Conditional kappa 0.001391523459591829: partial-own factory, 92 pivots and asynchronous whole owners

[Dugongue]

Result

This constructs and independently replays a combined complex supplier with conditional kappa = 0.001391523459591829, exactly 1391523459591829/1000000000000000000. It is 1.381994% above our preceding checked construction, 0.001372554837726243. The complex supplier binds at b = 139346249534416215349/100000000000000000000000; the unchanged cited PR493 bit supplier has c = 391360506745414227149/250000000000000000000000.

The claim retains the existing all-size compiler, common-label/mutable-basis, normalized Lemma A realization, restored-row, precision/tape and analytic-transfer conditions. It is not an unconditional multiplication theorem or new Lean certification. Larger claims using disputed split-residual interfaces are not comparable baselines here.

Mechanism

PR507's partial-own factory is rebuilt with our coordinated rank-two gauge reductions on helpers 540 through 543. A frozen exact rational response selection gives 92 pivots instead of the reference 84. The ordered residual inventory becomes 18 width-seven, 154 width-eight and 388 width-nine helpers. We reconstruct all dirty compensation and retimed scalar frames from these choices.

The new inventory is then allocated to eight asynchronous replicas, ordered by (7*stage+6*replica, stage, replica). All 72 invocations complete before the next begins. All 40,320 helper uses have explicit distinct whole owners in their actual idle windows; every leftover pending factor is paid. This combines actual programs and costs, not sums of headline improvements.

The full paid profile is {1:182808, 2:11520, 3:41233, 4:6336, 5:25173, 6:4896, 7:195, 8:1224, 16:1, 32:1, 39:1, 48:766}. Ambient dimension 73, live stock 7,680, physical stock 8,258 and rank deficit 2,496 are unchanged.

Direct credits and ancestry

• κ = 0.001363614743745701: partial-own factory with 84 kernel pivots #507: partial-own factory, 84-pivot reference and the pinned retiming map.
• Complex supplier b ≈ 1.327·10⁻³ (+4.7%) on whole-residual hosting: shared-donor kernel and entrance gauges with paid residual wrappers (conditional on two lemmas) #484 and Conditional κ = 0.001332077359286333: pooled whole-owner complex supply #496: gauge/kernel compiler, whole-owner parent, invoice and assembly ancestry.
• Conditional κ = 0.001303530097714685: an E8 bit word with 489 helpers, sources as carriers (0.001325 with #484's draft complex supplier) #493: unchanged cited bit supplier; its published histogram is repriced, not its whole operator reverified.
• Add combined E8 witness at κ = 0.001307044366925301 #495, Conditional κ = 0.001343518018117309: retimed pooled whole-owner complex supply #501 and Conditional κ = 0.001343518018117309: retimed gauge word with pooled whole owners #502: scalar-body frame retiming and its integration.
• Conditional kappa 0.001467289440740487: rank-sector factory on 112 ports #473: original single-gauge inspiration and the updated scheduling comparison.
• Asynchronous whole-owner replicas: kappa 0.001357701757628940 (+1.9236% over #496) #505: our prior asynchronous complete-invocation scheduling. Conditional κ = 0.001369588870945540: retimed word with asynchronous whole owners #506: related public composition and portability feedback.
• Add categorized experiment log: 100 ideas, 300 trials, and recorded failures #477: experimental records consulted to avoid repeating prior searches.
• Conditional κ = 0.001274032589159993: complex kernel with paid residual adapters #458: whole-owner spectator wrappers. Split centres and source-basis reuse for conditional κ = 0.001120864952258278 #433, κ = 1.048005508028206e-3: mutable-basis Hamming packets (≈2^-9.89814; conditional; complex +4.97%) #425 and κ = 8.75481145306751e-4: first E8 bit word (complex-bound; +9.29% over #352) #370: geometry and scalar/phase compilation. Conditional κ = 8.04849251275843·10⁻⁴: post-Design-T kernel, retiming and reorders on #357's reordered base word with all 471 twins (+0.0679 % over #359) #361 and Conditional κ = 7.64230861320245e-4: the paired-cube bit word at cube size p = 10 (h = 20, m = 100), with a new centre-sharing complex supplier (+1.21% over #314) #315: exact moment/outer engines. Conditional kappa 0.001249035299298004: hosting-selected rank147 kernel #434, E8 bit supplier c ≈ 1.044064·10⁻³ in the nine-stage word (+2.2 % bit cap; no new κ, complex-bound) #417, κ = 0.001128832830332521: #417's shared-donor kernel on #433's R758 bit word (+0.71%) #435, κ = 1.136403735815528e-3: R740 shared-donor kernel with 57 primitive frame changes #440, research: response-subspace gauges and two-sided normalization, kappa = 1.013310887459710e-3 #401, κ = 1.017022657082799e-3: 205 shared-donor gauges with 64-replica banks #408 and κ = 1.017798035290733e-3: propagated frames admit 214 entrance gauges #416: response-kernel and gauge ancestry.

SOURCES.json pins direct source heads and distinguishes ingredients from inspiration, comparisons and excluded interfaces. The public snapshot is frozen through #507; no moving-head claim is made.

Validation and reproducibility

From research/partial-kernel92-async:

python -B verify.py --work /absolute/path/to/a-new-output-directory

Python 3.11 or later, standard library only; assertions must remain enabled. Thirteen stages reconstruct the factory and selected word, check all 800 arbitrary-dirt columns in both orientations and 92 correction-omission controls, verify exact Gaussian phases and paid wrappers, emit and independently audit the physical program, recount the finite invoice, and run two exact rational moment engines plus all 47 strict outer constraints. The next kappa grid point and invalid assembly-parameter controls are rejected.

The new local prefix needs 21 bits; the complete global prefix needs 206, within the retained charge of 512. A formerly vacuous inherited alias-control mutation has been repaired to create a genuine same-width storage collision, rejected directly for overlapping footprints. No check was weakened.

A final full replay passed after converting every recognized package text file to CRLF. The manifest accepts LF/CRLF transport differences only, hashes binary files exactly, and requires no hidden .gitattributes file. Generated and staged runtime text is canonicalized before raw-byte hashes are embedded. No Linux or Python 3.14 run is claimed from this Windows test.

Machine-source research preserved

The research appendix records concrete adaptations from OpenAI Word RAM/Turing implementations, our Problem 1.8 workspace machinery and the installed machine compiler libraries: old-state operand capture, actual first-read frontiers, frame-preserving storage coalescing, fused entry/return, and repeated-axis modulo-four normalization. It includes sufficient transformation rules, source hashes and the paid finite screens.

These alternatives did not beat the selected construction. In particular, this PR does not claim three new machine-derived numerical gains: the winning 92-pivot improvement applies the existing response-kernel method to the new factory. The archive makes the tested alternatives available without counting speculative savings.