@photivra/engine 2.20.0 → 2.22.0

package/docs/API_REFERENCE.md

@@ -1,6 +1,6 @@
 # Public API reference
 
-Generated from the exact root exports for **@photivra/engine 2.20.0 / root contract 2.20.0** by `node scripts/generate-api-reference.mjs`. No module deep import is supported.
+Generated from the exact root exports for **@photivra/engine 2.22.0 / root contract 2.22.0** by `node scripts/generate-api-reference.mjs`. No module deep import is supported.
 
 Use [the developer guide](https://github.com/Photivra/photivra/blob/main/docs/DEVELOPERS.md) for executable paths and scientific boundaries. Types are contracts, not runtime validation: parse untrusted data, preserve units/reference frames, read assessment blockers, and retain evidence/limitations. Optional properties do not imply a universal default. Model/schema IDs remain independent of the distribution.
 
@@ -101,6 +101,28 @@ Start here to find a function by task. [Using Photivra from an agent](AGENT_GUID
 | [`SensorEquivalenceFieldOfView`](api/camera-sensor-equivalence.md#sensorequivalencefieldofview) | Type |
 | [`SensorEquivalenceQuantity`](api/camera-sensor-equivalence.md#sensorequivalencequantity) | Type |
 
+## capture/c2pa-embed.ts
+
+[Detailed contracts](api/capture-c2pa-embed.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/capture/c2pa-embed.ts)
+
+| Export | Kind |
+| --- | --- |
+| [`bindC2paManifestStore`](api/capture-c2pa-embed.md#bindc2pamanifeststore) | Runtime |
+| [`BindC2paManifestStoreInput`](api/capture-c2pa-embed.md#bindc2pamanifeststoreinput) | Type |
+| [`C2PA_ALGORITHMIC_MEDIA_SOURCE_TYPE`](api/capture-c2pa-embed.md#c2pa_algorithmic_media_source_type) | Runtime |
+| [`C2PA_MANIFEST_STORE_MAXIMUM_BYTES`](api/capture-c2pa-embed.md#c2pa_manifest_store_maximum_bytes) | Runtime |
+| [`C2paBoundFile`](api/capture-c2pa-embed.md#c2paboundfile) | Type |
+| [`C2paByteRange`](api/capture-c2pa-embed.md#c2pabyterange) | Type |
+| [`C2paDataHash`](api/capture-c2pa-embed.md#c2padatahash) | Type |
+| [`C2paFoundManifestStore`](api/capture-c2pa-embed.md#c2pafoundmanifeststore) | Type |
+| [`C2paHardBinding`](api/capture-c2pa-embed.md#c2pahardbinding) | Type |
+| [`calculateC2paDataHash`](api/capture-c2pa-embed.md#calculatec2padatahash) | Runtime |
+| [`CalculateC2paDataHashInput`](api/capture-c2pa-embed.md#calculatec2padatahashinput) | Type |
+| [`parseC2paManifestStore`](api/capture-c2pa-embed.md#parsec2pamanifeststore) | Runtime |
+| [`ParseC2paManifestStoreInput`](api/capture-c2pa-embed.md#parsec2pamanifeststoreinput) | Type |
+| [`setC2paManifestStore`](api/capture-c2pa-embed.md#setc2pamanifeststore) | Runtime |
+| [`SetC2paManifestStoreInput`](api/capture-c2pa-embed.md#setc2pamanifeststoreinput) | Type |
+
 ## capture/capture-export-metadata.ts
 
 [Detailed contracts](api/capture-capture-export-metadata.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/capture/capture-export-metadata.ts)
@@ -124,13 +146,57 @@ Start here to find a function by task. [Using Photivra from an agent](AGENT_GUID
 | --- | --- |
 | [`DNG_IMPORT_LIMITS`](api/capture-dng-import.md#dng_import_limits) | Runtime |
 | [`DNG_IMPORT_SCHEMA_VERSION`](api/capture-dng-import.md#dng_import_schema_version) | Runtime |
+| [`DngImportAsyncInput`](api/capture-dng-import.md#dngimportasyncinput) | Type |
 | [`DngImportInput`](api/capture-dng-import.md#dngimportinput) | Type |
+| [`importDngRawImage`](api/capture-dng-import.md#importdngrawimage) | Runtime |
 | [`ImportedDngBlackLevel`](api/capture-dng-import.md#importeddngblacklevel) | Type |
 | [`ImportedDngCfa`](api/capture-dng-import.md#importeddngcfa) | Type |
 | [`ImportedDngRaw`](api/capture-dng-import.md#importeddngraw) | Type |
 | [`ImportedDngRect`](api/capture-dng-import.md#importeddngrect) | Type |
+| [`ImportedDngSemanticMask`](api/capture-dng-import.md#importeddngsemanticmask) | Type |
 | [`parseDngRawImage`](api/capture-dng-import.md#parsedngrawimage) | Runtime |
 
