admin 541129542d Estimator: the HP figures published, raise only (R-1009 = A) — one verdict stricter ...

sizing.json + sizing-data.json rebuilt (23 figures up, none down; the checksum chain matches,
build_sizing_web --check OK). The household set without Immich on 8 GB: fits well -> fits, but
tight (2,731 x 1.5 = 4,096.5 > 4,096). Presets keep their verdicts. SIZING.md section 5 table,
bullets and memory table from the rule, plus the dated line; the per-app table re-rendered.
Test expectations moved with the rule. Lede, meta and method note name both machines. Data ?v=4.

Claude-Session: https://claude.ai/code/session_012qRErfCoiTkvDK9N5XHbzb

2026-10-14 15:27:56 +02:00

..

evidence

Estimator: the HP figures published, raise only (R-1009 = A) — one verdict stricter

2026-10-14 15:27:56 +02:00

tools

Sizing: the HP check can only RAISE a figure (R-1009, operator ruling A) — build_sizing.py

2026-10-14 15:24:07 +02:00

METHOD.md

Audit: what a Felhom box needs — floor, 64 apps' memory and CPU, power (N100 bench); R-993..R-995; R-231 due check read green

2026-10-13 13:23:51 +02:00

sizing.json

Estimator: the HP figures published, raise only (R-1009 = A) — one verdict stricter

2026-10-14 15:27:56 +02:00

SIZING.md

Estimator: the HP figures published, raise only (R-1009 = A) — one verdict stricter

2026-10-14 15:27:56 +02:00