Skip to content

P104 methods: {0}/V re-basis, absorption, range-G pre-loads, signed-row Rado/LU helpers, O+(8,2) blocks, t = −1 callers (no κ) - #7558

Open
DreamingOfClouds wants to merge 3 commits into
CrocSwap:mainfrom
DreamingOfClouds:research/p104-methods
Open

DreamingOfClouds wants to merge 3 commits into
CrocSwap:mainfrom
DreamingOfClouds:research/p104-methods

Conversation

@DreamingOfClouds

Copy link
Copy Markdown
Contributor

Package research/p104-methods writes down six construction methods from our P104 work. Each method comes with its scope and a standard-library exact checker. The package makes no κ claim and ships no new word. It builds on #7542 (helper rung merges), #7555 (structure laws) and #7557 (t = −1 laws), and does not repeat them.

Tags in the README: [P] proved in PROOF.md, [C] exact run in this package, [Q] lane number quoted but not reproduced here, open.

  1. {0}/V re-basis (rebasis.py). This is an exact test for deleting a paid helper rung while every row stays fixed. It has two parts, both proved [P]:

    • {0} freedom: registers leaving {0} admit exactly the unitriangular mixing.
    • V freedom: V-arrival rows only need to lie in W_V.

    Combined with local slots, the test found 761@1053, the 761 half of Conditional κ = .0036475398277 (+0.0126% over #7481; +0.0243% over #7334): two helper rung merges #7542. The checker rebuilds that merge from Clear1nsight's Conditional κ = 0.00364665195644: fuse helper 319’s paid lifts #7334 word (vendored with the same SHA256 as in Conditional κ = .0036475398277 (+0.0126% over #7481; +0.0243% over #7334): two helper rung merges #7542) and replays it exactly. All 1,569 rows and end charts match, and 761 goes from 0→5→7 to 0→7. It also checks the {0} and V facts separately. Seven controls are rejected.

  2. Absorption (absorption.py). A helper is settled against a register's outstanding junk instead of against Y: it lumps −λ·y at {0}, does H += Σd·X_r(K), then H −= Σd·X_r^f at V.

  3. Range-G pre-loads (preload.py). A span law shows that range-G y content is not excluded at any chart [P]. The owned gauge G·(x/2 + (11/2)·y) gives Y_t = x_t with no zero-Y park, exactly on all 104 columns of the PR968 Gram [C]. The charge half (carrier tokens) is open.

  4. Signed-row helpers, Rado/LU (rado_lu.py). Group feasibility is Rado's transversal condition, and it is realised by LU into elementary ADDs.

    • One helper carrying a free signed row fixes a deficit of 2 where a pure e_k helper fails [C].
    • Agrees with brute force on 118 seeded instances [C].
    • The lane's full fan-out build (408.68 vs 1,692.1 D0/call) is quoted, not reproduced [Q].
  5. O+(8,2) quotient (quotient.py). All [C]:

    • The vendored port Gram equals the PR968 Gram up to a verified permutation and ±1 switching.
    • The ports are 104 of the 120 nonsingular points of the quadric.
    • Two-graph sign law: product −1 exactly on the 704 c-sum triples, and one wrong cell per partner pair.
    • 282 twelve-port 4-spaces, plus one packing of 8 port-disjoint blocks.
    • The c-trick identity, and 33 disjoint c-sum triples. Their maximality is not claimed.
  6. t = −1 callers (callers.py).

Not reproduced here, and listed in the README: the general re-basis scan and its bilinear extension, a frame-legal E8 absorption layer, the full fan-out and U-junk builds, a HOST(−1,1) native, and Lemma A.

  • Validator: python3.12 -B scripts/research_package.py --package research/p104-methods → PASS_REGISTERED_RESEARCH_PACKAGE, in about 90 s.
  • The branch carries the two Pin the historical selected replay to its inspected commit #7257 CI-prerequisite commits by Clear1nsight, cherry-picked with authorship kept.
  • No human has reviewed the package.

— DreamingOfClouds with Claude Opus 5.5 (Anthropic), via Claude Code

🤖 Generated with Claude Code

Clear1nsight and others added 3 commits November 13, 2026 18:36
This is the exact documentation prerequisite retained by CrocSwap#7156. It points the onboarding check to the selected CrocSwap#7000/CrocSwap#7001 review. Credit to Rohan Arun and the source authors.
The PR CrocSwap#6292 rejection census and package plan are statements about 2485805. Load its exact sources and planner in a private sparse checkout, keeping all source hashes, 19 expected rejections, nine package selections and the empty-plan control. Subsequent edits to the live Makefile and CI guide no longer invalidate this historical check.

The original audit is by unpaidmoon, building on Dennis’s historical replay gate. The pinned repository view follows the precedent of Matarener’s PR CrocSwap#5434. Repair prepared with substantial OpenAI Codex assistance.
…ed-row Rado/LU helpers, O+(8,2) blocks, t = -1 callers (no kappa)

Six construction methods behind our P104 work, each with its scope and a
standard-library exact checker: the helper-761 rung merge of CrocSwap#7542 rebuilt
from CrocSwap#7334's word by the {0}/V re-basis test; the absorption identity, the
last-outstanding-cell rule and sigma-anchor D = 0 (frame-free A3/A4 algebra);
Y_t = x_t with J_t = (11/2)G_t on all 104 columns; the Rado transversal group
rule with LU into elementary ADDs; the O+(8,2) quotient of the ports (sign law,
12-port blocks, c-sum triples); the MONOHUB joint caller and four callers at
the 20-odd-op floor. Credit Clear1nsight for CrocSwap#7334 and Rohan Arun for CrocSwap#7113.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.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