[Wien] Default options in Geometry Optimizer
Dear Prof. Blaha and Wiens2k users, I had a few clarifications on the geometry optimizer ( I did not get it in the UG) 1. There is an option in the window of geometry optimizer. That is the Maximum number of structure changes. If we do not enter any number, what is the default value the optimizer takes. I did not enter any number, still the optimizer seems to be running. Thanks and with regards Suddhasattwa Ghosh -- next part -- An HTML attachment was scrubbed... URL: http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20100111/18ba830d/attachment.htm
[Wien] [Wien2k User Group] Volume optimization of Zirconium
I get more and more confused what you actually want to do. Do you want to optimize the c/a ratio or calculate the equation of state from a volume optimization? If I understand the output you supplied correctly, you tried to obtain the bulk modulus from the optimization of the c/a ratio, which doesn't make sense. Moreover, you must of course use more than just two points for the fit of the equation of states! Have you familiarized yourself with the TiC example which is explained in the UG? Best regards Michael -- Preisknaller: GMX DSL Flatrate f?r nur 16,99 Euro/mtl.! http://portal.gmx.net/de/go/dsl02
[Wien] [Wien2k User Group] Volume optimization of Zirconium
Dear Michael, I guess the values of the energy volume and pressure comes in the same window as the energy curve. I am not able to figure what you actually mean by saying that optimize c/a ratio and equation of state from volume optimization. The energy curve in fact comes from optimize.job and fitting parameters by fitting a suitable eos. Suddhasattwa -Original Message- From: wien-boun...@zeus.theochem.tuwien.ac.at [mailto:wien-bounces at zeus.theochem.tuwien.ac.at] On Behalf Of Michael Fischer Sent: Monday, January 11, 2010 1:57 PM To: wien at zeus.theochem.tuwien.ac.at Subject: [Wien] [Wien2k User Group] Volume optimization of Zirconium I get more and more confused what you actually want to do. Do you want to optimize the c/a ratio or calculate the equation of state from a volume optimization? If I understand the output you supplied correctly, you tried to obtain the bulk modulus from the optimization of the c/a ratio, which doesn't make sense. Moreover, you must of course use more than just two points for the fit of the equation of states! Have you familiarized yourself with the TiC example which is explained in the UG? Best regards Michael -- Preisknaller: GMX DSL Flatrate f?r nur 16,99 Euro/mtl.! http://portal.gmx.net/de/go/dsl02 ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
[Wien] [Wien2k User Group] Volume optimization of Zirconium
If this is the entire content of your case.outputeos: Vol energy de(Birch-Murnaghan) Pressure 314.2133 -14396.851 0.009235-1.432 314.2133 -14396.848 0.005909 -1.432 then you are make a eos fit through 1 single E(V) point... (actually through two, but they are the same). This doesn't make sense, as there are 4 free parameters in this eos. That's why you find very much rounded values for B and BP: V0 B(GPa) BP EO 309.5778 100. 5.000 -14396.842 (100 and 5, they are probably just the defaults). Make your eos fit trhough at least 4 data points (5-10 is better). They should all be listed as (:VOL, :ENE) in case.vol. Negative pressures correspond to volumes larger than the equilibrium volume. Stefaan Ghosh SUDDHASATTWA wrote: Dear Micheal, As suggested, I did the volume optimization with -5% to +5% The values are as shown for alpha-zirconium. V0 B(GPa) BP EO 309.5778 100. 