Hi
We are trying to compile lammps (in parallel) on Ranger supercomputing system again, after Ranger underwent some changes. Attached is a Makefile that TACC consulting (mainly Yaakoub) have generated for us. It’s compiling fine and i even used the executable generated lmp_tacc for my simulations. For system 1, the water diffusivity obtained with this new lmp_tacc is exactly same as that calculated by using lmp_ranger (old executable on ranger before ranger underwent any recent changes), however, for system 2, water diffusivity is 23 % lower than earlier value obtained from using lmp_ranger. When this was conveyed to TACC, i was told that he is using some vanilla flags for compilation, but it’s not clear if that is what is causing the trouble. I was advised to post this on lammps mailing list and so please let me know your opinion.
Thanks
Chetan
Makefile.tacc (2.8 KB)