+## capture/dng-opcodes.ts
+
+[Detailed contracts](api/capture-dng-opcodes.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/capture/dng-opcodes.ts)
+
+| Export | Kind |
+| --- | --- |
+| [`applyDngOpcodes`](api/capture-dng-opcodes.md#applydngopcodes) | Runtime |
+| [`ApplyDngOpcodesOptions`](api/capture-dng-opcodes.md#applydngopcodesoptions) | Type |
+| [`DngOpcode`](api/capture-dng-opcodes.md#dngopcode) | Type |
+| [`DngOpcodeArea`](api/capture-dng-opcodes.md#dngopcodearea) | Type |
+| [`DngOpcodeParameters`](api/capture-dng-opcodes.md#dngopcodeparameters) | Type |
+| [`DngStageImage`](api/capture-dng-opcodes.md#dngstageimage) | Type |
+| [`parseDngOpcodeList`](api/capture-dng-opcodes.md#parsedngopcodelist) | Runtime |
+
+## capture/dng-profile-maps.ts
+
+[Detailed contracts](api/capture-dng-profile-maps.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/capture/dng-profile-maps.ts)
+
+| Export | Kind |
+| --- | --- |
+| [`applyDngProfileGainTableMap`](api/capture-dng-profile-maps.md#applydngprofilegaintablemap) | Runtime |
+| [`ApplyDngProfileGainTableMapInput`](api/capture-dng-profile-maps.md#applydngprofilegaintablemapinput) | Type |
+| [`applyDngRgbTables`](api/capture-dng-profile-maps.md#applydngrgbtables) | Runtime |
+| [`ApplyDngRgbTablesInput`](api/capture-dng-profile-maps.md#applydngrgbtablesinput) | Type |
+| [`DngProfileGainTableMap`](api/capture-dng-profile-maps.md#dngprofilegaintablemap) | Type |
+| [`DngRgbImage`](api/capture-dng-profile-maps.md#dngrgbimage) | Type |
+| [`DngRgbTable`](api/capture-dng-profile-maps.md#dngrgbtable) | Type |
+| [`DngRgbTables`](api/capture-dng-profile-maps.md#dngrgbtables) | Type |
+| [`parseDngProfileGainTableMap`](api/capture-dng-profile-maps.md#parsedngprofilegaintablemap) | Runtime |
+| [`parseDngRgbTables`](api/capture-dng-profile-maps.md#parsedngrgbtables) | Runtime |
+
+## capture/dng-stage-image.ts
+
+[Detailed contracts](api/capture-dng-stage-image.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/capture/dng-stage-image.ts)
+
+| Export | Kind |
+| --- | --- |
+| [`developDngStageImage`](api/capture-dng-stage-image.md#developdngstageimage) | Runtime |
+| [`DevelopDngStageImageOptions`](api/capture-dng-stage-image.md#developdngstageimageoptions) | Type |
+| [`DngStageImageResult`](api/capture-dng-stage-image.md#dngstageimageresult) | Type |
+
 ## capture/environment-raw-producer.ts
 
 [Detailed contracts](api/capture-environment-raw-producer.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/capture/environment-raw-producer.ts)
@@ -180,6 +246,16 @@ Start here to find a function by task. [Using Photivra from an agent](AGENT_GUID
 | [`setManualFocusState`](api/capture-focus-control.md#setmanualfocusstate) | Runtime |
 | [`SetManualFocusStateInput`](api/capture-focus-control.md#setmanualfocusstateinput) | Type |
 
+## capture/jpeg-xl-tiles.ts
+
+[Detailed contracts](api/capture-jpeg-xl-tiles.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/capture/jpeg-xl-tiles.ts)
+
+| Export | Kind |
+| --- | --- |
+| [`createJxlOxideDngDecoder`](api/capture-jpeg-xl-tiles.md#createjxloxidedngdecoder) | Runtime |
+| [`DngJpegXlDecoder`](api/capture-jpeg-xl-tiles.md#dngjpegxldecoder) | Type |
+| [`JxlOxideModule`](api/capture-jpeg-xl-tiles.md#jxloxidemodule) | Type |
+
 ## capture/linear-encoding.ts
 
 [Detailed contracts](api/capture-linear-encoding.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/capture/linear-encoding.ts)
@@ -204,6 +280,16 @@ Start here to find a function by task. [Using Photivra from an agent](AGENT_GUID
 | [`LosslessJpegLimits`](api/capture-lossless-jpeg.md#losslessjpeglimits) | Type |
 | [`parseLosslessJpeg`](api/capture-lossless-jpeg.md#parselosslessjpeg) | Runtime |
 
+## capture/lossy-jpeg.ts
+
+[Detailed contracts](api/capture-lossy-jpeg.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/capture/lossy-jpeg.ts)
+
+| Export | Kind |
+| --- | --- |
+| [`LossyJpegImage`](api/capture-lossy-jpeg.md#lossyjpegimage) | Type |
+| [`LossyJpegLimits`](api/capture-lossy-jpeg.md#lossyjpeglimits) | Type |
+| [`parseLossyJpeg`](api/capture-lossy-jpeg.md#parselossyjpeg) | Runtime |
+
 ## capture/native-environment-optical-stage.ts
 
 [Detailed contracts](api/capture-native-environment-optical-stage.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/capture/native-environment-optical-stage.ts)
@@ -727,6 +813,16 @@ Start here to find a function by task. [Using Photivra from an agent](AGENT_GUID
 | --- | --- |
 | [`InvalidConfigurationError`](api/core-configuration-error.md#invalidconfigurationerror) | Runtime |
 
+## core/execution-error.ts
+
+[Detailed contracts](api/core-execution-error.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/core/execution-error.ts)
+
+| Export | Kind |
+| --- | --- |
+| [`ImageKernelExecutionError`](api/core-execution-error.md#imagekernelexecutionerror) | Runtime |
+| [`ImageKernelFailureReason`](api/core-execution-error.md#imagekernelfailurereason) | Type |
+| [`ImageKernelFallbackReason`](api/core-execution-error.md#imagekernelfallbackreason) | Type |
+
 ## core/image-formation.ts
 
 [Detailed contracts](api/core-image-formation.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/core/image-formation.ts)
@@ -976,6 +1072,26 @@ Start here to find a function by task. [Using Photivra from an agent](AGENT_GUID
 | [`GenericReleaseDriveMode`](api/equipment-release-capabilities.md#genericreleasedrivemode) | Type |
 | [`parseGenericReleaseCapabilityProfile`](api/equipment-release-capabilities.md#parsegenericreleasecapabilityprofile) | Runtime |
 
