Skill: refinement=R is a factor on the background spacing, not a grading ratio - #580
Open
lmoresi wants to merge 1 commit into
Open
Skill: refinement=R is a factor on the background spacing, not a grading ratio#580lmoresi wants to merge 1 commit into
lmoresi wants to merge 1 commit into
Conversation
…ing ratio The adaptive-meshing skill described refinement=R as the finest:coarsest cell-size ratio. It is the maximum local refinement on the background spacing h0: the metric targets h in [h0/R, h0*coarsening], and coarsening="auto" takes the budget-conserving R**(1/d). The finest:coarsest ratio is therefore R**(1+1/d) — about 11 for R=5 in 2-D, not 5. The skill is the first thing a session reads about the mover, so the error propagated into every model built from it. Also records that passing refinement takes the envelope branch, which ignores amp, lo/hi_percentile, mode and power. Underworld development team with AI support from Claude Code
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The
adaptive-meshingskill describedrefinement=Ras the finest:coarsest cell-size ratio.It is the maximum local refinement on the background cell size
h0. The metric targets the envelopeh ∈ [h0/R, h0·coarsening], andcoarsening="auto"takes the budget-conservingR**(1/d). The realised finest:coarsest ratio isR**(1+1/d)— about 11 forR=5in 2-D and 8.5 in 3-D, not 5.Source of truth:
metric_density_from_gradientinsrc/underworld3/meshing/smoothing/metrics.py.The skill is the first thing a session reads about the mover, so anyone sizing a mesh from it was asking for roughly twice the grading they thought. Also records that passing
refinementtakes the envelope branch, which ignoresamp,lo/hi_percentile,modeandpower.Underworld development team with AI support from Claude Code