Full T2₅ low-bit consumer

A local simplification is meaningful only if it survives composition into the complete T2₅ low-four-bit consumer, including Σ0(a5), Majority and inner carry.

Claim statusNO END-TO-END MINING ADVANTAGE DEMONSTRATED

Diagnostic work, exact algebra and local capabilities are not treated as end-to-end mining advantage.

Date2026-08-14

HardwareClassical exact algebraic verifier

Run scopeThree canonical variants

ReproducibilityRebuild all four component masks, verify their exact sum, preserve the original/fresh split and compare every predefined difference sign without dropping unfavorable components.

01 / Question & hypothesis

What was tested?

A local simplification is meaningful only if it survives composition into the complete T2₅ low-four-bit consumer, including Σ0(a5), Majority and inner carry.

02 / Scientific basis

Why the test is meaningful

T2₅ combines rotated state structure, a nonlinear Majority term and modular carry. Testing each part in isolation risks omitted-variable conclusions; the complete consumer is the correct connected interface.

T2₅ = Σ0(a5) + Maj(a5,b5,c5) (mod 2³²)low4(T2₅)=low4(Σ0+Maj+inner carry)replication requires the same survivor-minus-non-survivor sign in original and fresh cohorts

FIGURE / RESULT READINGVisual reading of the published metrics and gates for EXP-036; it summarizes the registered result, not a mining advantage.

03 / Method

How it was tested

Construct exact 16-lane masks for Σ0(a5).low4, Maj5.low4, inner carry 1–3 and T2₅.low4; compare nonlinear support on the fixed original split and repeat on the fresh cohort.

04 / Observed result

What happened

Semantic errors0

Replicated signs2 / 4

T2₅ simplificationNO

All semantic checks passed. Difference signs replicated for Majority and full T2₅, but not for Σ0 or inner carry. T2₅ nonlinear-support difference was +0.011 in the original cohort and +0.125 fresh, meaning survivors were not simpler under this metric.

05 / Validation

Exactness and statistical controls

The complete consumer was checked against direct arithmetic on 1,361 original and 300 fresh records. Reporting all four components makes the two non-replicating signs visible; no composite advantage was inferred.

06 / Interpretation

What the result means

The isolated attenuation does not survive as lower complexity in the full consumer. This is a negative operational direction even though the exact decomposition itself is validated.

Limitations

Nonlinear monomial count is not a timing benchmark.
The fixed group contrast cannot classify future candidates.
Only low-four-bit semantics are covered.

07 / Reproduction

Evidence trail

Rebuild all four component masks, verify their exact sum, preserve the original/fresh split and compare every predefined difference sign without dropping unfavorable components.

Canonical variants

SUBENGINE-V36ASUBENGINE-V36BSUBENGINE-V36C

Source: internally audited canonical reports. Local filesystem structure, private headers and operational identifiers are excluded from publication.