Repository navigation
Prove sharp 8H8 active-mode delivery cost with temporal reuse - #5745
Merged
Merged
Conversation
This was referenced Oct 30, 2026
Mister-Meeseeks
pushed a commit
that referenced
this pull request
Oct 31, 2026
On the fixed PR #5745 source, no three grade-below-8 response spaces meet the clean mode, while 41 dimension-1 frames do. The clean generation number satisfies 4 ≤ γ ≤ 41. The selected frontier is unchanged. Co-authored-by: unpaidmoon <unpaidmoon@users.noreply.github.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The released PR5720 repair grants a payload at frame280 but does not supply its transport. This result resolves the fixed source's complete eight-mode canonical delivery allowance: the sharp minimum is 8H8, even with arbitrary reuse of each payload at different times in its one frame. That equals the eight removed center jobs. The theorem covers any number of clean channels, and at most eight channels even if nuisance outputs may cancel jointly. It is a scoped paid-cost result, not a new kappa or general feedback barrier.
Scope, proof and exact contract, full evidence, temporal witnesses, complete emitted word, source pins, registered replay. Exact head: e82b64e. Base: refreshed canonical main df0137e, priority revision7. Selected conditional kappa .00310045 is unchanged. Canonical investigation.
The exact rational calculation covers 5,331 evaluated frame states, 3,963 projected endpoint columns and 5,715 distinct frame/column observations across time. Invertible coframed updates preserve their complete span; terminal full-frame saturation covers the higher-cost rank-nine case. No clean response occurs below rank eight, including arbitrary temporal reuse. Eight rank-eight witnesses span all modes. A rank argument extends the clean-channel bound to every at-most-eight-channel completion in the stated contract.
A legal completion charges all eight transports, uses coframed ADDs, and checks every one of the 632 original inputs, all 408 helper identities, every final frame and all owned zeros with the original native warp words retained. The eight new Gaussian maps have exact grade eight, including checked absolute phases. The complete paid profile equals the original: 2201H1+412H2+510H3+127H4+40H5+8H8. After proved-zero cleanup, this direct completion still adds 448 ADDs, sixteen CLEARs and eight live owned fields. The original center implementation dominates it. No compiler expansion of this word is warranted by these results.
CPython3.12 and standard library only. Registered normal and optimized replays pass with complete receipt/word equality. Controls cover useful cross-time nuisance cancellation, independent nuisance coordinates, a missing paid transport, double helper debit, a wrong injection sign and eight wrong native phases. Committed-byte checks, official export integrity and a fresh optimized/warning-strict read-only Git-free replay pass, with empty PATH and unchanged exported bytes/modes. Hosted status remains separate. Full source ancestry was not checked in the shallow checkout.
The lower bound uses PR5720's necessary scalar quotient and fixes canonical ground-to-frame delivery of original mode payloads. Nine or more jointly noisy channels, changing the payload's native frame/address contract, changing the source word or recomputing payloads from changed live data remain open. The positive word inherits the frozen source's existing native contracts; it does not re-execute every historical physical wrapper or supply caller, precision, stock, tape, compiler or outer admission. A different same-paid-profile word could have a different ordinary bill; only the emitted completion is shown dominated.
Credit flyingviper/PR5720, its original filtered-native packet, Cao Yifan, Cinna Davis, Dugongue and all retained mode/native/controller/caller authors. PR5704 supplies the unchanged compact phase engine and its classical stabilizer-simulation lineage. The archive is unchanged and its license retained. The original center construction and helper compensation remain predecessor work. Ali Romman with substantial OpenAI Codex assistance; no subagents; Apache-2.0.
Next: give a fully charged at-least-nine-channel noisy completion, or an exact changed native payload contract. Repeating a clean same-frame timing search is unnecessary for this pinned word. Fresh current-main, priority, PR and Discussion checks found no overlapping F9 delivery claim; unpublished work is outside that coverage.