Home › Evidence › Records › eng-0227
⚠ This notebook has stopped taking notes: newest record is 23 days old (eng-0280), against an expected cadence of 14 days.

eng-0227

window recorded
citable URL: https://halobench.com/records/eng-0227/ — this address never moves; the anchor /records/#eng-0227 keeps resolving
254.4 Whmean 163.08 W − 10.1 W idle floor (idle-ho003-patched-throughput) = 153 W attributable
window 2026-08-22T05:37:52Z → 2026-08-22T07:11:28Z · provenance recorded
method wall-meter · aihydra energy sensor via Home Assistant (counter-difference on sensor.hardware_ai_hydra_energy, raw ~10s samples queried by hborchestrator via Nabu Casa API, idle baseline 10.1 W) · no contention
covers run-0562 run-0563 run-0564 run-0565 run-0566
PATCHED q8_0 throughput ladder aggregate (run-0562..0566). Counter 31.682629 -> 31.952827 kWh = 270.20 Wh total, 15.76 Wh idle -> 254.44 Wh active. Build ce7689f, q8_0/q8_0 KV. Slightly MORE efficient than the f16 control ladder (254.44 vs 272.84 Wh). Wh-total only (energy join yaml patched-throughput).

Wh, never joules — read as a cumulative-counter difference across the window; delta over the idle baseline is the only figure that means anything