5.000 -14396.842 Vol energy de(Birch-Murnaghan) Pressure 314.2133 -14396.851 0.009235-1.432 314.2133 -14396.848 0.005909 -1.432 ; Are these values possible. I am really amazed to see the value of pressure which is negative The ground state energy is I guess -14396.842. Am I correct? With spin orbit coupling, the ENE value was -14396.892 which are pretty close. Then why these pressure values are negative. I seek one more clarification. In the energy plot curve option, there comes by using *. Scf files. If we want to select a few files and not all the files for plotting, then do we have to enter the file names with commas separating them? I ask this question because the points from the previous volume optimization also are coming in the current plot. Thanks Suddhasattwa Ghosh -Original Message- From: wien-bounces at zeus.theochem.tuwien.ac.at [mailto:wien-bounces at zeus.theochem.tuwien.ac.at] On Behalf Of Michael Fischer Sent: Sunday, January 10, 2010 3:44 AM To: wien at zeus.theochem.tuwien.ac.at Subject: Re: [Wien] [Wien2k User Group] Volume optimization of Zirconium Hello, I can only guess on this problem, but I am quite sure that changes of the c/a ratio from -25% to +25% are much too large, since this probably leads to an overlap of the RMT spheres. Try a set of calculations for smaller changes of the c/a ratio, e.g. from -5% to +5%, and make sure beforehand that the RMT spheres do not overlap in the most strongly distorted structures. Regards Michael ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien -- Stefaan Cottenier Center for Molecular Modeling (CMM) Ghent University Technologiepark 903 (2nd floor) BE-9052 Zwijnaarde Belgium http://molmod.Ugent.be email: Stefaan . Cottenier /at/ UGent . be
[Wien] [Wien2k User Group] Volume optimization of Zirconium
optimizing c/a and optimizing the volume are two different things. The fit with some eos should only be done for volume optimization, but of course NOT for c/a optimizations. Maybe this is the reason for your scatter ??? When you read the UG, chapeter 5.3 you will find more information. The analysis with an eos is only for VOLUME !!! and needs an input file case.analysis with volume/e-tot. For EACH volume you can do a c/a optimization and then put the best energy for this volume into case.analysis . In order to be able to do this, you need to understand what you are doing. Another possibility is to use parabolfit_lapw (see UG). However, there is even an alternative listed in the UG (did you read it ?). It is called 2Doptimize This is probably what you want to use in your case. Ghosh SUDDHASATTWA schrieb: Dear Michael, I guess the values of the energy volume and pressure comes in the same window as the energy curve. I am not able to figure what you actually mean by saying that optimize c/a ratio and equation of state from volume optimization. The energy curve in fact comes from optimize.job and fitting parameters by fitting a suitable eos. Suddhasattwa -Original Message- From: wien-bounces at zeus.theochem.tuwien.ac.at [mailto:wien-bounces at zeus.theochem.tuwien.ac.at] On Behalf Of Michael Fischer Sent: Monday, January 11, 2010 1:57 PM To: wien at zeus.theochem.tuwien.ac.at Subject: [Wien] [Wien2k User Group] Volume optimization of Zirconium I get more and more confused what you actually want to do. Do you want to optimize the c/a ratio or calculate the equation of state from a volume optimization? If I understand the output you supplied correctly, you tried to obtain the bulk modulus from the optimization of the c/a ratio, which doesn't make sense. Moreover, you must of course use more than just two points for the fit of the equation of states! Have you familiarized yourself with the TiC example which is explained in the UG? Best regards Michael -- P.Blaha -- Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna Phone: +43-1-58801-15671 FAX: +43-1-58801-15698 Email: blaha at theochem.tuwien.ac.atWWW: http://info.tuwien.ac.at/theochem/ --
[Wien] [Wien2k User Group] Volume optimization of Zirconium
Dear Stefaan, I have only given two points. There are in fact 20 points. However, I want to ask you about E0 which more or less matches with the ENE after the SCF cycle with spin orbit coupling. Suddhasattwa -Original Message- From: wien-boun...@zeus.theochem.tuwien.ac.at [mailto:wien-bounces at zeus.theochem.tuwien.ac.at] On Behalf Of Stefaan Cottenier Sent: Monday, January 11, 2010 1:18 PM To: A Mailing list for WIEN2k users Subject: Re: [Wien] [Wien2k User Group] Volume optimization of Zirconium If this is the entire content of your case.outputeos: Vol energy de(Birch-Murnaghan) Pressure 314.2133 -14396.851 0.009235-1.432 314.2133 -14396.848 0.005909 -1.432 then you are make a eos fit through 1 single E(V) point... (actually through two, but they are the same). This doesn't make sense, as there are 4 free parameters in this eos. That's why you find very much rounded values for B and BP: V0 B(GPa) BP EO 309.5778 100. 