QMCE Sesja 14: An Unresolved Elo Effect for a Chess Meta-Engine Overlay under Author-Declared Thresholds, with an Analytical Bound Against the Queueing-Overload Explanation and a Measured Depth-per-Clock-Second Deficit in One Run

QMCE (Quantum Meta-Chess Engine) was integrated as a Path A overlay on Amelia Chess via PyO3 0.20 in-process embedding (sessions 10–14, 2026-04-26..05-17). Nine match estimates across eight configurations give centers in -12.2 to +9.9 Elo, every 95% interval bracketing zero as reported (the eight pre-cloud ones are legacy game-level approximations, §4.1.3); three same-binary long-TC probes pool to -0.0 ± ~19 Elo (N=634). A queueing-overload explanation (near-saturation PyO3 GIL fetch) is bounded well below its own overload band in every tabulated production scenario: the fetch duty cycle peaks at 0.286 in production (0.036 warm), against an H0 boundary of 0.5 and overload at 0.9. That is a bound, not a test: the model's arithmetic makes the declared H1 bands unreachable (§3.2). A depth-per-clock-second analysis of one run crossed its author-declared cutoff (sample mean ratio 0.874 < 0.95), but it is unpaired and has no uncertainty interval, so no inferential support is claimed: the summaries show no deficit at the median or p95 and localize it nowhere. A targeted 4.02× microbenchmark speedup on the supposed hot path produced no detectable Elo change (-12.2 ± 32.9). The completed 1500-game cloud follow-up (putative replication; binaries never identified, §4.1.1) gives +3.0 ±12.3 Elo over 749 colour-swapped pairs; that interval contains both zero and the +5 Elo H1 boundary, so H1 is unresolved rather than rejected. Thresholds and the retain-neutral rule are author-declared and dated in version control, but the commits do not establish priority over the numbers they judge (§3.1). We present the full record, our cross-review's errors included, as a micro-scale case study consistent with, not evidence for, frontier-scale claims. Our claim is narrower: the compute we could afford could not decide whether the hypothesis was right; the cheap analytical calculation gave the firmest result here.