B72 target-boundary tri-state

What was tested?

After the B72 prefix, exact lexicographic comparison can classify many states as final ACCEPT or REJECT while safely deferring only prefix ties to the remaining suffix.

Why the test is meaningful

For two equal-length bit strings compared lexicographically, the first differing bit decides the result. If the known prefix is smaller than the target, later bits cannot reverse ACCEPT; if larger, they cannot reverse REJECT; equality alone requires DEFER.

How it was tested

Evaluate the registered 72-bit boundary on two independent 20,000-record B72-conditioned cohorts, emit one of three actions and check action safety plus the final full-PoW truth for every deferred case.

What happened

Replicate A produced 19,782 exact rejects, 139 exact accepts and 79 defers; replicate B produced 19,786, 143 and 71. Defer rates were 0.395% and 0.355%. Action-safety and B72-condition errors were zero; mean incremental boundary work was 158 units.

Exactness and statistical controls

Every ACCEPT and REJECT was checked as irreversible under all suffixes, while every DEFER was evaluated against full PoW. Two replicates produced closely aligned action rates.

What the result means

The tri-state boundary is an exact control primitive that prevents unsafe early decisions. It is not itself a submotor and does not establish a cheaper path to valid hashes.

Limitations

• Action frequencies are conditional on B72.
• The 158-unit accounting is representation-specific.
• Deferred cases still require suffix evaluation.

Evidence trail

Perform unsigned lexicographic comparison at the frozen boundary, record all three actions and verify each against the full 256-bit target in two independent cohorts.

Canonical variants

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