5.000 -14396.842 (100 and 5, they are probably just the defaults). Make your eos fit trhough at least 4 data points (5-10 is better). They should all be listed as (:VOL, :ENE) in case.vol. Negative pressures correspond to volumes larger than the equilibrium volume. Stefaan Ghosh SUDDHASATTWA wrote: Dear Micheal, As suggested, I did the volume optimization with -5% to +5% The values are as shown for alpha-zirconium. V0 B(GPa) BP EO 309.5778 100. 5.000 -14396.842 Vol energy de(Birch-Murnaghan) Pressure 314.2133 -14396.851 0.009235-1.432 314.2133 -14396.848 0.005909 -1.432 ; Are these values possible. I am really amazed to see the value of pressure which is negative The ground state energy is I guess -14396.842. Am I correct? With spin orbit coupling, the ENE value was -14396.892 which are pretty close. Then why these pressure values are negative. I seek one more clarification. In the energy plot curve option, there comes by using *. Scf files. If we want to select a few files and not all the files for plotting, then do we have to enter the file names with commas separating them? I ask this question because the points from the previous volume optimization also are coming in the current plot. Thanks Suddhasattwa Ghosh -Original Message- From: wien-bounces at zeus.theochem.tuwien.ac.at [mailto:wien-bounces at zeus.theochem.tuwien.ac.at] On Behalf Of Michael Fischer Sent: Sunday, January 10, 2010 3:44 AM To: wien at zeus.theochem.tuwien.ac.at Subject: Re: [Wien] [Wien2k User Group] Volume optimization of Zirconium Hello, I can only guess on this problem, but I am quite sure that changes of the c/a ratio from -25% to +25% are much too large, since this probably leads to an overlap of the RMT spheres. Try a set of calculations for smaller changes of the c/a ratio, e.g. from -5% to +5%, and make sure beforehand that the RMT spheres do not overlap in the most strongly distorted structures. Regards Michael ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien -- Stefaan Cottenier Center for Molecular Modeling (CMM) Ghent University Technologiepark 903 (2nd floor) BE-9052 Zwijnaarde Belgium http://molmod.Ugent.be email: Stefaan . Cottenier /at/ UGent . be ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
[Wien] [Wien2k User Group] Volume optimization of Zirconium
Dear Prof.Blaha, Thanks for the information. I have read 2D optimization in UG. I initially wanted to do a 2D optimization; however in the volume optimization window, there are only 7 options and the 8th option (2D optimization by Prof. Jamal) is not there. Suddhasattwa -Original Message- From: wien-boun...@zeus.theochem.tuwien.ac.at [mailto:wien-bounces at zeus.theochem.tuwien.ac.at] On Behalf Of Peter Blaha Sent: Monday, January 11, 2010 3:45 PM To: A Mailing list for WIEN2k users Subject: Re: [Wien] [Wien2k User Group] Volume optimization of Zirconium optimizing c/a and optimizing the volume are two different things. The fit with some eos should only be done for volume optimization, but of course NOT for c/a optimizations. Maybe this is the reason for your scatter ??? When you read the UG, chapeter 5.3 you will find more information. The analysis with an eos is only for VOLUME !!! and needs an input file case.analysis with volume/e-tot. For EACH volume you can do a c/a optimization and then put the best energy for this volume into case.analysis . In order to be able to do this, you need to understand what you are doing. Another possibility is to use parabolfit_lapw (see UG). However, there is even an alternative listed in the UG (did you read it ?). It is called 2Doptimize This is probably what you want to use in your case. Ghosh SUDDHASATTWA schrieb: Dear Michael, I guess the values of the energy volume and pressure comes in the same window as the energy curve. I am not able to figure what you actually mean by saying that optimize c/a ratio and equation of state from volume optimization. The energy curve