[Wien] optimization notes- Marks2004
Sorry, Prof. Marks Can't understand the answer. - Original Message - From: Laurence Marks Date: Saturday, September 17, 2011 17:15 Subject: Re: [Wien] optimization notes- Marks2004 To: A Mailing list for WIEN2k users > Do "ls .min* " -- the file is .min_hess not case.min_hess > > Files with a "." in front of them often are hidden, i.e. they do not > show up with a normal ls command. > > 2011/9/17 Viktor Zano : > > I am running wien version 10.1 on a Intel machine (quad-?core) with > > operating system ubunto 10.04 LTS (the Lucid lynx). > > The purpose of my calculations is relaxation of the structure (mini > > positions). > > Since it didn't converge, I used Prof. Marks "optimization > notes": I used > > lower k-points (30 in IBZ instead of 120), 5 mRy/bohr, Energy conv' > > creterion: 0.005, using the command line (I didn't use the -NI > switch):> min -j 'runsp_lapw -l -i 50 -ec 0.005 -fc 5.0 it'. > > > > It did well and converges even bellow?2?mRy/bohr, Energy conv' > < 0.02. > > Now, I want to do the same calculation but with 120 k-points > in IBZ,?2 > > mRy/bohr, Energy conv' creterion: 0.0001. Acording > to?"optimization notes", > > in order to use the Hessian estimate instead of the default, I > have to copy > > .min_hess to .minrestart and delete old case.tmpM and > case.finM. The problem > > is I don't have .min_hess but I have .hess. I don't have > .minrestart, but > > have .inm_resart_st. How can I progress (I can use the new > relaxed ?atomic > > positions, but what about the hesian?) > > > > _ > > Victor Y. Zenou > > PhD student > > Department of Materials Engineering > > BGU? > > ___ > > Wien mailing list > > Wien at zeus.theochem.tuwien.ac.at > > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien > > > > > > > > -- > 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 > Research is to see what everybody else has seen, and to think what > nobody else has thought > Albert Szent-Gyorgi > ___ > Wien mailing list > Wien at zeus.theochem.tuwien.ac.at > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien > _ Victor Y. Zenou PhD student Department of Materials Engineering BGU? -- next part -- An HTML attachment was scrubbed... URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20110917/2db4bf1a/attachment.htm>
[Wien] optimization notes- Marks2004
I am running wien version 10.1 on a Intel machine (quad-?core) with operating system ubunto 10.04 LTS (the Lucid lynx). The purpose of my calculations is relaxation of the structure (mini positions).? Since it didn't converge, I used Prof. Marks "optimization notes": I used lower k-points (30 in IBZ instead of 120), 5 mRy/bohr, Energy conv' creterion: 0.005, using the command line (I didn't use the -NI switch): min -j 'runsp_lapw -l -i 50 -ec 0.005 -fc 5.0 it'. It did well and converges even bellow?2?mRy/bohr, Energy conv' < 0.02. Now, I want to do the same calculation but with 120 k-points in IBZ,?2 mRy/bohr, Energy conv' creterion: 0.0001. Acording to?"optimization notes", in order to use the Hessian estimate instead of the default, I have to copy .min_hess to .minrestart and delete old case.tmpM and case.finM. The problem is I don't have .min_hess but I have .hess. I don't have .minrestart, but have .inm_resart_st. How can I progress (I can use the new relaxed ?atomic positions, but what about the hesian?)? _ Victor Y. Zenou PhD student Department of Materials Engineering BGU? -- next part -- An HTML attachment was scrubbed... URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20110917/b8dfb687/attachment.htm>
[Wien] optimization notes- Marks2004
Do "ls .min* " -- the file is .min_hess not case.min_hess Files with a "." in front of them often are hidden, i.e. they do not show up with a normal ls command. 2011/9/17 Viktor Zano : > I am running wien version 10.1 on a Intel machine (quad-?core) with > operating system ubunto 10.04 LTS (the Lucid lynx). > The purpose of my calculations is relaxation of the structure (mini > positions). > Since it didn't converge, I used Prof. Marks "optimization notes": I used > lower k-points (30 in IBZ instead of 120), 5 mRy/bohr, Energy conv' > creterion: 0.005, using the command line (I didn't use the -NI switch): > min -j 'runsp_lapw -l -i 50 -ec 0.005 -fc 5.0 it'. > > It did well and converges even bellow?2?mRy/bohr, Energy conv' < 0.02. > Now, I want to do the same calculation but with 120 k-points in IBZ,?2 > mRy/bohr, Energy conv' creterion: 0.0001. Acording to?"optimization notes", > in order to use the Hessian estimate instead of the default, I have to copy > .min_hess to .minrestart and delete old case.tmpM and case.finM. The problem > is I don't have .min_hess but I have .hess. I don't have .minrestart, but > have .inm_resart_st. How can I progress (I can use the new relaxed ?atomic > positions, but what about the hesian?) > > _ > Victor Y. Zenou > PhD student > Department of Materials Engineering > BGU? > ___ > Wien mailing list > Wien at zeus.theochem.tuwien.ac.at > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien > > -- 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 Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi
