Fwd: Re: Use of feff8 with D-artemis?
On Thursday, July 12, 2012 05:31:19 PM you wrote:
I wonder why the new Artemis (from Demeter) is limited to feff6 or feff7 as the old ifeffit can deals with feff8 and feff9.
Is it a wish? When you published data using feff7, the reviewers first note that you should have used more actual version.
I would challenge a reviewer to come up with a reliable study demonstrating that Feff8 or Feff9 provides a statistically significant improvement to EXAFS analysis. Obviously, I am not saying that full multiple scattering, self-consistency, and the various potential improvements in Feff9 are not relevant to XANES calculations or to Feff's treatment of other spectroscopies. What I am saying is that it has not been rigorously demonstrated that those things have an impact on the calculation of the extended EXAFS such that the analysis is improved in a statistically meaningful way. In short, I find that such arguments from reviewers boil down to "Feff8 is 33% better than Feff6 because 8 is 33% bigger than 6!"
Is it planned that D-Artemis will work with feff8 too?
I am using the other programs of Demeter but still have to stick to the old Artemis as none of my files can be read by D-Artemis (always error due to feff8 or feff9 files by importing old project)
That said, there is no good reason that artemis should not be able to use Feff8 or 9 if you want to do so. I suppose that laziness is the best explanation for why it doesn't yet. The Atoms portion of Demeter writes Feff8 files. The Atoms portion of Artemis simply has the line of code for using that part of Demeter commented out. That's trivial to fix. Getting Demeter to use Feff8 instead of Feff6 is not a lot of work, but it is also not a trivial amount of work. I will put this at the top of my to do list. Given that I will soon be leaving for conferences in distant places on the globe, I probably won't be able to release a version of Demeter that uses Feff8 effectively until Spetember. B -- 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 ----------------------------------------- -- 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
Hi Bruce,
On Thu, Jul 12, 2012 at 4:04 PM, Bruce Ravel
On Thursday, July 12, 2012 05:31:19 PM you wrote:
I wonder why the new Artemis (from Demeter) is limited to feff6 or feff7 as the old ifeffit can deals with feff8 and feff9.
Is it a wish? When you published data using feff7, the reviewers first note that you should have used more actual version.
I would challenge a reviewer to come up with a reliable study demonstrating that Feff8 or Feff9 provides a statistically significant improvement to EXAFS analysis. Obviously, I am not saying that full multiple scattering, self-consistency, and the various potential improvements in Feff9 are not relevant to XANES calculations or to Feff's treatment of other spectroscopies. What I am saying is that it has not been rigorously demonstrated that those things have an impact on the calculation of the extended EXAFS such that the analysis is improved in a statistically meaningful way.
In short, I find that such arguments from reviewers boil down to "Feff8 is 33% better than Feff6 because 8 is 33% bigger than 6!"
I would say that there are some demonstrated cases where Feff8 is slightly better than Feff6 at modeling EXAFS. The most notable cases are when H is in the input file -- Feff6 is terrible at this. Feff8 has gone through many revisions, so minor version might matter here. Some of the earliest versions actually gave worse fits (in terms of goodness-of-fit statistics) to EXAFS than Feff6. Versions around 8.2.4 (if memory serves) ended up with approximately the same goodness-of-fits as Feff6. And, although few cases have been tested, and the process is hard to generalize, if multi-pole loss terms are used in Feff8.5 or later, the improvement in goodness of fit is actually quite noticeable. Whether these differences have any impact of the accuracy of fitted values is harder to determine, of course.
Is it planned that D-Artemis will work with feff8 too?
I am using the other programs of Demeter but still have to stick to the old Artemis as none of my files can be read by D-Artemis (always error due to feff8 or feff9 files by importing old project)
This is unfortunate, but I'm willing to stick up for Bruce on this. Feff8 (I have to admit I do not have a copy of Feff9) is really a very different program from Feff6. Although the Feff team does great physics, the several different calculations that Feff8 can do are quite difficult to work with in any automated way. Artemis goes through great effort to hide just how awful Feff6 and Feff8 are in this respect. Inevitably, there are subtle differences between the inputs and outputs of Feff6 and Feff8, and being able to support all variations in what Feff8 might be able to do is a tall order. Asking why Feff from around 2012 cannot write EXAFS outputs that are compatible with Feff code from around 2000 or 1996 is a reasonable question. Sadly, Bruce (nor I) have any influence on this. All said, Bruce's approach of Feff8 not bothering to support Feff8 for EXAFS seems completely reasonable to me. If the Feff team were interested in seeing Feff8 being used in DArtemis, they would have to be the ones to make that happen, as they have refused to let Bruce or I work on this. --Matt
On Thursday, July 12, 2012 05:00:00 PM Matt Newville wrote:
I would challenge a reviewer to come up with a reliable study demonstrating that Feff8 or Feff9 provides a statistically significant improvement to EXAFS analysis. Obviously, I am not saying that full multiple scattering, self-consistency, and the various potential improvements in Feff9 are not relevant to XANES calculations or to Feff's treatment of other spectroscopies. What I am saying is that it has not been rigorously demonstrated that those things have an impact on the calculation of the extended EXAFS such that the analysis is improved in a statistically meaningful way.