+## equipment/sensor-profile-body.ts
+
+[Detailed contracts](api/equipment-sensor-profile-body.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/equipment/sensor-profile-body.ts)
+
+| Export | Kind |
+| --- | --- |
+| [`SENSOR_PROFILE_ISO_ANCHOR`](api/equipment-sensor-profile-body.md#sensor_profile_iso_anchor) | Runtime |
+| [`SensorProfile`](api/equipment-sensor-profile-body.md#sensorprofile) | Type |
+| [`SensorProfileValueKind`](api/equipment-sensor-profile-body.md#sensorprofilevaluekind) | Type |
+
+## equipment/sensor-profiles.ts
+
+[Detailed contracts](api/equipment-sensor-profiles.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/equipment/sensor-profiles.ts)
+
+| Export | Kind |
+| --- | --- |
+| [`resolveSensorProfile`](api/equipment-sensor-profiles.md#resolvesensorprofile) | Runtime |
+| [`SENSOR_PROFILE_IDS`](api/equipment-sensor-profiles.md#sensor_profile_ids) | Runtime |
+| [`SENSOR_PROFILES`](api/equipment-sensor-profiles.md#sensor_profiles) | Runtime |
+
 ## exposure/auto-iso-slowest-shutter.ts
 
 [Detailed contracts](api/exposure-auto-iso-slowest-shutter.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/exposure/auto-iso-slowest-shutter.ts)
@@ -1314,6 +1430,19 @@ Start here to find a function by task. [Using Photivra from an agent](AGENT_GUID
 | [`DisplayViewCorrectionStateSet`](api/facade-display-view.md#displayviewcorrectionstateset) | Type |
 | [`serializeDisplayViewCorrectionStates`](api/facade-display-view.md#serializedisplayviewcorrectionstates) | Runtime |
 
+## facade/exposure-build-up.ts
+
+[Detailed contracts](api/facade-exposure-build-up.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/facade/exposure-build-up.ts)
+
+| Export | Kind |
+| --- | --- |
+| [`EXPOSURE_BUILD_UP_MAXIMUM_FRACTIONS`](api/facade-exposure-build-up.md#exposure_build_up_maximum_fractions) | Runtime |
+| [`EXPOSURE_BUILD_UP_STREAM_BASE`](api/facade-exposure-build-up.md#exposure_build_up_stream_base) | Runtime |
+| [`simulateExposureBuildUp`](api/facade-exposure-build-up.md#simulateexposurebuildup) | Runtime |
+| [`SimulateExposureBuildUpFrame`](api/facade-exposure-build-up.md#simulateexposurebuildupframe) | Type |
+| [`SimulateExposureBuildUpOptions`](api/facade-exposure-build-up.md#simulateexposurebuildupoptions) | Type |
+| [`SimulateExposureBuildUpResult`](api/facade-exposure-build-up.md#simulateexposurebuildupresult) | Type |
+
 ## facade/fast-methods.ts
 
 [Detailed contracts](api/facade-fast-methods.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/facade/fast-methods.ts)
@@ -1323,6 +1452,52 @@ Start here to find a function by task. [Using Photivra from an agent](AGENT_GUID
 | [`SIMULATE_CAPTURE_FAST_METHODS`](api/facade-fast-methods.md#simulate_capture_fast_methods) | Runtime |
 | [`SimulateCaptureFastMethod`](api/facade-fast-methods.md#simulatecapturefastmethod) | Type |
 
+## facade/image-plane-enable.ts
+
+[Detailed contracts](api/facade-image-plane-enable.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/facade/image-plane-enable.ts)
+
+| Export | Kind |
+| --- | --- |
+| [`prepareImagePlaneScenes`](api/facade-image-plane-enable.md#prepareimageplanescenes) | Runtime |
+
+## facade/mixed-illumination.ts
+
+[Detailed contracts](api/facade-mixed-illumination.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/facade/mixed-illumination.ts)
+
+| Export | Kind |
+| --- | --- |
+| [`MixedIllumination`](api/facade-mixed-illumination.md#mixedillumination) | Type |
+| [`MixedIlluminationBlend`](api/facade-mixed-illumination.md#mixedilluminationblend) | Type |
+| [`MixedIlluminationPoint`](api/facade-mixed-illumination.md#mixedilluminationpoint) | Type |
+| [`MixedIlluminationPointSource`](api/facade-mixed-illumination.md#mixedilluminationpointsource) | Type |
+| [`MixedIlluminationSource`](api/facade-mixed-illumination.md#mixedilluminationsource) | Type |
+
+## facade/quality-paths.ts
+
+[Detailed contracts](api/facade-quality-paths.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/facade/quality-paths.ts)
+
+| Export | Kind |
+| --- | --- |
+| [`prepareQualityPaths`](api/facade-quality-paths.md#preparequalitypaths) | Runtime |
+
+## facade/representative-light-sources.ts
+
+[Detailed contracts](api/facade-representative-light-sources.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/facade/representative-light-sources.ts)
+
+| Export | Kind |
+| --- | --- |
+| [`createRepresentativeLightSourceIlluminant`](api/facade-representative-light-sources.md#createrepresentativelightsourceilluminant) | Runtime |
+| [`createRepresentativeLightSourceSpectrum`](api/facade-representative-light-sources.md#createrepresentativelightsourcespectrum) | Runtime |
+| [`getRepresentativeLightSourceSet`](api/facade-representative-light-sources.md#getrepresentativelightsourceset) | Runtime |
+| [`REPRESENTATIVE_LIGHT_SOURCE_IDS`](api/facade-representative-light-sources.md#representative_light_source_ids) | Runtime |
+| [`REPRESENTATIVE_LIGHT_SOURCE_SET_ID`](api/facade-representative-light-sources.md#representative_light_source_set_id) | Runtime |
+| [`REPRESENTATIVE_LIGHT_SOURCE_SET_VERSION`](api/facade-representative-light-sources.md#representative_light_source_set_version) | Runtime |
+| [`REPRESENTATIVE_LIGHT_SOURCE_WAVELENGTH_RANGE_NANOMETERS`](api/facade-representative-light-sources.md#representative_light_source_wavelength_range_nanometers) | Runtime |
+| [`RepresentativeLightSource`](api/facade-representative-light-sources.md#representativelightsource) | Type |
+| [`RepresentativeLightSourceBand`](api/facade-representative-light-sources.md#representativelightsourceband) | Type |
+| [`RepresentativeLightSourceId`](api/facade-representative-light-sources.md#representativelightsourceid) | Type |
+| [`RepresentativeLightSourceSet`](api/facade-representative-light-sources.md#representativelightsourceset) | Type |
+
 ## facade/scene-radiance.ts
 
