Dear Stefaan, Thanks for the suggestion. My calculation failed within 10 iterations. In some cases, it failed at the zeroth iteration by printing out L2main: QTL-B error. So I had to change case.in1 manually. But for other cases, it failed after the 3rd iteration, which motivated me to try -in1new.
I notice the issue is discussed at the end of UG and online FAQ. However, it seems not trivial to fix it. Surely I appreciate if you can also share the experience. Jianxin On Sep 19, 2009, at 11:58 PM, Stefaan Cottenier wrote: > > If this calculation was started from scratch, then do not apply > in1new during the say 10 first iterations (= -in1new 10) (that makes > no sense together with an upper limit of 6 iterations as you use, > one typically needs 40 of them). > > Also, if this started from scratch, it might be more stable not to > apply -orb yet. First roughly converge without -orb, save, then > continue with -orb. > > Stefaan > > > Jian-Xin Zhu wrote: >> Dear Prof. Blaha and Wien2k users, >> I am running scf calculations using the following command >> runsp_lapw -orb -cc 0.001 -i 6 -in1new 1 -p >> After the first LAPW0 finishes, I see a warning >> epl: Subscript out of range. >> I dig into the package (version wien2k_09.1) and find the variable >> is in write_in1_lapw, which is in turn called by runsp_lapw (and >> should also by run_lapw). As I understand, it is related to the >> option -in1new. >> What can be done to fix it? >> Thanks, >> Jianxin >> _______________________________________________ >> 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