In short, I find that such arguments from reviewers boil down to "Feff8 is 33% better than Feff6 because 8 is 33% bigger than 6!"
I would say that there are some demonstrated cases where Feff8 is slightly better than Feff6 at modeling EXAFS. The most notable cases are when H is in the input file -- Feff6 is terrible at this.
Feff8 has gone through many revisions, so minor version might matter here. Some of the earliest versions actually gave worse fits (in terms of goodness-of-fit statistics) to EXAFS than Feff6. Versions around 8.2.4 (if memory serves) ended up with approximately the same goodness-of-fits as Feff6. And, although few cases have been tested, and the process is hard to generalize, if multi-pole loss terms are used in Feff8.5 or later, the improvement in goodness of fit is actually quite noticeable.
Whether these differences have any impact of the accuracy of fitted values is harder to determine, of course.
Alrighty, fair enough. I admit to being a little vague about what the different minor versions of Feff8 do. And I certainly agree that the multi-pole loss terms represent a substantial improvement for the interpretation of EXAFS. That said, your last sentence is the gist of my comment. To my knowledge (and I freely admit that my knowledge may be limited and inaccurate), no one has done a rigorous statistical assessment of how EXAFS analysis is affected by the various bits of theory that are in Feff8 and are not in Feff6. It's a shame that hasn't been done. It seems like rather low-hanging fruit. Regardless of the outcome, such a study would certainly have an impact on how I develop and promote my software. The other relevant issue here is that the specific version of Feff6 that comes with Ifeffit and Demeter is the *only* version of Feff that can be freely redistributed. The substantial majority of the users of my software want a package that can be installed with one download and one double-click. That specific version of Feff6 is the only one that I can guarantee will be installed correctly along with the rest of my software. If having routine, well-integrated access to later versions of Feff is important to the people on this list, I am not the person you need to talk to. You need to bring it up with John. B -- 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
On Fri, Jul 13, 2012 at 7:54 AM, Bruce Ravel
On Thursday, July 12, 2012 05:00:00 PM Matt Newville wrote:
I would challenge a reviewer to come up with a reliable study demonstrating that Feff8 or Feff9 provides a statistically significant improvement to EXAFS analysis. Obviously, I am not saying that full multiple scattering, self-consistency, and the various potential improvements in Feff9 are not relevant to XANES calculations or to Feff's treatment of other spectroscopies. What I am saying is that it has not been rigorously demonstrated that those things have an impact on the calculation of the extended EXAFS such that the analysis is improved in a statistically meaningful way.
In short, I find that such arguments from reviewers boil down to "Feff8 is 33% better than Feff6 because 8 is 33% bigger than 6!"
I would say that there are some demonstrated cases where Feff8 is slightly better than Feff6 at modeling EXAFS. The most notable cases are when H is in the input file -- Feff6 is terrible at this.
Feff8 has gone through many revisions, so minor version might matter here. Some of the earliest versions actually gave worse fits (in terms of goodness-of-fit statistics) to EXAFS than Feff6. Versions around 8.2.4 (if memory serves) ended up with approximately the same goodness-of-fits as Feff6. And, although few cases have been tested, and the process is hard to generalize, if multi-pole loss terms are used in Feff8.5 or later, the improvement in goodness of fit is actually quite noticeable.
Whether these differences have any impact of the accuracy of fitted values is harder to determine, of course.
Alrighty, fair enough. I admit to being a little vague about what the different minor versions of Feff8 do. And I certainly agree that the multi-pole loss terms represent a substantial improvement for the interpretation of EXAFS.
That said, your last sentence is the gist of my comment.
To my knowledge (and I freely admit that my knowledge may be limited and inaccurate), no one has done a rigorous statistical assessment of how EXAFS analysis is affected by the various bits of theory that are in Feff8 and are not in Feff6.
