Non-omm ase #245
Non-omm ase #245
Conversation
|
I've updated the |
|
Meaning of simulation switching/reset not quite consistent right now. The ASE dynamics can't really be reset without reconstructing, and we can't do that for arbitrarily provided dynamics -- so the number of steps etc continues to accumulate (maybe this is fine?). ASE OMM always uses a default set of dynamics that can't be overriden, and are replaced each reset. Instead it could continue to reuse provided dynamics or something? |
Ragzouken
force-pushed
the
feature/non-omm-ase
branch
from
18bcc9a
to
8234f21
Compare
Ragzouken
force-pushed
the
feature/non-omm-ase
branch
from
8234f21
to
1d268c4
Compare
Ragzouken
force-pushed
the
feature/non-omm-ase
branch
from
80ed10f
to
6d4aa2f
Compare
Ragzouken
added
bug
Ragzouken
removed
bug
| "outputs": [], | ||
| "source": [ | ||
| "from ase.md import Langevin\n", | ||
| "import ase.units as ase_units\n", | ||
| "dynamics = Langevin(atoms, timestep=1.0 * ase_units.fs, temperature_K=300 * ase_units.kB, friction=1.0e-03)" | ||
| "dynamics = Langevin(atoms, timestep=1.0 * ase_units.fs, temperature_K=300, friction=1.0e-03)" |
There was a problem hiding this comment.
When setting the temperature using temperature_K I think you only need to specify the number in Kelvin, rather than converting to units of kT by multiplying by the Boltzmann constant kB (see the first note on this page - this is all I changed here!
There was a problem hiding this comment.
Mostly looks good to me, though I haven't checked how everything looks in VR. The main problem I found is that the particle_count key in FrameData doesn't exist (and probably should for any simulation, even an empty one), and I'm not sure if there are any other keys that should be being set that aren't being set at the moment.
Fixes #241