Sander Vandenhaute

> This is something I've discussed with the OpenMM people without coming to a great conclusion. Obviously, the ideal scenario would be where `openmm-torch` is able to collect the neighborlist...

> There is not. `torch_cluster` is only without PBC, and we need full triclinic PBC. My vote again goes for trying to translate the ASE neighborlist — which is thoroughly...

The contribution of the total force on a given atom i from each of the other atoms j is not kept track of in OpenMM, so the formula you are...

@chen850512337 : sorry for the late reply. [This](http://d-nb.info/1079218084/34) PhD thesis, chapter 5, states the definition of virial stress in terms of a derivative of the energy with respect to components...

Good to know! I typically just use the Nvidia GPU for compute only, not for screen rendering. I suppose I cannot just tell viamd to use the nvidia one —...

Right, but given that the pytorch evaluation constitutes most of the calculation time (at least for relatively small systems and complex torch models), this would turn out to be rather...

Hi @b-mazur , The energy of the isolated atoms depends on the specific pseudopotential used as well as the functional, so tabulating those would be quite an amount of work....

It's one of the things we're fixing in the next release. For CP2K in particular, it's currently still necessary to put `cpu_affinity: none` in the .yaml. Perhaps that could fix...

I honestly don't know. If you have tried both `mpirun -np X` and `mpirun -np X -bind-to core -rmk user -launcher fork` then I wouldn't know. The MPI in the...

@b-mazur the first release candidate for v4.0.0 is out, in case you want to try again.