NVT simulations on recent versions of lammps

LAMMPS LAMMPS Mailing List Mirror
1

Hi

Is anyone else having problems with NVT simulations? We have compiled it on two separate platforms, in parallel on a linux platform (a variant of red hat) and in serial on mac with multiple input files. Both are giving the same type of problem. Small errors are accumulating during the simulation untill eventually the simulation box "explodes" and all we get out for the thermochemistry is Nan. We are running the same jobs that ran fine on a version of lamps from Spring of 2010 again on a parallel linux machine (Rocks) and on a mac in serial.

matthew

2

Hi Matthew,

Is anyone else having problems with NVT simulations? We have compiled it on two separate platforms, in parallel on a linux platform (a variant of red > hat) and in serial on mac with multiple input files. Both are giving the same type of problem. Small errors are accumulating during the simulation untill > eventually the simulation box "explodes" and all we get out for the thermochemistry is Nan. We are running the same jobs that ran fine on a version of > lamps from Spring of 2010 again on a parallel linux machine (Rocks) and on a mac in serial.

I was just wondering how would the box explode if it was an NVT
ensemble? Are you running NPT instead?

Ray