in fact comes from optimize.job and fitting parameters by fitting a suitable eos. Suddhasattwa -Original Message- From: wien-bounces at zeus.theochem.tuwien.ac.at [mailto:wien-bounces at zeus.theochem.tuwien.ac.at] On Behalf Of Michael Fischer Sent: Monday, January 11, 2010 1:57 PM To: wien at zeus.theochem.tuwien.ac.at Subject: [Wien] [Wien2k User Group] Volume optimization of Zirconium I get more and more confused what you actually want to do. Do you want to optimize the c/a ratio or calculate the equation of state from a volume optimization? If I understand the output you supplied correctly, you tried to obtain the bulk modulus from the optimization of the c/a ratio, which doesn't make sense. Moreover, you must of course use more than just two points for the fit of the equation of states! Have you familiarized yourself with the TiC example which is explained in the UG? Best regards Michael -- P.Blaha -- Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna Phone: +43-1-58801-15671 FAX: +43-1-58801-15698 Email: blaha at theochem.tuwien.ac.atWWW: http://info.tuwien.ac.at/theochem/ -- ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
[Wien] Running on Teragrid (queenbee.loni-lsu)
I run Wien2k with MVAPICH2 + ifort 11.0.074 + Intel MKL 10.1.0.015 on a regular basis. It is very stable. I used maximum 256 cores and can provide further details, if necessary. Oleg -- Oleg Rubel, PhD Thunder Bay Regional Research Institute Homepage: http://www.tbrri.com/~orubel/ Laurence Marks L-marks at northwestern.edu 01/10/10 10:24 AM Has anyone got anywhere with running Wien2k on the Teragrid (specifically Queenbee). I'm using version 9.1 with ifort 10.1/cmkl and have compiled using both mvapich and mvapich2. So far mvapich works, but larger lapw1_mpi using 256 cores hang which I cannot get mvapich2 to even run the straight mpi benchmark. N.B., has anyone got mvapich2 and/or mpich2 working with mkl for larger mpi jobs? -- Laurence Marks Department of Materials Science and Engineering MSE Rm 2036 Cook Hall 2220 N Campus Drive Northwestern University Evanston, IL 60208, USA Tel: (847) 491-3996 Fax: (847) 491-7820 email: L-marks at northwestern dot edu Web: www.numis.northwestern.edu Chair, Commission on Electron Crystallography of IUCR www.numis.northwestern.edu/ Electron crystallography is the branch of science that uses electron scattering and imaging to study the structure of matter. ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
[Wien] wien2k.v09.2 on a single node shared memory machine
You are right, the latest siteconfig_lapw has a bug and does not honor the shared memory settings during installation. Nothing else was changed in the code and you can fix this by adding a line to $WIENROOT/parallel_options: setenv USE_REMOTE 0 PS: Aparently hardly anybody uses the shared memory option, because this restricts the use of WIEN2k on more than one PC. PPS: There is a new setting MPI_REMOTE, which directs/avoids calling the mpirun command via ssh. Saeid Jalali schrieb: Dear Peter, The latest version of the code, v09.2, needs passwordless ssh for running in parallel even on a single multi cores node in the shared memory set up. This would not be necessary as mentioned also in Sec. 5.5.1 of the usersguide, and was not also the case for earlier versions of the code, e.g. v08.2. We could run in parallel in our shared memory machine only after doing the following unnecessary steps: 1) ssh-keygen ?t dsa 2) cd .ssh 3) cat id_dsa.pub authorized_keys 4) chmod 644 authorized_keys Many users may not observe the above discussed problem, as they are using clusters and not just a single node alone. Sincerely yours, S. Jalali /_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/ Saeid Jalali Asadabadi, Department of Physics, Faculty of Science, University of Isfahan (UI), Hezar Gerib Avenue, 81744 Isfahan, Iran. Phones: Dep. of Phys. :+98-0311-793 2435 Office :+98-0311-793 4176 Fax No. :+98-0311-793 2409 E-mail :sjalali at phys.ui.ac.ir Homepage:http://sci.ui.ac.ir/~sjalali www:http://www.ui.ac.ir /_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/ ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