 [Detailed contracts](api/facade-scene-radiance.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/facade/scene-radiance.ts)
@@ -1337,6 +1512,8 @@ Start here to find a function by task. [Using Photivra from an agent](AGENT_GUID
 | [`CHART_SIZE_DEFAULT_SURROUND_REFLECTANCE`](api/facade-scene-radiance.md#chart_size_default_surround_reflectance) | Runtime |
 | [`ChartSizeOption`](api/facade-scene-radiance.md#chartsizeoption) | Type |
 | [`createPlanarSceneRadianceProvider`](api/facade-scene-radiance.md#createplanarsceneradianceprovider) | Runtime |
+| [`ImagePlaneImage`](api/facade-scene-radiance.md#imageplaneimage) | Type |
+| [`ImagePlaneScene`](api/facade-scene-radiance.md#imageplanescene) | Type |
 | [`PHOTOPIC_MAXIMUM_LUMINOUS_EFFICACY_LM_PER_W`](api/facade-scene-radiance.md#photopic_maximum_luminous_efficacy_lm_per_w) | Runtime |
 | [`PlanarSceneRadianceProvider`](api/facade-scene-radiance.md#planarsceneradianceprovider) | Type |
 | [`resolveCaptureIlluminant`](api/facade-scene-radiance.md#resolvecaptureilluminant) | Runtime |
@@ -1451,9 +1628,6 @@ Start here to find a function by task. [Using Photivra from an agent](AGENT_GUID
 | [`ImageKernelBackend`](api/gpu-image-kernel-execution.md#imagekernelbackend) | Type |
 | [`ImageKernelCodeErrorBound`](api/gpu-image-kernel-execution.md#imagekernelcodeerrorbound) | Type |
 | [`ImageKernelExecution`](api/gpu-image-kernel-execution.md#imagekernelexecution) | Type |
-| [`ImageKernelExecutionError`](api/gpu-image-kernel-execution.md#imagekernelexecutionerror) | Runtime |
-| [`ImageKernelFailureReason`](api/gpu-image-kernel-execution.md#imagekernelfailurereason) | Type |
-| [`ImageKernelFallbackReason`](api/gpu-image-kernel-execution.md#imagekernelfallbackreason) | Type |
 | [`ImageKernelRelativeErrorBound`](api/gpu-image-kernel-execution.md#imagekernelrelativeerrorbound) | Type |
 | [`ImageKernelSeededRealizationBound`](api/gpu-image-kernel-execution.md#imagekernelseededrealizationbound) | Type |
 | [`WEBGPU_F32_TRANSFER_ABSOLUTE_BOUND`](api/gpu-image-kernel-execution.md#webgpu_f32_transfer_absolute_bound) | Runtime |
@@ -1563,6 +1737,14 @@ Start here to find a function by task. [Using Photivra from an agent](AGENT_GUID
 | [`CalculateProjectedMotionBlurInput`](api/motion-projected-motion.md#calculateprojectedmotionblurinput) | Type |
 | [`ProjectedMotionBlur`](api/motion-projected-motion.md#projectedmotionblur) | Type |
 
+## motion/subject-motion.ts
+
+[Detailed contracts](api/motion-subject-motion.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/motion/subject-motion.ts)
+
+| Export | Kind |
+| --- | --- |
+| [`SimulateCaptureSubjectMotion`](api/motion-subject-motion.md#simulatecapturesubjectmotion) | Type |
+
 ## motion/temporal-exposure-sum.ts
 
 [Detailed contracts](api/motion-temporal-exposure-sum.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/motion/temporal-exposure-sum.ts)
@@ -1801,6 +1983,17 @@ Start here to find a function by task. [Using Photivra from an agent](AGENT_GUID
 | [`resolveFrontOfLensFilterTransmission`](api/optics-front-of-lens-filter.md#resolvefrontoflensfiltertransmission) | Runtime |
 | [`ResolveFrontOfLensFilterTransmissionInput`](api/optics-front-of-lens-filter.md#resolvefrontoflensfiltertransmissioninput) | Type |
 
+## optics/generic-lens-lateral-color.ts
+
+[Detailed contracts](api/optics-generic-lens-lateral-color.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/optics/generic-lens-lateral-color.ts)
+
+| Export | Kind |
+| --- | --- |
+| [`calculateLateralColorFringe`](api/optics-generic-lens-lateral-color.md#calculatelateralcolorfringe) | Runtime |
+| [`CalculateLateralColorFringeInput`](api/optics-generic-lens-lateral-color.md#calculatelateralcolorfringeinput) | Type |
+| [`LateralColorDisplacement`](api/optics-generic-lens-lateral-color.md#lateralcolordisplacement) | Type |
+| [`LateralColorFringe`](api/optics-generic-lens-lateral-color.md#lateralcolorfringe) | Type |
+
 ## optics/generic-nd-filter-presets.ts
 
 [Detailed contracts](api/optics-generic-nd-filter-presets.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/optics/generic-nd-filter-presets.ts)
@@ -2097,6 +2290,16 @@ Start here to find a function by task. [Using Photivra from an agent](AGENT_GUID
 | [`ZernikeCoefficient`](api/optics-zernike-wavefront.md#zernikecoefficient) | Type |
 | [`ZernikeOrder`](api/optics-zernike-wavefront.md#zernikeorder) | Type |
 
