Target-byte 0x35 tri-state refinement

For cases deferred by V50, comparison with the next registered target byte 0x35 should again produce exact ACCEPT, REJECT or a much smaller DEFER set.

Claim statusNO END-TO-END MINING ADVANTAGE DEMONSTRATED

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

Date2026-08-23

HardwareClassical exact algebraic verifier

Run scopeThree canonical variants

ReproducibilitySelect only V50 DEFER states, compare the next digest byte with 0x35 using unsigned order and verify ACCEPT/REJECT/DEFER against full PoW in both cohorts.

01 / Question & hypothesis

What was tested?

For cases deferred by V50, comparison with the next registered target byte 0x35 should again produce exact ACCEPT, REJECT or a much smaller DEFER set.

02 / Scientific basis

Why the test is meaningful

Lexicographic comparison is recursively composable. Conditioning on prefix equality makes the next byte sufficient unless it also equals the target byte; this preserves decision safety without assuming anything about later digest bits.

given V50=DEFER, compare byte b with 0x35b<0x35⇒ACCEPT; b>0x35⇒REJECT; b=0x35⇒DEFERP(DEFER next)=P(b=0x35 | prior prefix tie)

FIGURE / RESULT READINGActions among records already deferred by the preceding boundary.

03 / Method

How it was tested

Generate two 20,000-record cohorts conditioned on valid V50 deferral, compare the next byte, record actions and verify every action against the complete target predicate.

04 / Observed result

What happened

Defer rate A / B0.330% / 0.365%

Reject rate A / B78.61% / 78.77%

Safety errors0

Replicate A recorded 4,212 accepts, 15,722 rejects and 66 defers; replicate B recorded 4,174, 15,753 and 73. Defer rates were 0.330% and 0.365%. Prior-condition, action-safety and semantic errors were zero; incremental work was 152 units.

05 / Validation

Exactness and statistical controls

Conditioning integrity was checked before evaluation, preventing contamination by non-deferred V50 states. Both replicates audited full PoW truth after tri-state classification.

06 / Interpretation

What the result means

V51 safely resolves more prefix ties at one byte boundary. Its high reject fraction is a property of the conditioned comparison, not evidence that more valid hashes are found.

Limitations

The target byte is fixed to 0x35 for this experiment.
Conditioned action rates are not unconditional probabilities.
No hardware advantage was tested.

07 / Reproduction

Evidence trail

Select only V50 DEFER states, compare the next digest byte with 0x35 using unsigned order and verify ACCEPT/REJECT/DEFER against full PoW in both cohorts.

Canonical variants

SUBENGINE-V51ASUBENGINE-V51BSUBENGINE-V51C

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