Home Developers Evidence math.numerical.ode

Public evidence

math.numerical.ode

Machine-checked verification for calculation version 1.1.1 · tier engine_tested

CVP status

CVP VERIFIED · CVP protocol 1.0.0-proposed · Core assurance · all published methods

Scalar first-order IVP. Discovery /euler-method and /runge-kutta lock method. Not rk45, not systems, not PDE, not CAS.

Declared partition coverage
PASS · 7/7 declared partitions (exponential-range, shifted-time, scaled-results, numerical-range, euler, rk4, invalid-domain)
Versions
Calculation 1.1.1 · CVP protocol 1.0.0-proposed · Evidence 2026-10-18.ode-input-polynomial-surface-v3
Verification revision
2026-10-18.ode-input-polynomial-surface-v3 · 1/1 property · digest f6deeacd0aae

Verification

Legacy regression suite: 9/9 calculation tests · Production surface contract 5/5

Three distinct suites: educational examples on the calculator page, this legacy regression count (9/9 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.

    This calculator CURRENT · Public schema 1.1.1 matches · Semantic contract ✓ · Production attested · Public/cache ✓ · Origin ✓

CVP verification suite

Protocol checks from the Evidence Manifest. Distinct from the legacy regression suite above. Verification revision 2026-10-18.ode-input-polynomial-surface-v3. Calculation version 1.1.1. CVP 1.0.0-proposed. 141/141 golden · 2/2 CVP boundary · 17/17 invalid · 1/1 property · 12/12 O3 · 8/8 cross-interface · 1/1 cross-calculator · 5/5 CVP contract. digest f6deeacd0aae. Relative 1e-8 comparison of y and exact against listed fixed-step references; not a whole-domain accuracy guarantee or a user-tolerance acceptance.

  • Golden (oracle-backed) 141 / 141
  • CVP boundary 2 / 2
  • Invalid input 17 / 17
  • Property 1 / 1
  • O3 high-precision 12 / 12
  • Cross-interface 8 / 8
  • Cross-calculator 1 / 1
  • CVP contract 5 / 5
  • Interfaces PASS · UI (SSR) / REST / MCP
  • Interface claim ui-ssr is query-result HTML, not a live browser session. Error-path engine·REST·MCP 2/2 (status, code, calculation_version). SSR compared on URL-canonical requested calculations; empty query is idle (not an error) and JSON-typed object/array inputs are REST/MCP-only.
  • Verified scope all published methods
  • Assurance Core
  • release artifacts aligned PASS
  • schema output required declared PASS
  • schema valid input executes PASS
  • engine success output validates schema PASS
  • rest success payload validates schema PASS

Fingerprint sha256:561c36f3d0381a72bbd6f3e64fcc2ed76606765e189a55b5a1329fa99a328f31

Input schema sha256:76f7d15fb6093a1cfca27199d6571322d789a599ac7012371c89573ef2159f22 · Output schema sha256:89d59e3d5b954d05402cc590fd56c6aedbe6f972553fd3a9441d862632f62cf1 · 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.

Build & scope

Engine identity and declared limitations.

Tool
ordinary-differential-equation
Engine
ordinary-differential-equation
Engine build
2026-10-18
Evidence core
2026-10-18T12:51:28.122Z
Rendered
2026-10-23T01:27:11.653Z
Last reviewed
—

Assumptions

  • The Result Card includes an engine-linked Live graph of numerical y(t) using the same Overview / Focus template as Log (Overview window t0…t1; current point is published y(t1); same Euler/RK4 stepper as REST/MCP). Exact y(t1) is a reference when it differs. Not rk45, not systems, not PDE, not CAS.
  • Scalar first-order only
  • Polynomial quadrature and baseline use exact dyadic intermediates over binary64 inputs and actual representable time nodes; output remains binary64, not symbolic.
  • y, exact and signed difference publish12 significant digits; model inputs and h retain binary64 precision. The graph is sampled to at most257 points, not a complete trajectory export.
  • Polynomial degree ≤ 8

Limitations

  • Not rk45 / adaptive / Dormand–Prince
  • Not systems of ODEs or second-order IVP
  • Not PDE / CAS / symbolic solve
  • n ≤ 10000
  • t1 = t0 → INVALID_INPUT
  • n must be an integer; fractional steps and unknown inputs are rejected
  • abs_err is a legacy signed difference before publication rounding; diagnostics names signed/absolute/relative error explicitly
  • Fixed-step method completion is not accuracy acceptance; inspect diagnostics and warnings
  • Finite decimal text only; aliases must agree; explicit empty modes/fields do not default.
  • Polynomial intermediates and quadrature sums use exact binary64-input dyadic arithmetic; final results are binary64 projections.
  • Raw input/model fields and h retain binary64 precision; y, exact and error publish12 significant digits. Live graph uses at most257 samples and is not a complete trajectory export.

Sources

Standards and references supporting this tool specification. Methods are listed separately — they are algorithms, not bibliographic sources.

Verification disclosure

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.