Incorporating FEFF8 into IFEFFIT
Bruce, I have just recieved a copy of FEFF8. Is there a way to incorporate it into Artemis on the windows version of IFEFFIT? In theory, we should be able to play a "shell game" with the names of the executable files and get FEFF8 to work while IFEFFIT thinks it's running FEFF6l (by changing the FEFF6l.exe file to something else, and rename the FEFF8 file to FEFF6l.exe) and then replace the DAT files to those for FEFF8. Is this possible or is it just better to run FEFF8 independently and import the .inp files into Artemis? Rich Mayes Barnes Group Dept. of Chemistry Univ. of Tennessee Knoxville, TN 37996
On Thursday 28 September 2006 07:57, Richard Mayes wrote:
I have just recieved a copy of FEFF8. Is there a way to incorporate it into Artemis on the windows version of IFEFFIT? In theory, we should be able to play a "shell game" with the names of the executable files and get FEFF8 to work while IFEFFIT thinks it's running FEFF6l (by changing the FEFF6l.exe file to something else, and rename the FEFF8 file to FEFF6l.exe) and then replace the DAT files to those for FEFF8.
Is this possible or is it just better to run FEFF8 independently and import the .inp files into Artemis?
There are at least two ways to do what you want to do: 1. You can run any version of feff outside of Artemis and import the feff.inp file using "Open files" in the File menu. This will import the feff.inp file, produce the interpretation page, and ask you haw many paths you'd like to import. 2. Go to "Edit preferences" in the Edit menu. Change feff->feff_executable to point at feff8. You may need to make that a fully resolved filename -- i.e. someting like "/usr/local/bin/feff8" or "C:\Program Files\whatever\feff8". Also, change atoms->feff_version to 8 so that the Atoms part of Artemis writes out the correct kind of feff.inp file. From there, everythiung should work normally. You may also wish to play around with atoms->template. By default, Artemis will write a feff.inp file for feff8 that does *not* do self-consistency. I made that choice because the self-consistency loop is time consuming and I want the user to make a concious decision about running feff that way. Two more notes: ** It has already been suggested that the preferences page for feff->feff_executable should involve a file selection dialog ** This question needs an entry on the Ifeffit FAQ page. B -- Bruce Ravel ---------------------------------------------- bravel@anl.gov Molecular Environmental Science Group, Building 203, Room E-165 MRCAT, Sector 10, Advanced Photon Source, Building 433, Room B007 Argonne National Laboratory phone and voice mail: (1) 630 252 5033 Argonne IL 60439, USA fax: (1) 630 252 9793 My homepage: http://cars9.uchicago.edu/~ravel EXAFS software: http://cars9.uchicago.edu/~ravel/software/
I've always been told that when doing Fourier filtering, the R-space window shouldn't have sharp corners. Artemis, for instance, allows you to vary this, and by default puts shoulders on the window. I have a user who now wants the sharp corners, claiming that it provides the best filtering. He describes this as "doing the FT in a region of R-space". What does the Assembled Multitude think of the risk of termination effects with a rectangular window? mam
Just to clarify...do you mean you're doing the fits in q-space (i.e. the back transform)? Or do you actually mean that the window used in k-space to get to R-space shouldn't have sharp corners? After all, if you're fitting in R-space, Artemis doesn't use an R-space window per se, just a fitting range. --Scott Calvin Sarah Lawrence College At 12:36 PM 9/28/2006, you wrote:
I've always been told that when doing Fourier filtering, the R-space window shouldn't have sharp corners. Artemis, for instance, allows you to vary this, and by default puts shoulders on the window. I have a user who now wants the sharp corners, claiming that it provides the best filtering. He describes this as "doing the FT in a region of R-space". What does the Assembled Multitude think of the risk of termination effects with a rectangular window? mam
This is about going to what you guys call q-space. It's often of interest to be able to display the contributions from specific
shells,
thus showing off such things as beat-nodes. For instance, in phyllomanganates, the Mn-O first shell is often not of great interest,
but
the split Mn-Mn second shell is.
mam
----- Original Message -----
From: "Scott Calvin"
Just to clarify...do you mean you're doing the fits in q-space (i.e. the back transform)? Or do you actually mean that the window used in k-space to get to R-space shouldn't have sharp corners? After all, if you're fitting in R-space, Artemis doesn't use an R-space window per se, just a fitting range.
--Scott Calvin Sarah Lawrence College
At 12:36 PM 9/28/2006, you wrote:
I've always been told that when doing Fourier filtering, the R-space window shouldn't have sharp corners. Artemis, for instance, allows you to vary this, and by default puts shoulders on the window. I have a user who now wants the sharp corners, claiming that it provides the best filtering. He describes this as "doing the FT in a region of R-space". What does the Assembled Multitude think of the risk of termination effects with a rectangular window? mam
_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
participants (4)
-
Bruce Ravel -
Matthew Marcus -
Richard Mayes -
Scott Calvin