Slice builds each phase's coverage argument and the detail argument (centre,
axis, radius in noise space); AffineBound and its gate read them whole.
AffineWarpStaysEnclosed checks the rays against those arguments and their
derivation from CentredWarp; AffineCompositionIsTighterThanItsRanges sees the
shared-eps composition and the intersection.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The 2D ball carries no evaluation allowance, like iv_snoise_ball(float3); the
caller that rounds the flow's point adds 2^-14 + L2 2^-20 (|centre|_1 + radius),
the convention af_snoise carries for the 3D point.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Upstream Ashima's step compares are cyclic, so where x0's components compare equal
(the origin, every (t,t,t), all lattice diagonals) the corners stop being a simplex
and snoise jumps (about 0.77 near the origin, up to 1.56 further out). The z-x compare
is now strict, giving Gustavson's simplexnoise1234 order in C#, both HLSL copies and
the generated GLSL. Off the z=x tie set values are bit-identical. SNOISE_LIPSCHITZ
re-pinned 10.099261 -> 7.9640074. A value change: next cultmath-unity release is
Breaking (follow-up snoise-tie-release). Soul s1 held (cut-snoise-tie.h2).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Soul s1 (cut-snoise-tie) measured 1.55 to 1.56 at magnitudes 64 and above;
0.77 holds only near the origin. Comment and design text only.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Two families appended after phacelle, cultmath_snoise(vec3) and
cultmath_snoise_grad(vec3) over TiePoint3 (seven of eight cases on lattice
diagonals, one of eight on (t, t, t)); the shared Random leaves every earlier
family's draws unchanged. Both carry the tolerance text "ulp-bounded, on the
simplex tie set" and share a glsl name with the random families, so a consumer
tells them apart by index or tolerance.
Regenerated on Yggdrasil (Linux, the fixture's platform) with
CULTMATH_WRITE_GLSL=1. CultMath.glsl: the two tie lines, the constant, one
comment sentence. Fixture: 41 cases of iv_snoise_ball(vec3) and iv_fbm_ball move
with L, everything else is byte-identical before the appended families. Full
suite on Yggdrasil: 290 of 290.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
MeasureLipschitz on e89fd687 (Yggdrasil): sampled 7.1933103, refined 7.2400064,
x 1.10 = 7.9640074, witness (229.92526, 73.37927, -66.14595). The old 9.181147
sat on the diagonal tie. C#, both Interval.hlsl copies, the pin test's witness
and design.md's provenance numbers move together; CultMath.glsl follows with the
regenerated fixture.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The cyclic non-strict step(x0.yzx, x0.xyz) made the four corners a non-simplex
where x0's components are equal: the origin and every (t, t, t) read a different
function (snoise(0) was -0.4358730, jumps up to 0.77). One strict compare, z
against x, gives simplexnoise1234's order. Off the tie set old and new return
the same bits.
NoiseGradTests.SnoiseIsContinuousOnTheSimplexDiagonal fails against the old
rule (0.054 within 1 ulp of (-53.6, -53.6, -53.6)). The origin golden is 0 and a
(1.25, 1.25, 1.25) diagonal row joins it. SNOISE_LIPSCHITZ and the GLSL follow
in the next commits, so the ratio pin and CommittedGlslEqualsLowering are red here.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
(c) is FogField.DeepWell (ruling wells-scale-order-of-magnitude), measured under
the centred-warp interval bound and the affine bound on the same seeds; the
shallow wells are the printed (c-band) row. The warp-enlarged ball and the
warp-carrying detail ball in FogField.Bound are gone. (c) asserts affine cheaper
than the centred interval at both warps, with the 0.5 cull floor under both
(ruling affine-wells-contract-cheaper-than-interval).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
npm Publish and Python build lines run the guard under env -u NODE_OPTIONS;
the shell/defaults ban is dropped (deliberate edits are review's job);
check-changelog-semver reads Keep a Changelog's dated heading; semver-policy.md
states the threat model and the first-release versions. Verdict
cut-release-guard-tidy.s1 holds with no findings.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A domain built so span integration holds by construction: sums of
nonnegative compact ellipsoid kernels in a mass-conserving multiplicative
cascade, per-kernel affine warp, all nonlinearity moved into placement.
Measured on Yggdrasil against the shipped void and wells, with a physical
reference (Beer-Lambert, single scattering with HG).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The python-build step runs the same guard whose output is the dist PyPI
uploads, so a NODE_OPTIONS preload in the workflow or job env could stub
its checks exactly as it could on the npm Publish line. It now starts
under env -u NODE_OPTIONS; the pin's exact text and semver-policy.md say
so. The pin test's if: always() edit matched the npm line's old text and
had become a no-op; it now matches the pinned line.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- The npm Publish step starts the guard under `env -u NODE_OPTIONS`, and the
pin's exact text carries it, so a workflow or job NODE_OPTIONS preload
cannot run inside the guard (preload-fakes-guard).
- The pin's shell/defaults ban and its test are gone: it defended only a
deliberate edit, which is review's job (suite-gate-structural-not-parsed),
and refused a harmless defaults.run.working-directory. A step-level shell
still fails the exact comparison (shell-ban-deliberate-only).
