Repository navigation
Conversation
Merged
This branch was previously deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Closes #56.
The energy kernel only caught a trial water atom on top of a system atom (
r2 < 1e-6) when the system wasn't alchemical. In an alchemical system, a non-ghost atom went through the hard LJ and Coulomb path with no clamp, so an exact overlap gave inf or NaN rather than the overlap energy. The check now applies to every non-ghost atom, whether or not the system is alchemical. Ghost atoms still use the soft-core path, which clampsr2.In practice this was very unlikely to matter: a random trial has to land within 0.001 Å of an atom, and an inf or NaN energy was already rejected by the acceptance kernel. Only an exact overlap of opposite charges with no LJ on either atom would give −inf and be accepted.
The energy kernel launch is moved from
move()into a private_compute_energy()method, so that a test can score trial waters placed by hand.test_overlap_in_alchemical_systemputs a trial water's oxygen on a non-ghost atom insd12. It checks that this pair gets the overlap energy and that every energy for the trial is finite, on both CUDA and OpenCL.