Virtual move Monte Carlo #1969
Draft
Virtual move Monte Carlo #1969
Conversation
This file contains 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
ClusterData is a more accurate name than LinkerData since it stores information about the cluster that's being built up, which does contain information about the linkers, but that's not it's main purpose.
The way I implemented the rotate moves, the cluster's center of rotation was always the origin. This has the effect that the particles became confined to the insphere of the simulation box. I only noticed this when I ran simulations of larger systems.
and some other stuff.
Must totally abort the cluster move if any of the linker particles start or end up in the inactive region. Perhaps an optimization opportunity: once a linkee has been definitely added to the cluster, check its current and trial position to see if it starts or ends up in the inactive region. That would give you a much earlier early exit (compared to what we do now) since you wouldn't have to continue checking the rest of the current or following linkers' neighbors for overlaps/pair energies.
This is the mode that allows for the cluster cleaving version of the algorithm as described in the SI of Whitelam and Geissler.
I was getting arbitrary overlaps in systems before this.
…l-move-monte-carlo
tcmoore3
force-pushed
the
feature-virtual-move-monte-carlo
branch
from
80d37d7 to
5403695
Compare
Best to get the main implementation into HOOMD and then test experimental features since it will be easier to validate the straightforward implementation.
…option Also count the number of AABBTree builds triggered by the VMMC updater.
Move buildAABBTree to end of loop so that I minimize the number of times I get the array handles.
This was causing the tree to never be rebuilt because the particles were wrapped back into the primary image before the displacement check was performed. And since the diaplacement check used the minimum plane-to-plane distance, the particle could never really move far enough to trigger a rebuild. This may have been leading to rare overlaps.
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.
Description
Motivation and context
Resolves #???
How has this been tested?
Checklist:
sphinx-doc/credits.rst) in the pull request source branch.CHANGELOG.rstfollowing the established format.