+## output/camera-channel-response.ts
+
+[Detailed contracts](api/output-camera-channel-response.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/output/camera-channel-response.ts)
+
+| Export | Kind |
+| --- | --- |
+| [`calculateCameraChannelResponse`](api/output-camera-channel-response.md#calculatecamerachannelresponse) | Runtime |
+| [`CalculateCameraChannelResponseInput`](api/output-camera-channel-response.md#calculatecamerachannelresponseinput) | Type |
+| [`CameraChannelResponse`](api/output-camera-channel-response.md#camerachannelresponse) | Type |
+
 ## output/capture-corrected-sdr.ts
 
 [Detailed contracts](api/output-capture-corrected-sdr.md) · [Source](https://github.com/Photivra/photivra/blob/main/src/output/capture-corrected-sdr.ts)
@@ -2183,6 +2386,7 @@ Start here to find a function by task. [Using Photivra from an agent](AGENT_GUID
 
 | Export | Kind |
 | --- | --- |
+| [`ChannelDispersion`](api/output-developed-lens-corrections.md#channeldispersion) | Type |
 | [`GENERIC_LENS_CORRECTION_CHANNEL_WAVELENGTHS_NM`](api/output-developed-lens-corrections.md#generic_lens_correction_channel_wavelengths_nm) | Runtime |
 
 ## output/generic-lens-corrections.ts
@@ -2191,6 +2395,8 @@ Start here to find a function by task. [Using Photivra from an agent](AGENT_GUID
 
 | Export | Kind |
 | --- | --- |
+| [`calculateChannelEffectiveDispersion`](api/output-generic-lens-corrections.md#calculatechanneleffectivedispersion) | Runtime |
+| [`CalculateChannelEffectiveDispersionInput`](api/output-generic-lens-corrections.md#calculatechanneleffectivedispersioninput) | Type |
 | [`createGenericLensCorrectionProfile`](api/output-generic-lens-corrections.md#creategenericlenscorrectionprofile) | Runtime |
 | [`CreateGenericLensCorrectionProfileInput`](api/output-generic-lens-corrections.md#creategenericlenscorrectionprofileinput) | Type |
 
@@ -2336,7 +2542,9 @@ Start here to find a function by task. [Using Photivra from an agent](AGENT_GUID
 | [`calculatePrintPlan`](api/output-print-plan.md#calculateprintplan) | Runtime |
 | [`calculatePrintSizeLimit`](api/output-print-plan.md#calculateprintsizelimit) | Runtime |
 | [`parsePrintPlanInput`](api/output-print-plan.md#parseprintplaninput) | Runtime |
+| [`PrintContainLayout`](api/output-print-plan.md#printcontainlayout) | Type |
 | [`PrintedImageSize`](api/output-print-plan.md#printedimagesize) | Type |
+| [`PrintFit`](api/output-print-plan.md#printfit) | Type |
 | [`PrintLength`](api/output-print-plan.md#printlength) | Type |
 | [`PrintNativeSource`](api/output-print-plan.md#printnativesource) | Type |
 | [`PrintPlan`](api/output-print-plan.md#printplan) | Type |
@@ -3194,6 +3402,8 @@ Start here to find a function by task. [Using Photivra from an agent](AGENT_GUID
 
 | Export | Kind |
 | --- | --- |
+| [`SimulateCaptureMidGreyFigure`](api/sensor-sensor-characteristics.md#simulatecapturemidgreyfigure) | Type |
+| [`SimulateCaptureNativeSensorReport`](api/sensor-sensor-characteristics.md#simulatecapturenativesensorreport) | Type |
 | [`SimulateCaptureNoiseComponents`](api/sensor-sensor-characteristics.md#simulatecapturenoisecomponents) | Type |
 | [`SimulateCaptureSensorCharacteristics`](api/sensor-sensor-characteristics.md#simulatecapturesensorcharacteristics) | Type |
 | [`SimulateCaptureSensorReport`](api/sensor-sensor-characteristics.md#simulatecapturesensorreport) | Type |

