Repository navigation
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Result and scope
Finite global order is impossible at the sharp exact-sum Gaussian radial optimum. Every injective Gaussian-pair map with exact sum, Euclidean error at most one and forward radial slack at most
1/sqrt(2)has exactly two possible actions on each sum-one shell of odd difference taxicab radius R: two cycles of length R, or one cycle of length 2R. These unbounded periods exclude every finite global iterate identity.Read the proof, exact receipt, exhaustive ring checker and scope record. The proof forces every optimal map into the two ring matchings by finite-square capacities; no matching or symmetry ansatz is assumed. It also gives a strictly positive necessary radial gap for each fixed identity F^m=id, without claiming the general gap is optimal.
For F^4=id alone, the sharp slack is
sqrt(19)-sqrt(13). The new lower argument uses two invariant squares and a ten-input ring whose only map orders are five and ten. It does not assume F^2=swap, phase covariance, wire-swap covariance or quarter-turn symmetry. The upper construction is CubeJerry's existing global square-to-swap map, which also satisfies F^4=id. The sharp fourth-power conclusion explicitly uses that pinned source theorem.No new kappa, physical provider, native tariff, full caller or selected-assumption discharge is supplied. The classification is on the sum-one fiber; larger blocks, extra state, changed error bounds and relaxed sum constraints are outside scope. All five selected assumptions remain. This completes the advance claim.
Evidence and reproduction
Normal and optimized CPython 3.12.16 replay passed on macOS arm64. The checker enumerates every matching in 33 complete rings, for odd R from 1 through 65. It finds exactly two per ring. Direct orbit traversal and a separate gcd calculation agree on all 66 maps. All 2244 exterior edges from the corresponding squares have cost strictly above
1/sqrt(2)and direct Gaussian error one.The receipt retains both complete ten-point maps and every exterior edge of the two critical squares, with exact beta comparisons. Sixteen prescribed finite orders check the necessary positive gap and incompatible ring orders. The credited attaining four-cycle has squared norms 13/19 and direct error checks. Collision, forbidden outward-edge and strict-boundary controls pass. A credited affine control checks 625 inputs through four steps and shows that F^4=id does not force F^2=swap within the exact-sum error-one class. The written argument proves the all-ring classification; finite checks corroborate it. A finite cycle alone is not a global upper construction.
Metadata, source history, committed bytes, modes, eight-file export and archive integrity pass. A read-only optimized replay without Git or a usable PATH passes and preserves all nine payload/manifest hashes and modes.
With CPython 3.12 and its standard library, from the repository root:
Expected markers:
Hosted CI did not certify this package. The plan job failed in the shared
test_current_ci_section_matches_selected_record: the base guide names frontier6030 while the selected record requiresmake frontier7000-verify. The log reports 760 tests, one failure and 24 skipped; package replay was skipped. This package changes no shared CI file. Local replay remains the scientific evidence. Full predecessor package replays, selected cold replay, Lean and Docker were not run. Independent human and formal-kernel review are unclaimed. The fourth-power upper map is not promised to use conjugation as its inverse. Other prescribed finite-order optima remain open.Sources and credit
Prepared by @aliromman with substantial OpenAI Codex assistance. Jerry Chen (@CubeJerry) supplied the half-grid framework, prior square-to-swap theorem, global upper construction and attaining cycle in PR #7570, pinned all-input proof. That proof was read; its full package replay was not run. Those upper results are retained, not newly claimed.
Ali Romman's PR #7696, pinned proof, supplies the global optimum, individual-ring matching and boundary geometry. The new result forces every optimal map into those choices and computes their periods. The checker adapts the earlier coordinate and integer-comparison methods; it is not blind review.
The proper-weakening control uses Ali Romman's minimum-error affine center family in PR #6509, pinned proof. That family is prior work.
Ali Romman's PR #7698, pinned proof, motivates the finite-order question through its fourth-power failure. Its provider and ports are not changed. Credit details retain Nick Petalas's exact-sum question and finite witness and the earlier Gaussian/native lineage. New work is Apache-2.0.
Base: canonical main
e54523517d0f03e58352e647c403a5af3e525a57, priority revision 8. Carduniform-transfer-audit, scope changed to this local precision/order implication. Live handoffs, refreshed PR/Discussion overlap and source-reference intake show no competing classification or fourth-power-only lower claim. PR #7572 concerns affine translation periods, a different condition. Closed overlap uses the complete snapshot and an updated-order delta with no later changes. Sparse local and unpublished work remain coverage limits.