Skip to content

Prove two rank-q maps in one gap can fill every helped core block (no κ) - #273

Draft
unpaidmoon wants to merge 2 commits into
cursor/upstream-e545235-base-03aafrom
cursor/two-rank-q-same-gap-cb2d
Draft

unpaidmoon wants to merge 2 commits into
cursor/upstream-e545235-base-03aafrom
cursor/two-rank-q-same-gap-cb2d

Conversation

@unpaidmoon

Copy link
Copy Markdown
Owner

Change and mathematical scope

Proof, not a κ change. On the paired-star one-gap schedule, two matrices of rank at most q in the same internal gap, each with P_s Z P_f = 0, can make every helped core block equal I_{2q}.

The first step from open PR CrocSwap#7053 does not. Two copies of that rank-q witness on complementary leaf halves, or split onto complementary core directions, do not reach rank 2q while obeying the square equation. With B12 = B21 = 0 both outside blocks have rank at most q for any number of internal matrices, because each factors through a projector image of dimension q. That failed step is not unrestricted nonexistence.

The positive pair uses B12 = B21 = I_q, over any field of characteristic two. Each factor has rank q and kills the image of the first half. For both half orders the three blocks of the sum are I_{2q}. A helped leaf can meet A_j + B_j = I_d with external block zero. The uniform lift of CrocSwap#7053 to external rank q(g - 1) therefore does not apply to every placement. The dropped-leaf bound q(g - 3) of CrocSwap#7025 still applies.

With those identity blocks the parity count is sharp for the rank inequalities:

  • odd g, even stage: at least q(g - 1), for any internal matrices, because only one even leaf is helped;
  • odd g, odd stage: at least q(g - 3), and a gap tuple meets it;
  • even g: at least q(g - 2) at every stage, and a gap tuple meets it.

At g = 31, q = 26 the odd-g numbers are 780 and 728. The tuples are not words.

The selected frontier stays κ = 363005817/100000000000. This does not construct a word, does not retake the one-map theorem, and does not replace CrocSwap#7025, CrocSwap#7005, CrocSwap#7050, or CrocSwap#7126.

Task and overlap

Validation

Evidence class: proof over any characteristic-2 field, for the stated halves and matrices. The q = 1 census is an exhaustive check of every rank-at-most-1 matrix on the eight coordinates (all scalar (B12, B21), both half orders). It is not a census of words. The gap tuples are inequality witnesses, not words.

python3 -B scripts/research_package.py --package research/two-rank-q-same-gap

Python 3.12, no dependencies. Both registered commands (python -B and python -O -B) print:

PASS_TWO_RANK_Q_SAME_GAP blocks=I_2q singular_next_cap=q+rank(B21) odd_stage_external=728 even_stage_external=780 uniform_q(g-1)_lift=does_not kappa=363005817/100000000000

scripts/check_research_commit.py --package research/two-rank-q-same-gap --check-source-history --check-modes reports PASS_RESEARCH_COMMIT_BYTES for the package at 66405802f1eaac23d7b19759db1fc5d68b409397, with source history ancestor e54523517d0f03e58352e647c403a5af3e525a57.

Failure controls: clearing one column of the first map makes the next block leave I_2, and one map alone is not the next identity.

Credit

The one-gap schedule and q(g - 3) are open PR CrocSwap#7025. The three blocks, the one-map cap, the rank-q witness, and this question are open PR CrocSwap#7053. The paired-star contract and the commuting map-count are from PR CrocSwap#5747 (Ali Romman), including PR CrocSwap#5673, PR CrocSwap#5631, and PR CrocSwap#5617. Cinna Davis supplied the controller and simplex lineage those packages retain. Open PR CrocSwap#7005 is the consecutive-halves bound. Open PR CrocSwap#7126 is the every-stage factor.

Prepared by unpaidmoon with substantial Cursor Grok assistance. Apache-2.0.

Open next questions

  1. At an odd stage, can external gaps of total rank q(g - 3), together with these two maps, meet A_j + B_j = I_d on every leaf at g = 31, q = 26? First step: stage t = 1, place the maps, set the three helped external blocks to zero, and test the cover on the remaining leaves. The monomial transport is not rebuilt here, so a passing rank tuple would still not be a word. This is not Prove one rank-q internal map raises the paired-star external bound to q(g-1) CrocSwap/integer-mult-bounds#7053 question 2 and not the every-stage question.
  2. For 0 < rank(B21) < q, can two rank-at-most-q maps attain next-block rank q + rank(B21) and own-block rank 2q together? First step: at q = 4, take B21 of rank 1 and B12 = I_4. The identity witness uses B21 = I_q, so its functionals need not kill the smaller image.

Workflow feedback (optional)

CrocSwap#1200

Open in Web Open in Cursor 

cursoragent and others added 2 commits November 11, 2026 10:43
The complementary copies of the rank-q witness do not reach rank 2q.
With invertible off-diagonal blocks an explicit pair makes each helped
block the identity, so the uniform q(g-1) lift does not survive.
No kappa change.

Co-authored-by: unpaidmoon <unpaidmoon@users.noreply.github.com>
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