other machine, with 256 Mb, I could run both simultaneously. Aha! I'm on to something! When I took 128 Mb out of this 2nd computer, though, I could still start Athena and Artemis simultaneously! Rats, foiled again! Unfortunatley, I couldn't bump up that 1st computer to 256 Mb to see if that would enable both to run simultaneously on that computer. In the past, on the first computer, I have had to shut down other programs to get enough memory to run either Athena or Artemis. Maybe, people with this problem will just have to adopt my old approach and put up with this glitch. Mark ------------------------------ Mark P. Jensen Chemistry Division Argonne National Laboratory 9700 South Cass Avenue Argonne, Illinois 60439-4831 USA 630-252-3670 630-252-7501 FAX mjensen(a)anl.gov

1 0

RE: [Ifeffit] Minor bugs in the new versions
by Scott Calvin 28 Apr '04

28 Apr '04

At 02:57 PM 4/28/2004 -0500, you wrote: >1. Lack of complaints doesn't mean that he problem is rare. > Mark brings up an interesting issue. I'd kind of like to get the sense of the community on this. Given that, unlike Ifeffit itself, Artemis and Athena are beta versions (a "0" starts the version number), all of us who use that software are beta testers. Thus part of our responsibilities, in exchange for getting to use really nifty and cutting-edge software, is to report bugs to Bruce. Unfortunately, reporting bugs in a responsible way is somewhat time-consuming (not nearly as time-consuming as writing the software in the first place, of course!). For example, a really minor bug which has already been discussed is the one that occasionally makes the right panel of Athena get wide. It's not something that bothers me, but it's clearly a bug. I don't feel like I'm being helpful, though, if I report to Bruce that something like this "occasionally" happens without being able to make it happen reproducibly. So I never reported it. If, on the other hand, some bug is either making it harder for me to do what I want to do, or it seems likely to cause some other unsuspecting user grief, or I can see exactly under what circumstances it occurs, I generally investigate and make a bug report. So my question is what perspective everyone (and especially Bruce) has on the trade-off between tracking down every bug so as to give Bruce full reports and spending the time actually using this great software for analysis. --Scott Calvin Sarah Lawrence College

1 0

RE: [Ifeffit] k range
by Kelly, Shelly D. 28 Apr '04

28 Apr '04

Hi Steven, > On Tue, 27 Apr 2004, Steven Sangyun Lim wrote: > > > My question is how low can I set the k range? Is there any minimum > > range for reliable result? Please let me get out of this dilemma... > Scott and Matt both answered this question differently than I would. I guess I am very practical. If I was asking this question I would want a set of rules that would help me figure it out the k-range. So here one way to do that. 1) Look at the data in Athena to determine a good k-range to use. Here is how. 1a) As Matt suggested find a node in the chi(k) data that is around 3 inverse angstroms and set that to the kmin value. 1b) make about 6 copies of the chi(k) data set, all with the same kmin value. All the other FT values should be as you used in Artemis. (k-weight, window, dk) 1c) highlight each data set choosing a larger and larger node for kmax. Start really safe value around 8 inverse angstroms. Then pick the next node for the next copy and so forth using up all the data. For example, the first copy of the data set has a k-range [3.2 to 8.2] the next copy has k-range [3.2 to 9.3] and so forth....until the last has a k-range of [3.2 to 14] 1d) Plot the FT of the first three data sets all together (using the purple buttons to select them and the purple R button). Look at the data with the magnitude and the real or imaginary part. With a small k-range the FT signal will be very smooth curve in the magnitude, real and imaginary parts. You will tend to see two types of changes as you increase the k-range. 1) peaks will become more resolved, larger amplitude and thinner. --this is a good kind of change 2) eventually you will see noise in the FT of the data. The noise will look like high frequency pulses in the real and imaginary part of the spectra all over in R. From one data set to the next random signals will appear and disappear as you include noise in the data. --this is the bad kind of change. Keep comparing three data sets, drop the first one then add another with slightly larger k-range. You should be able to find the cross over from the good kind of a change to the bad kind of a change. 1e) By using this comparison you should be able to choose a k-range that is large but not too large. It should result in a conservative k-range to use as a starting point for your modeling. Be careful about the k-weight that you use when you make this comparison. It needs to be the same that you are using in Artemis to model the data. Larger k-weights will have smaller "good" k-ranges for the data. 2) Pull your data into Artemis. Use the good k-range that you determined from step one. 2b) Determine a model that works and look at the results including the uncertainties. 2c) Increase the k-range and compare the results. Try to decrease kmin and increase kmax. Do one thing at a time and see how the results vary. You will need to make a lot of comparisons, see how your results and uncertainties change as you adjust these values. 2d) Decide on a k-range that gives you the "best" results. 3) You may want to go back into Athena and use a standard to adjust the low k region of the chi(k) data, depending on how your FT looks at low R. This can help you stretch the kmin value. HTH Shelly

4 3

RE: [Ifeffit] k range
by Kelly, Shelly D. 28 Apr '04

28 Apr '04

Hi Steven, In general you can use as small of a k-range as you want but you loose the number of variable that you can determine and your uncertainty in those variable gets really big too. If those trade offs are acceptable then there is no problem. I assume that you want to determine a few variables so Scotts equation is all you need to determine the minimum k-range. Artemis also lists the number of independent points in the log file. That number has to be greater than the number of variables you can determine in the fit. In practice, you need about twice as many independent points as variables to get reasonable uncertainties. But if what you say about your data is true then the metal peak is not real. In the chi(k) data the signal for each peak starts at k=0 and goes until the sigma2 term damps it out, usually around 8 to 10 inverse angstroms for oxygen, but usually much larger values for atoms with more electrons. It is physically impossible for the signal of the metal peak to only exist between 6 and 8 inverse angstroms. I would compare the data set that you are having troubles with to the other data sets that you indicated gave a consistent peak in the FT. I would bet that the troubled data set has some artifact in it that is distorting the FT. Shelly -----Original Message----- From: Steven Sangyun Lim [mailto:steven.lim@yale.edu] Sent: Tuesday, April 27, 2004 6:31 PM To: ifeffit(a)millenia.cars.aps.anl.gov Subject: [Ifeffit] k range Thank you all very much for the immediate help. One more question, may I??? What can the minimum kmax be? I think it can be about 8 according to Scott. If I use kmin=2, can it be 7 or 6? The reason why I ask this is that when I plot R space with my sample, which has been reduced and needs to be confirmed if there is any metal cluster, the metal peak shows at kmax=8, but it disappears at kmax=6 or 7. At kmax=10, it is hard to tell because the spectrum is very noisy (it shows split peak). When there are extremely small metal clusters (even at CN=0.6), I can see a distinct metal peak that does not change no matter what k range and weight I use. I wonder any further advice can be suggested. Thank you for your help in advance. Steven

1 0

Substrate material question
by Paul Fons 28 Apr '04

28 Apr '04

OK, I admit this isn't an athena question (but it is about how to take the raw data for it!). I want to do biaxial strain measurements. The idea is to deposit a thin film of the material to be investigated on a substrate and then deform the plate with a form of cylindrical symmetry. One can then solve for the stress of a plate using elasticity theory. The twist is that I need a plate that is stiff and (surprise) x-ray transparent ... and heat resistant to about 250 C. Kapton would be a wonderful choice, but is only made in thicknesses up to 175 microns (hardly stiff enough). There must be someone out there in the collective iffeffit world who knows of an appropriate material. Doing some web surfing, I came across another product by DuPont called Vespar that *may* be available in thicker forms. The specs I am looking for are: a thin (semiconducting) film has to stick to the surface, heat resistant to <=250 C, thick enough to be elastically stiff, and x-ray transparent. Any suggestions? Thanks for any help that may come my way in advance! Dr. Paul Fons Senior Researcher National Institute for Advanced Industrial Science & Technology METI Center for Applied Near-Field Optics Research (CANFOR) AIST Central 4, Higashi 1-1-1 Tsukuba, Ibaraki JAPAN 305-8568 tel. +81-298-61-5636 fax. +81-298-61-2939 email: paul-fons(a)aist.go.jp The lines below are in a Japanese font 〒３０５−８５６８茨城県つくば市東１−１−1 つくば中央第４近接場光応用工学センターポール・フォンス主任研究官

5 9

k range
by Steven Sangyun Lim 27 Apr '04

27 Apr '04

Thank you all very much for the immediate help. One more question, may I??? What can the minimum kmax be? I think it can be about 8 according to Scott. If I use kmin=2, can it be 7 or 6? The reason why I ask this is that when I plot R space with my sample, which has been reduced and needs to be confirmed if there is any metal cluster, the metal peak shows at kmax=8, but it disappears at kmax=6 or 7. At kmax=10, it is hard to tell because the spectrum is very noisy (it shows split peak). When there are extremely small metal clusters (even at CN=0.6), I can see a distinct metal peak that does not change no matter what k range and weight I use. I wonder any further advice can be suggested. Thank you for your help in advance. Steven

1 0

k range
by Steven Sangyun Lim 27 Apr '04

27 Apr '04

Dear all, I always appreciate the big useful help from everyone. I have a basic question. I think a similar question has been arose long time ago. (But I lost the mail..) My sample has very small amount (under 1 wt%) and cluster size (< 1nm). The spectrum is always noisy at high k range, so I use usually k=2 to 10 or 12. It depends on the noise of the spectrum. Of course, if I go further down, it is better. My question is how low can I set the k range? Is there any minimum range for reliable result? Please let me get out of this dilemma... Thank you always. Steven

3 2