docs(showpieces): Homotopy Basins - #352
Draft
ofloveandhate wants to merge 4 commits into
Draft
Conversation
…nding) Third showpiece: rasterize the coefficient plane of x^d - d x - c and run the real total-degree gamma-trick homotopy at every pixel (all d paths, DoublePrecisionTracker). Hue = landing-fingerprint phase, brightness = tracker step count (the swept discriminant glows), stripes = fingerprint magnitude bands, speckle = genuine path failures. Exact coefficients throughout; no randomness -> deterministic tracked data. PNGs land in the next commit (generated via tools/refresh_doc_artifacts.py). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…hero) Generated via the new refresh manifest entry: 2x supersampled, box-downsampled to 800x450 / 1200x675. ~17M tracked paths, ~14.5 min on 12 cores. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…t–end correlation
The swept-discriminant arc formula on the page was the degree-5 one
(-4ζ(1+μ)^{-1/4}) under the degree-9 discussion; the correct elimination
gives c(μ) = -8ζ(1+μ)^{-1/8} - μ, μ = γt/(1-t). Also renames the coloring
quantity honestly: s(c) = Σ ζ^k x_k is the start–end correlation (the
unweighted landing sum is permutation-blind and identically zero here),
and distinguishes the cosmetic log|s| stripes from the singular arcs.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…d maps Reframe both windows so the family's whole branch-point ring is inside the picture (d=9: |c|=8 incl. ±8i, previously cropped; d=5: |c|=4 incl. ±4i). Each degree now renders twice from one computation: the full-bleed frame and an annotated map with labeled Re(c)/Im(c) axes, every arc center c=-(d-1)ζ marked on its dotted ring (axis points as ±R/±Ri, diagonals in exact polar form). Four PNGs total; refresh manifest updated. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Contributor
Author
|
I really enjoyed using Claude to help write this. But the wording is spongy and overdramatic. I will manually revise before accepting. |
Contributor
Author
|
I will only accept this as part of a complete human-at-the-keyboard revision of the docs, which is badly needed. Using Claude got me off the ground. It's time to do the writing myself. |
ofloveandhate
marked this pull request as draft
July 16, 2026 09:17
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.
What
A third Showpieces gallery entry: Homotopy Basins
Rasterize a window of the coefficient plane of
f(x;c) = x^d − d·x − c; at each pixel run the total-degree gamma-trick homotopyH = γt(x^d − 1) + (1−t)(x^d − dx − c)and track alldstart roots with theDoublePrecisionTracker. Every channel of the image is tracked data:s(c) = Σ ζᵏ·xₖ(c)(each path contributes start root × landing point; jumps exactly where the swept discriminant permutes destinations)log|s|(cosmetic domain coloring; whirlpools are the zeros ofs)Four frames from two computations (~17M tracked paths, ~15 min on 12 cores): per degree (d = 5 teaching, d = 9 hero) a full-bleed render and an annotated map with labeled Re(c)/Im(c) axes and every arc center
c = −(d−1)ζmarked on its ring. Each window contains the family's whole branch-point ring. The docs page derives the arc parameterizationc(μ) = −8ζ(1+μ)^{−1/8} − μ, μ = γt/(1−t)— the gamma trick, photographed — and cross-links the Monodromy Loom (same mathematics from solution space).Notes
fractions.Fraction; pixel coordinates snapped to rationals); no randomness anywhere, so the tracked data is fully deterministic.bertini.recording(False): millions of throwaway raw-tracker solves, deliberately below the records seam.tools/refresh_doc_artifacts.pywith a SLOW note — this is the heaviest doc artifact.sphinx -b dummy, no new warnings). No ADR: docs-only gallery addition; the load-bearing choices are documented in the generator's docstring and the page itself.🤖 Generated with Claude Code