Please specify which version of the code you are using P.
On Fri, 2012-09-21 at 11:01 +0200, Thomas Gruber wrote: > Dear all, > > I am doing phonon calculations with splitting in q-points and > irreducible representations. Do to negative frequencies at the last > q-point I do some convergence test only with the last q-point. It is > possible to calculate the whole q-point at once without calculating the > other and get the frequencies with "start_q" and "last_q". But I have to > split into representations because the system is too big. When I run > final calculation with "recover=.true." then the calculation finds all > *.xml files but it stop the the error code: > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > from openfilq : error # 1 > file LiSi.k8.q4.tr2-14.0.4.4.wfc not found > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > I found out that when I run a final calculation with several q-points > and "start_q=1" it works to every q-point, but if start_q is not equal > to 1 then I get the above error message. Why is it possible to calculate > a q-point at once but not with splitting? > > Best regards, > > Thomas Gruber > -- Paolo Giannozzi, IOM-Democritos and University of Udine, Italy