package/docs/QUICK_START.md

@@ -48,7 +48,7 @@ the example above is [simulate-capture.mjs](examples/simulate-capture.mjs).
 | `raw` | The packed native RAW: codes, black levels, saturation flags and per-phase white levels |
 | `plan` | The native RAW plan the engine executed |
 | `request` | The complete underlying request: `environmentRaw` (the native environment RAW input), `siteInput(index)` (rebuilds the physical input of any sensor site), `development` (the RAW development input) and `whiteBalanceState` |
-| `sensor` | 2.20 (#555, #535): the sensor characteristics used (pitch, peak QE, full well, read noise of each conversion gain and the dual-gain switch ISO, readout time, ADC bit depth, fixed-pattern noise), the calculated dynamic range (`dynamicRangeStops`, and per ISO setting) and, on exact requests, the mid-gray noise figure (`midGrey`); see [Sensor characteristics](#sensor-characteristics-555) |
+| `sensor` | 2.20 (#555, #535): the sensor characteristics used (pitch, peak QE, full well, read noise of each conversion gain and the dual-gain switch ISO, readout time, ADC bit depth, fixed-pattern noise), the calculated dynamic range (`dynamicRangeStops`, and per ISO setting) and, on exact requests, the mid-gray noise figure (`midGrey`); 2.22 (#595): the same for one native photosite with the EMVA 1288 and 8 MP-normalized dynamic range (`native`); see [Sensor characteristics](#sensor-characteristics-555) |
 | `exposure` | Mode, resolved settings, EV100, `isoControl` (`"manual"` or `"automatic"`) and, with `iso: "auto"`, `autoIso` (how the ISO was chosen and what limited it), metered scene luminance, the deviation from the standard exposure in stops (`standardExposureDeviationStops`), the illuminance at which the settings give the standard exposure (`standardExposureIlluminanceLux`), from 2.14 the deviation from the standard exposure of an 18 % grey card under the scene's illuminance (`greyCardDeviationStops`, incident metering) and the exposure-mode resolver's full output |
 | `scene` | Chart parameters, illuminant, distance, framed size and the scene's SHA-256 digest (it covers caller illuminant and reflectance spectra; see `sourceSha256`) |
 | `diagnostics` | Radiance query count, highlight and rendering counters, per-stage timings (`timingsMs`) and, from 2.1, the opt-ins that ran (`options`); see [Cost of each option](#cost-of-each-option) |
@@ -88,6 +88,7 @@ modelling choices (such as "no lens PSF") as `kind: "model"`.
 | `chartSize` | fill the frame | 2.3: `{ widthM, heightM?, surroundReflectance? }` places the chart at its physical size (see [Fixed-size chart](#fixed-size-chart-23)). |
 | `illuminant` | Planckian 6500 K | Also `"cie-a"`, `{ kind: "planckian", temperatureKelvin }`, `{ kind: "daylight", correlatedColorTemperatureKelvin, basis }` or `{ kind: "spectrum", spectrum }`. |
 | `illuminanceLux` | 21 481 (π·80 000/11.7) | See [Exposure](#exposure). |
+| `illumination` | off | 2.22 (#550): mixed lighting, two to eight sources (`{ illuminant, illuminanceLux }`) blended across the chart by a linear gradient, a radial falloff or regions, and point sources (`points`) with inverse-square falloff, instead of `illuminant` and `illuminanceLux`. The frame center's light sets the grey card and its strongest source the `"scene-illuminant"` white balance (see [Mixed lighting](https://github.com/Photivra/photivra/blob/main/docs/MIXED_ILLUMINATION.md)). |
 | `output.width`, `output.height` | 64 × 64 | Output pixels, 1–4094 each. The sensor has one extra site on each side for demosaicing (see `output.borderHandling`). |
 | `output.maximumPixels` | off | Opt-in large images (2.2): the most native sensor sites the capture may allocate, up to `NATIVE_RAW_LIMITS.ceilingPixels` (2^27). Sides may then reach 16 384 sites including the margin (see [Large images](#large-images)). |
 | `output.region` | whole frame | 2.3: `{ x, y, width, height }` renders that rectangle of the frame, byte-identical to the whole render (see [Region rendering](#region-rendering-23)). |
@@ -109,7 +110,7 @@ modelling choices (such as "no lens PSF") as `kind: "model"`.
 | `output.whiteBalance` | `"scene-illuminant"` | Also `"auto"`, `"none"`, `{ kelvin, tint? }` and `{ channelGains }`. |
 | `output.jpegQuantizationStep` | 2 | JPEG only: constant step 1–255, not a manufacturer quality scale. |
 | `sensor.pixelPitchMicrometers` | 6 | 0.5–1000 (2.5; 0.5–100 before). The sensor is `width × pitch` wide (see [Geometry](#geometry)); the ISO readout scales with the pixel area (see [ISO and pixel size](#iso-and-pixel-size)). Above 100 µm the capture is a downsized view of a whole frame (see [Whole-frame views](#whole-frame-views-25)). |
-| `sensor.spectralStepNanometers` | 20 | Spectral quadrature over 380–780 nm. |
+| `sensor.spectralStepNanometers` | 20; 5 on exact requests under a spiky light (2.22.0, #590) | Spectral quadrature over 380–780 nm. A light is spiky when R(20) = 20²·κ/24 > 0.01, κ = Σ\|Δ²S\| / ΣS on a 1 nm grid (the midpoint rule's error term, DLMF §3.5(i)): the fluorescent and LED representative sources. Exact requests (DNG, JPEG, PNG, `fast: false`) then take 5 nm; fast ones keep 20 nm, add the assumption `sensor.spectralSpikiness` and report `diagnostics.fast.applied["sensor.spectralStepNanometers"]: true` (exact ones `false`). An explicit step wins and adds no key. |
 | `sensor.temporalSampleCount` | 1, or 8 when the camera rotates | Midpoints across each site's exposure window. |
 | `sensor.noiseSeed` | 1 | Output is deterministic for a given seed. |
 | `sensor.antiAliasing` | off | 2.1: an optical low-pass (anti-aliasing) filter: `{ kind: "four-spot", separationSites? }`, `{ kind: "two-spot", axis, separationSites? }` or `{ kind: "point-splitting", components }` (see [Anti-aliasing filter](#anti-aliasing-filter)). Each record also takes `fast`. |
@@ -118,6 +119,7 @@ modelling choices (such as "no lens PSF") as `kind: "model"`.
 | `motion.angularVelocityRadPerSec` | still | `{ pitch?, yaw?, roll? }` in rad/s (see [Motion](#motion)). |
 | `motion.shutter` | `"global"` | Or `"rolling"`. |
 | `motion.rollingShutterSkewSeconds` | the body's readout skew | Rolling only. |
+| `motion.subject` | still subject | 2.22.0: `{ velocityMPerSec?: { x?, y? }, angularVelocityRadPerSec?, region?, fast? }`, a moving subject (see [Subject motion](#subject-motion)). |
 | `signal`, `onProgress` | none | An `AbortSignal` cancels the running task; `onProgress` reports `raw`, `white-balance`, `development`, `diffraction-correction` and `sharpening` (#571: output rows, with those options) and `jpeg` tiles. The last event of each stage has `completedTileCount === totalTileCount`. The capture yields to the host about every 8 ms with a `MessageChannel` message (else `setImmediate`, else `setTimeout`), so other queued tasks run in between: an abort sent from another task, such as a Web Worker `message` event, takes effect at the next yield. |
 
 Unknown fields are rejected, so a typo never silently falls back to a default.
@@ -442,7 +444,10 @@ independent of pixel size, so the same exposure gives the same image brightness
 - So the facade multiplies the system conversion gain (e⁻ per code) and the pre-ADC saturation
   signal of every ISO regime by the area ratio pitch²/36 µm². Codes, black level, ADC and digital gain
   are unchanged. At 6 µm the ratio is exactly 1 and nothing is changed (byte-identical to 2.1.0).
-- Read noise (in e⁻) and the synthetic dark current (1 e⁻/s) stay per pixel.
+- Above 6 µm a pixel stands for N = pitch²/36 µm² of the tier's 6 µm photosites, so (from 2.22.0, #599) its
+  read noise is the photosite's summed in quadrature, ×√N, as for a [sensor profile](#sensor-profiles-554)
+  (EMVA 1288 Release 4.0, §6.4: binning multiplies the pixel area). Below 6 µm the read noise (in e⁻) stays
+  per pixel, and the synthetic dark current (1 e⁻/s) stays per pixel at every pitch.
 
