Public evidence
Machine-checked verification for calculation version 1.6.3 · tier internally_reviewed
CVP status
CVP VERIFIED · CVP protocol 1.0.0-proposed · Engineering assurance · Z↔Γ + match-steps + auto-match + stub + Touchstone + f-sweep + Z(f) + S21/K + stab-geom + chart-xy + gain-geom + noise-geom
CVP covers z→Γ, Γ→z, invalid-domain, z↔Γ round-trip, looking-in match-steps (series-X/R, lumped L·C, tline, open/short stub, transformer, parallel-RLC), auto-match (L-section, λ/4, single-stub), Touchstone 1.x S11 (MA/RI/DB), numerical f-sweep of looking-in (±span around f0), Custom Z(f) linear interpolation of normalized z at the query frequency (frozen load), numerical 2-port S21 + Rollett K/μ (last .s2p row; same model as Analyze S21/K readout), Gonzalez load/source stability-circle centre+radius and origin/inside stable-side flags in the Γ-plane (same model as Analyze Stability readout; includes K≈1, |Δ|≈1, and den=0 sentinels), canonical Γ→chart XY at the default SVG viewport (size=520, pad=28, y-up in Γ / y-down in SVG; same affine map as UI toPx, including f-sweep locus points), default-viewport SVG load/Zin marker cx/cy from buildSmithChartSvg at size=520 (and the same map at size=640), unilateral GTU source/load constant-gain circles at absolute −1/−3 dB (same model as Analyze Gain −1/−3 dB readout; |S11|=1 omits Gs), and constant-NF circles at NFmin+0.5 / NFmin+1 dB from the last Touchstone 1.x noise row (NFmin, Γopt mag/ang, rn; same model as Analyze NF +0.5/+1 dB readout; rn=0 and missing noise omit circles). A Node URL-query contract (ohm↔norm conversion that preserves Z, visible R/X beating hidden z_r/z_i, illegal smith_state parse, and Basic/Match/Analyze tier) is exercised as properties; it is not a live browser session. CVP-XCAL-01 compares this IUT to electrical.rf.gamma_z (Γ↔z), electrical.rf.vswr (|Γ|→VSWR/RL), and electrical.rf.lc_match on the overlapping real-RL L-section; complex-z auto-match vs LC Match is N/A because LC Match encodes real RL↔Z0 only. Verification revision 2026-09-07.p14 distinguishes this CVP batch from earlier 1.6.3 manifests and from the legacy 21-test Evidence JSON. Live CSS-resized SVG, grid/path d= markup, stability/gain/noise overlay drawing, drag/hover, history, bilateral MAG/MSG, 3 dB bandwidth, and interactive workspace features are outside this verification scope.
Legacy regression suite: 21/21 calculation tests · Production surface contract 10/10
Three distinct suites: educational examples on the calculator page, this legacy regression count (21/21 golden + boundary), and the CVP verification suite below. They are not the same number. Property, metamorphic, round-trip, O3, and interface statuses/checks are recorded in the CVP Manifest . SKIP (0/0) means that suite has no published vectors.
Protocol checks from the Evidence Manifest. Distinct from the legacy regression suite above. Verification revision 2026-10-28.metric-scope-v3. Calculation version 1.6.3. CVP 1.0.0-proposed. 43/43 golden · 5/5 CVP boundary · 3/3 invalid · 53/53 property · 5/5 metamorphic · 4/4 round-trip · 12/12 O3 · 3/3 cross-interface · 3/3 cross-calculator · 10/10 CVP contract. digest e16e9c6d4139.
Fingerprint sha256:7ace186bc38978d6caea5c8694605a6e778f40e10ebfd9d79331688a07e99ef5
Input schema sha256:a05ef25ca7cf35bc01f6722b882e746a008756fc0dd54724052d34e5ddac35b7 · Output schema sha256:5736600d32f028e3ae48b600659781b97c032b017db38f2bb489284177b06234 · Build-time SHA-256 digests prove published artifacts agree with each other (release_integrity.scope=workspace). Production attestation GETs canonical URLs twice (public/cached view and origin), including Capability HTML, via npm run attest:production --write
(published at /.well-known/calculatorx-production-attestation.json) and fails if the public Capability page advertises a different calculation_version. This is artifact consistency, not an independent immutable release log.
Engine identity and declared limitations.
Standards and references supporting this tool specification. Methods are listed separately — they are algorithms, not bibliographic sources.
Aggregate outcomes, scope and artifact commitments are public. Complete test inputs, expected values, execution records and oracle infrastructure are retained privately. This summary does not independently reproduce the full verification run.