It's a shame that hasn't been done. It seems like rather low-hanging fruit. Regardless of the outcome, such a study would certainly have an impact on how I develop and promote my software.
John and I have run statistical tests of fits using various feff versions in the past. I still have the test suite, and I believe they've run the suite several times on various versions of Feff8. Getting definitive tests is not as easy as it might appear, as you'd really want perfect data on simple, well-known structures (like, I would not expect Mn2O3 or even MnO or FeO to be good candidates). I believe that we used some Cu metal data you and I measured long ago, and some Ag and Au metal foil data. I have been made skeptical of the quality of the Ag data used. In any event, the tests were all on dense metal systems, which is a very skewed test, and not highly relevant for many analyses. Like you say, while Feff6 is not perfect, demonstrating a robust improvement is challenging. And I agree that the issue is worth re-visiting. What we would need is a representative test suite of very good data on simple structures, probably with XRD data. --Matt
Hello, I know you have lot's of work with demeter and even more with the new version based on Larch, but I wanted to know if it is still foreseen that the use of feff84 will be implemented? Even just the possibility to load feffnnnn files already calculated (if easier). I proceed to a test on Am aquo(III) species doing one shell analysis using feff84 and feff6.exe (as provided in Demeter) and I get "different" results. As all my previous data are analysed with feff84 (or feff9 with no significant differences) and I can not use Artemis Demeter. And all files are incompatible with Demeter artemis (as feff8 is used) so that I would have to start from zero again... I attached to my e-mail the old artemis project with both fits and the project with demeter artemis but only feff6 results. As expected Demeter artemis or old artemis give identical results using for both feff6. Thanks, Kathy Dardenne -----Original Message----- From: ifeffit-bounces@millenia.cars.aps.anl.gov [mailto:ifeffit-bounces@millenia.cars.aps.anl.gov] On Behalf Of Bruce Ravel Sent: Friday, July 13, 2012 2:55 PM To: XAFS Analysis using Ifeffit Subject: Re: [Ifeffit] Fwd: Re: Use of feff8 with D-artemis? On Thursday, July 12, 2012 05:00:00 PM Matt Newville wrote:
I would challenge a reviewer to come up with a reliable study demonstrating that Feff8 or Feff9 provides a statistically significant improvement to EXAFS analysis. Obviously, I am not saying that full multiple scattering, self-consistency, and the various potential improvements in Feff9 are not relevant to XANES calculations or to Feff's treatment of other spectroscopies. What I am saying is that it has not been rigorously demonstrated that those things have an impact on the calculation of the extended EXAFS such that the analysis is improved in a statistically meaningful way.
In short, I find that such arguments from reviewers boil down to "Feff8 is 33% better than Feff6 because 8 is 33% bigger than 6!"
I would say that there are some demonstrated cases where Feff8 is slightly better than Feff6 at modeling EXAFS. The most notable cases are when H is in the input file -- Feff6 is terrible at this.
Feff8 has gone through many revisions, so minor version might matter here. Some of the earliest versions actually gave worse fits (in terms of goodness-of-fit statistics) to EXAFS than Feff6. Versions around 8.2.4 (if memory serves) ended up with approximately the same goodness-of-fits as Feff6. And, although few cases have been tested, and the process is hard to generalize, if multi-pole loss terms are used in Feff8.5 or later, the improvement in goodness of fit is actually quite noticeable.
Whether these differences have any impact of the accuracy of fitted values is harder to determine, of course.
Alrighty, fair enough. I admit to being a little vague about what the different minor versions of Feff8 do. And I certainly agree that the multi-pole loss terms represent a substantial improvement for the interpretation of EXAFS. That said, your last sentence is the gist of my comment. To my knowledge (and I freely admit that my knowledge may be limited and inaccurate), no one has done a rigorous statistical assessment of how EXAFS analysis is affected by the various bits of theory that are in Feff8 and are not in Feff6. It's a shame that hasn't been done. It seems like rather low-hanging fruit. Regardless of the outcome, such a study would certainly have an impact on how I develop and promote my software. The other relevant issue here is that the specific version of Feff6 that comes with Ifeffit and Demeter is the *only* version of Feff that can be freely redistributed. The substantial majority of the users of my software want a package that can be installed with one download and one double-click. That specific version of Feff6 is the only one that I can guarantee will be installed correctly along with the rest of my software. If having routine, well-integrated access to later versions of Feff is important to the people on this list, I am not the person you need to talk to. You need to bring it up with John. B -- 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
participants (3)
-
Bruce Ravel -
Kathy Dardenne -
Matt Newville