 What you see: brightness stays the same, and per-pixel noise rises as pixels shrink, because a
 smaller pixel collects fewer electrons for the same exposure (shot-noise SNR = √N_e). At f/8, 1/125 s,
@@ -487,7 +492,7 @@ r.sensor.dynamicRangeStops; r.sensor.midGrey?.snrDecibels;
 | --- | --- | --- | --- |
 | `peakQuantumEfficiency` | 0.05–0.9 | 0.5 | Multiplies every channel's EQE curve by value / 0.5: the curves keep their shape and the colour matrix, and the collected charge scales in proportion. 0.9 keeps the blue channel's peak below 1. |
 | `fullWell` | 100–1e10 e⁻ at the pitch | the body's base-ISO saturation at 6 µm, × pitch²/36 µm² | The charge at the RAW white level at base ISO (the saturation capacity of EMVA 1288 Release 4.0). `link: "pixel-area"` (default) scales `electrons`, the full well at `referencePixelPitchMicrometers` (default 6), by (pitch / reference)²; `link: "none"` uses it at any pitch. |
-| `readNoise` | 0–100 e⁻; switch ISO 100–1e7 | the body's (tiers: 2/1.1, 3/1.4 and 6/2.5 e⁻, switch at ISO 800) | Read noise of the low and the high conversion gain, and `highGainFromIso`, the ISO from which the high gain is used. A switch above the body's top ISO keeps the low gain throughout. |
+| `readNoise` | 0–100 e⁻; switch ISO 100–1e7 | the body's (tiers: 2/1.1, 3/1.4 and 6/2.5 e⁻ per 6 µm photosite, ×√N above 6 µm; switch at ISO 800) | Read noise of one physical photosite (2.22.0, tiers and profiles alike; a rendered site that sums N photosites takes it ×√N) for the low and the high conversion gain, and `highGainFromIso`, the ISO from which the high gain is used. A switch above the body's top ISO keeps the low gain throughout. |
 | `readoutSeconds` | 1e-6–1 s | the body's (8, 15, 30 ms) | The rolling shutter's top-to-bottom skew (`motion.shutter: "rolling"`; `motion.rollingShutterSkewSeconds` still overrides it). |
 | `adcBitDepth` | 12, 14, 16 | the body's (consumer 12, others 14) | The black level scales by 2^Δbits (1 024 at 14 bits → 4 096 at 16), the saturation code is 2^bits − 1 and the conversion gain follows, so the same charge reaches the same fraction of the code range. A 16-bit RAW writes a 16-bit DNG with white level 65 535. |
 | `fixedPatternNoise` | `true`/`false` | `false` | The body's synthetic fixed-pattern profile (PRNU, DSNU, hot pixels, row and column offsets), the same pattern in every capture. Not available in `simulateDisplayView`, which refuses it. |
@@ -514,6 +519,32 @@ gain) in electrons. With no read noise the range is limited by the ADC alone: lo
 The Professional tier gives 14.7 stops at ISO 100, 12.7 at ISO 800 (high gain) and 9.7 at ISO 6400;
 the Consumer tier 11.7, 10.0 and 7.0.
 
+**Native-pixel dynamic range (#595)** (`sensor.native`, from 2.22.0): the figures above are those of
+the rendered site, which at a pitch other than the body's own stands for N = (pitch / native pitch)²
+photosites (`native.photositesPerSite`). `sensor.native` gives the same figures for one native
+photosite: a generic tier's declared 6 µm reference, or a sensor profile's own pitch, with the request's
+sensor characteristics applied as at that pitch. It has the photosite's `fullWellElectrons`,
+`readNoiseElectrons`, engineering `dynamicRangeStops` and `dynamicRangeByIso`, and, on exact requests,
+its `midGrey` figure (the site's charge × 1/N). For N independent photosites summed with their read noise,
+S_sat,N = N·S_sat,0 and σ_N = √N·σ_0, so the rendered range is the native range + ½·log₂N, limited by the
+ADC, and the rendered mid-gray SNR is √N times the native one.
+
+- `native.emva1288`: EMVA 1288 Release 4.0 Linear (§2.7, Eq. 26–28, free from emva.org): the absolute
+  sensitivity threshold μ_e.min = √(σ_d² + σ_q² + 1/4) + 1/2 (σ_d the read noise, dark current excluded),
+  `dynamicRangeStops` = log₂(μ_e.sat / μ_e.min) and `dynamicRangeDecibels`, and the saturation capacity
+  per µm² (Table 2). This is the figure to show as "dynamic range (EMVA 1288)". It is slightly lower than
+  the engineering figure: 14.36 against 14.67 stops for the Professional tier at ISO 100.
+- `native.normalized8Megapixel`: **a convention, not part of EMVA 1288**, for comparing bodies of
+  different resolutions: the frame's native photosites (`framePhotosites`, the output frame's area over
+  the native photosite's) resampled to 8 million pixels, each summing framePhotosites / 8 000 000
+  photosites with their read noise (the binning rule above): DR + ½·log₂(framePhotosites / 8 000 000),
+  for the engineering and the EMVA figure. No ADC limit is applied to the sum. A 6000 × 4000 frame of
+  6 µm photosites adds 0.79 stops.
+
+All are model figures from the engine's noise model, not measurements. `sensor.readNoise` is per **physical
+photosite** on a tier and a profile alike (2.22.0): `native.readNoiseElectrons` is the value given, and a rendered site
+that sums N photosites takes it ×√N.
+
 **Noise figure** (#535, `sensor.midGrey`): computed only when the request resolves to exact (`fast:
 false`, the default for DNG, JPEG and PNG exports; about 0.3 ms); a fast request (the sRGB default)
 omits it. The expected signal-to-noise ratio of an 18 % Lambertian
@@ -530,6 +561,29 @@ SNR is 88.5 (38.9 dB); at ISO 6400 with 1/64 of the light, S ≈ 123 e⁻ and th
 the photon shot noise grows as √S, so less light means more noise whatever the ISO. The display view
 reports the same `sensor` record as its capture.
 
