On Jan 2, 2013, at 21:31 , Axel Kohlmeyer wrote: > this is a good idea in theory, but doesn't work in practice.
maybe not even in theory. The number of possible combinations of OS/compiler/libraries/OpenMP/MPI/whatever and of the respective versions is just too big. I used to save in the user guide sufficiently detailed reports of trouble, but most reports are not informative enough, to put it mildly, and anyway it was a waste of time. By the way: years ago I have seen two supposedly identical machines, installed with supposedly the same software, yield different results (one working, one "segmentation fault") P. --- Paolo Giannozzi, Dept of Chemistry&Physics&Environment, Univ. Udine, via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222
