Re: [Wien] Force on frozen atoms in a MSR1a
I also asume that you are constraining positions in case.inM . There are of course cases where constraining atomic positions makes a lot of sense, but in any case if a scheme offers a feature, it should work. And in fact, it works perfectly well with MSR1a. We use it often and it does NOT prevent the automatic stop. I guess you are fooled by the fact that the forces on the other atoms might already be smaller than the convergence criterium, however, MSR1a still moves around the positions of some atoms and if the movement is too large, the convergence criterium is not fulfilled. You can see this in the :FR label This feature is in particular useful if there is a soft potential energy landscape and small forces move atoms a long way. In this way "false" convergence is circumvented. You can always change to MSR1 by using touch .minstop and it will continue to fully converge the forces at fixed positions. Eventually you may observe that the converged forces are bigger than the threshold and the small non-scf forces came from a balance of non-scf and atomic movements. Am 13.05.2014 16:36, schrieb Salman Zarrini: Dear Wien2k user, All the convergence criteria including energy, charge and total force on all the free atoms have already met their thresholds in my MSR1a calculation, however, the total force on all the frozen atoms are far away from the force convergence threshold, and so, the job is still running. So, I was wondering if the frozen atoms are included from force convergence in a MSR1a minimization? If so, is there any way to exclude frozen atoms and if does it make sense? Cheers, Salman Zarrini ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html -- - Peter Blaha Inst. Materials Chemistry, TU Vienna Getreidemarkt 9, A-1060 Vienna, Austria Tel: +43-1-5880115671 Fax: +43-1-5880115698 email: pbl...@theochem.tuwien.ac.at - ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
Re: [Wien] Force on frozen atoms in a MSR1a
N.B., the only time it would make sense is if you are calculating something like an activation energy barrier, in which case just do "touch .stop". Fixing atoms, for instance of a substrate, is bad physics in my opinion. __ Laurence Marks Dept Mat Sci & Eng Northwestern University www.numis.northwestern.edu 847 491 3996 On May 13, 2014 9:47 AM, "Laurence Marks" wrote: > I assume by "frozen atoms" you mean. ones where you forced via > case.inM there to be no movement. This make no sense in MSR1a. > > No sense. > > Contrary to what people think, it does NOT improve the convergence to > fix the positions of certain atoms, and I consider it to be > inappropriate physics. The reason is that the convergence of MSR1a > depends upon the number of clusters of eigenvalues the combined > density-atoms problem, not on the number of atoms. Indeed, this is > also true of PORT and other methods such as conjugant gradients. Alas, > the unpublished literature in DFT (and other fields) gets this wrong. > > N.B., if instead by "frozen atoms" you mean ones which have via > symmetry no forces, this is different and indicates that there is > something wrong with your structure. > > On Tue, May 13, 2014 at 9:36 AM, Salman Zarrini > wrote: > > > > Dear Wien2k user, > > > > All the convergence criteria including energy, charge and total force > > on all the free atoms have already met their thresholds in my MSR1a > > calculation, however, the total force on all the frozen atoms are far > > away from the force convergence threshold, and so, the job is still > > running. So, I was wondering if the frozen atoms are included from > > force convergence in a MSR1a minimization? If so, is there any way to > > exclude frozen atoms and if does it make sense? > > > > Cheers, > > > > Salman Zarrini > > > > > > ___ > > Wien mailing list > > Wien@zeus.theochem.tuwien.ac.at > > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien > > SEARCH the MAILING-LIST at: > http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html > > > > -- > Professor Laurence Marks > Department of Materials Science and Engineering > Northwestern University > www.numis.northwestern.edu 1-847-491-3996 > Co-Editor, Acta Cryst A > "Research is to see what everybody else has seen, and to think what > nobody else has thought" > Albert Szent-Gyorgi > ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
Re: [Wien] Force on frozen atoms in a MSR1a
I assume by "frozen atoms" you mean. ones where you forced via case.inM there to be no movement. This make no sense in MSR1a. No sense. Contrary to what people think, it does NOT improve the convergence to fix the positions of certain atoms, and I consider it to be inappropriate physics. The reason is that the convergence of MSR1a depends upon the number of clusters of eigenvalues the combined density-atoms problem, not on the number of atoms. Indeed, this is also true of PORT and other methods such as conjugant gradients. Alas, the unpublished literature in DFT (and other fields) gets this wrong. N.B., if instead by "frozen atoms" you mean ones which have via symmetry no forces, this is different and indicates that there is something wrong with your structure. On Tue, May 13, 2014 at 9:36 AM, Salman Zarrini wrote: > > Dear Wien2k user, > > All the convergence criteria including energy, charge and total force > on all the free atoms have already met their thresholds in my MSR1a > calculation, however, the total force on all the frozen atoms are far > away from the force convergence threshold, and so, the job is still > running. So, I was wondering if the frozen atoms are included from > force convergence in a MSR1a minimization? If so, is there any way to > exclude frozen atoms and if does it make sense? > > Cheers, > > Salman Zarrini > > > ___ > Wien mailing list > Wien@zeus.theochem.tuwien.ac.at > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien > SEARCH the MAILING-LIST at: > http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html -- Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu 1-847-491-3996 Co-Editor, Acta Cryst A "Research is to see what everybody else has seen, and to think what nobody else has thought" Albert Szent-Gyorgi ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
[Wien] Force on frozen atoms in a MSR1a
Dear Wien2k user, All the convergence criteria including energy, charge and total force on all the free atoms have already met their thresholds in my MSR1a calculation, however, the total force on all the frozen atoms are far away from the force convergence threshold, and so, the job is still running. So, I was wondering if the frozen atoms are included from force convergence in a MSR1a minimization? If so, is there any way to exclude frozen atoms and if does it make sense? Cheers, Salman Zarrini ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
