Skip to content

Prove integral four-odd-c labels miss length-6 support (no κ) - #272

Draft
unpaidmoon wants to merge 1 commit into
cursor/crocswap-main-e545235-300efrom
cursor/four-odd-c-length6-absence-300e
Draft

unpaidmoon wants to merge 1 commit into
cursor/crocswap-main-e545235-300efrom
cursor/four-odd-c-length6-absence-300e

Conversation

@unpaidmoon

Copy link
Copy Markdown
Owner

Change and mathematical scope

Proof for integral labels. No new κ and no word.

In the PR2085/2112 register model, no six nonzero integral rank-one labels sum to 4I, meet target support, and have at least four labels of odd c. For every pair of distinct old projectors, the length-6 equation

R + S + T + U = 4I - E - F

therefore has no integral solution that meets support and has at least four odd-c labels. Odd c means the integer coordinate of P is odd. A rational label with a non-integral coordinate is outside the theorem.

The PR CrocSwap#7100 witness A, B, P, Q, 2A, 2B still meets support, with odd-c count exactly two. Count four is the next even value. This package does not classify the count-two solutions.

The pair step is an exhaustive census of a finite set: for nine goals, the discriminant is negative for |b| ≥ 5, and b ∈ {-4,...,4} yields only the four projector pairs and the two even splittings of 2I. The case split on the two even labels is then finite. A max-norm-4 scan is a control, not that census.

Selected κ stays 363005817/100000000000. The eight-call minimum is unchanged.

Task and overlap

Base commit e54523517d0f03e58352e647c403a5af3e525a57 (canonical CrocSwap main, re-fetched at submission and unchanged). Fork main is a different history, so this draft uses cursor/crocswap-main-e545235-300e, which is that same commit, as its base. The upstream target is CrocSwap main.

Head 2cea1378258e93a140dd71a7027730a3df2c8cb0. Package research/four-odd-c-length6-absence.

Task card bit-new-register-model, scope changed. This is the integral half of open question 1 in PR CrocSwap#7100.

Overlap checked against open upstream PRs and fork branch names at submission. A GitHub issue search for the odd-c length-6 question returned only PR CrocSwap#7100. No fork branch already carries a four-odd-c follow-up.

Validation

Python 3.12, no third-party packages.

python3 -B scripts/research_package.py --package research/four-odd-c-length6-absence

Registered status: PASS_REGISTERED_RESEARCH_PACKAGE with outcome completed. Both the plain replay and the python3 -O -B replay print PASS four-odd-c-length6-absence checks. Checks raise, so -O does not strip them.

scripts/check_research_commit.py --package research/four-odd-c-length6-absence --check-source-history --check-modes passed at 2cea1378258e93a140dd71a7027730a3df2c8cb0.

The checker confirms the polynomial elimination, the |b| ≥ 5 discriminant bound, the finite pair census, the parity types, every EE/OE budget at odd-c count 4 or 6, the count-two boundary witness, and the max-norm-4 control.

Not run: a classification of count-two solutions, a rational-scale search, a native or supplier replay, a proof assistant, and scripts/research_selected_baseline.py. The selected record is provenance only.

Credit

The register model is Cinna Davis's PR2085/2112 formal model. The determinant, the subset-sum necessity, and the call-versus-target inequality are the PR2910 package as recorded in apoursh's two-signed-frame floor, PR CrocSwap#6190. The four-call support proof is PR CrocSwap#6952. The one-old length-5 proof is PR CrocSwap#7083. The length-6 support witness whose open question this note answers for integral labels is PR CrocSwap#7100.

Prepared by unpaidmoon with substantial Cursor Grok assistance.

Open next questions

Does any pair of nonzero rational rank-one frames, not both integral, sum to 4I-H for a target H? The pair census is integral. A rational complement pair would allow a size-4 carrier. First step: one rational solution of X+Y = 4I-(A+P), or a discriminant obstruction over Q. Blocker: this package enumerates integer b only.

Does R+S+T+U = 4I-E-F have a non-integral rational support solution on which the inherited lattice character detects at least four calls? The integral odd-c count is empty, so a length-6 word would have to leave Z^3. First step: one rational support solution with a denominator greater than 1, then the subset scan and the character on its lattice. Blocker: the odd-c test used here requires integer coordinates.

Workflow feedback (optional)

CrocSwap#1200

Open in Web Open in Cursor 

The length-6 equation still meets support at odd-c count 2. No integral solution reaches count 4 or 6, so that count does not give a shorter word.

Co-authored-by: unpaidmoon <unpaidmoon@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants