Commit Graph

3 Commits

Author SHA1 Message Date
giannozz 07685af429 Better attempt at fixing the strange behavior of force constants in
a large cell: a more general convergence criterion for the sum over
G vectors is introduced


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5016 c92efa57-630b-4861-b058-cf58834340f0
2008-06-25 11:24:22 +00:00
giannozz e0ce902420 Misc changes to phonon postprocessing tools: i) calculation of rigid-ion
term made simpler and faster if many atoms are present. It should make
any difference in final results, but on a 40-atom cell q2r.x now yields
correct results, while the previous version didn't. ii) Large automatic
arrays in ASR routine replace by allocatable so that for many atoms
the routine doesn't instantly crash any longer. The fancy ASR badly
need some serious work to reduce the amount of cpu time and memory
they use, or else they are unusable on large and not-so-large systems.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5011 c92efa57-630b-4861-b058-cf58834340f0
2008-06-20 11:37:33 +00:00
giannozz e68d8d07f0 1) two scripts for output file analysis added to pwtools/
(courtesy of CS)
2) phonon-related auxiliary codes moved from pwtools/ to PH/
   I think it is a more appropriate place; moreover the D(q)=>D(r) step
   (q2r) might be called at the end of the phonon dispersion calculation


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3970 c92efa57-630b-4861-b058-cf58834340f0
2007-06-18 13:47:18 +00:00