- docs/semver-policy.md states that the guard and pin stop accidental edits,
that the cleaned environment removes known variables rather than proving no
inherited one matters (env-denylist-overclaim), recommends a
required-reviewer environment for the npm job, and gives the first-release
note with the valid next versions.
- check-changelog-semver accepts `## [x.y.z] - YYYY-MM-DD` and
`## x.y.z - YYYY-MM-DD` beside the plain forms (changelog-first-entry).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Suite, semver check and package tests run without NODE_OPTIONS, NODE_TEST_* or
the publish token; tag/version match and package tests moved into the guard,
so no step condition can skip them. Verdict cut-release-guard-env.s1 holds;
its Low findings (threat model wording, shell ban, changelog heading) follow.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The GLSL is generated from the HLSL by one lowering, mirrored against C# by a
committed fixture evaluated on its own arguments (portable across OSes), with
the frustum ball checked by two-sided enclosure and Phacelle by a scaled
absolute bound. design.md conflict: keeps the branch's sqrt-precision text and
drops the 6e-4 sentence interval-ops r4 removed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
iv_frustum_ball widened by its derived float32 bound, envelope-then-noise
fields, the analytic constant-density body, one shipped void (Rh 198, S 50),
LOD agreement against a converged march, and the wells contract as ruled
(cultmath-tapes:ruling:wells-overhead-plus-floor). Verdict
cut-interval-ops.s3 holds; one Low finding (LodStep footprint branch unpinned)
is recorded for the next interval cut.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A job or workflow defaults.run.shell can run the exact Publish line through any
command, with the token in its env. Only continue-on-error leaves the pin: a
check that fails inside the guard refuses the release whatever the step allows.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Measured on the shipped void and the Aetheria wells at warp 60: affine warp
spans, simplex cells and lattice points per span, exact-against-march cost and
error per octave, high-octave variance, clamp crossings and transmittance
error; plus Aetheria's triangle noise. Verdict: exact integration does not
beat sampling for optical depth on either field.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
gpu-snoise-parity is Asura's FXC and Unity readback of snoise, not a GLSL run; phacelle's GLSL
evaluator is the cultmath-tapes follow-up glsl-browser-parity-run.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
GoldenFixtureMatchesCSharp draws arguments only under CULTMATH_WRITE_GLSL=1: the draw calls MathF.Pow,
whose bits differ by OS. Checking decodes each committed case's arguments and evaluates C# on them.
phacelle is compared across platforms within 2^-20 max(|v|, 1) per component; iv_exp keeps one ulp.
The iv_frustum_ball check gains the upper side r + d + 2^-19 (r + |c|_1) that NoiseBoundTests pins.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The parity section states the enclosure rule, restores that the evaluator concatenates
CultMath.Phacelle.glsl for the phacelle family, and names CultLib's browser-pane run
(follow-up gpu-snoise-parity) as that evaluator. The frustum-ball bound no longer assumes
a correctly rounded GLSL sqrt: proven for C# and HLSL, measured on GLSL.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
iv_frustum_ball's sqrt is not correctly rounded on WebGL2, so its family is ulp-bounded and
its entry carries the enclosure rule the consumer applies in double; GoldenFixtureMatchesCSharp
applies it to C#'s results too. A phacelle family (normalization 0.5, platform exp, sin, cos)
is appended after iv_exp; every earlier family's cases are byte-identical.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
EveryPublicFunctionIsLowered compares (name, parameter types) after the lowering's own
step 3, so a dropped overload such as cultmath_pcg3d(vec2) fails. TwoFileHeaderContract
pins distinct guards, the MIT and MPL-2.0 provenance labels, and the MIT file's pointer
to CultMath.Phacelle.glsl. GlslLowering exposes step 3 as LowerTypeNames; output unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
release-after-suite.mjs now takes the run's git ref and, from one process, checks
the tag against the declared prefix and manifest version, runs the release-checks
suite, the semver policy check and the package's own tests, then releases. The
checks run without NODE_OPTIONS, NODE_TEST_*, or NODE_AUTH_TOKEN; only the release
action gets the token. The workflow's separate tag, semver and test steps are gone,
and so is the pin's continue-on-error/shell/defaults ban, which no release check
needs any more.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The release script refuses a CultMath.dll that differs from the tracked bytes
(clearing every obj configuration, CultMath source pinned to CRLF, a git
checkout required). CI publishes through one guarded entrypoint that runs the
release-checks suite first, replacing the deleted if:-condition pin (ruling
asura:ruling:suite-gate-structural-not-parsed). Verdict cut-3-gate.s4: merge;
node-env-skips-suite (Medium) and always-skips-other-gates (Low) follow before
the next release tag.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every figure is from r4's IV-REPORT lines: (a)-(d) with (c)'s oracle ceiling,
efficiency, cull fraction and probe overhead; (e) at the shipped void with the
two depth errors against the converged march; the analytic body and its pin;
the [w(z1), 1] weight-bound deviation; the contracts as asserted, (c) in the
words of ruling wells-overhead-plus-floor.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