+### Sensor profiles (#554)
+
+`sensor.profile` takes one of eight representative sensor profile records whose values come from
+published specifications (`SENSOR_PROFILES`, or `resolveSensorProfile(id)`; unnamed, by format,
+resolution, architecture and era), checked strictly, so a modified copy also works. It
+replaces the body's ISO range, readout regimes (full well, read noise of each conversion gain, the
+dual-gain switch and the high gain's storage, ADC bit depth), readout time and fixed-pattern noise,
+and sets the default pitch and peak QE; the characteristics above still override it one at a time.
+
+```ts
+import { resolveSensorProfile, simulateCapture } from "@photivra/engine";
+const phone = await simulateCapture({ scene: { preset: "uniform" }, sensor: { profile: resolveSensorProfile("phone-50mp-stacked") } });
+phone.sensor.fullWellElectrons;   // 8 600 e⁻ at 1.197 µm
+```
+
+Each profile's base ISO is re-rated from its full well, QE and pixel area, so the same ISO reaches
+the RAW white at the same exposure on every profile (the assumption `sensor.profile` records the
+values and the rounding). The phone profile takes the catalogue's three phone lenses
+(`lens: { catalogue: "pv-24-1.8-phone" }`), and they need it. With a profile, a catalogue lens is framed on
+the profile's real sensor (the largest centred region of the output's aspect; assumption `sensor.profileView`), so it
+shows its real field of view. Values, sources, the framing and the ISO rating:
+[SENSOR_PROFILES.md](https://github.com/Photivra/photivra/blob/main/docs/SENSOR_PROFILES.md).
+
 ## Exposure
 
 The facade meters the scene like a reflected-light meter:
@@ -1689,6 +1743,48 @@ const pan = await simulateCapture({
 console.log(pan.assumptions.find((a) => a.id === "sensor.shutter")?.value, pan.exposure.settings.shutterSeconds);
 ```
 
+## Subject motion
+
+`motion.subject` (2.22.0, #551) moves the subject on the scene plane during the exposure: the whole scene, or one
+rectangle of it (`region: { centerXM, centerYM, widthM, heightM }` in metres at rest) over a still background.
+
+- `velocityMPerSec: { x, y }` is a translation in m/s on the plane (+x right, +y up as seen from the camera).
+  `angularVelocityRadPerSec` adds a rotation about an axis parallel to the optical axis, counter-clockwise as seen from
+  the camera: the optical axis for the whole scene, the region's center for a region.
+- The model is rigid motion of a planar layer. At time t from the first opening (camera rotation's time reference), a
+  subject point at rest q is at R(ωt)(q − c) + c + v·t. A plane point shows the subject where one is there, else the
+  region's rest rectangle shows a neutral 18 % background (what the subject revealed), else the still scene. The background
+  is lit by the plane's irradiance at that point: the illuminant, or with `illumination` (#550) the blend and point sources there.
+- Every scene query reads the subject at its own time. The blur is therefore the same temporal midpoint quadrature as
+  camera rotation (`sensor.temporalSampleCount`, default 8 with a moving subject), and an edge streaks over
+  v·T·m on the sensor (m = v′/D, the thin-lens image distance over the subject distance). With few samples a long streak
+  shows separate copies; raise the count. A test checks the length within sampling error.
+- A rolling shutter skews a moving subject: each row's exposure window opens later, so it sees the subject later. A test
+  checks the skew against the site windows' start offsets.
+- Panning: a camera yaw of −v/D rad/s (the engine's rotation convention) follows a subject moving at v m/s in +x. The
+  subject stays sharp and the background streaks.
+- It combines with `lens.diffraction`, `lens.defocus` (global shutter; their site lattices are formed at each midpoint), the
+  lens projection, distortion and vignetting, and camera rotation. `lens.pupilSampling` is refused.
+- Display view: sideways motion under a global shutter only. It refuses a rotating subject and a rolling shutter with a
+  reason; `simulateCapture` renders both.
+- Fast method (2.20.0 convention): with the default count and a still camera, `fast` takes one midpoint per site of the
+  streak plus one when that is fewer than 8 (taps less than one site apart). Exact keeps 8, and an explicit
+  `sensor.temporalSampleCount` always wins. Fast is the default for sRGB samples and the display view, exact for DNG,
+  JPEG and PNG.
+- Off by default: a request without `motion.subject` is byte-identical to 2.20.0.
+
+```ts
+import { simulateCapture } from "@photivra/engine";
+
+// A subject 4 mm wide moving right at 0.2 m/s, with the camera panning to follow it.
+const pan = await simulateCapture({
+  scene: { preset: "bar-groups" }, subjectDistanceM: 2, lens: { focalLengthMm: 50 },
+  motion: { angularVelocityRadPerSec: { yaw: -0.1 }, subject: { velocityMPerSec: { x: 0.2 }, region: { centerXM: 0.002, centerYM: 0, widthM: 0.004, heightM: 0.004 } } },
+  output: { width: 32, height: 16 }
+});
+console.log(pan.assumptions.find((a) => a.id === "motion.subject")?.value);
+```
+
 ## Performance
 
 Almost all of the cost is engine time per sensor site; the facade itself adds little. Since 2.0, the
@@ -2029,6 +2125,64 @@ developed image in 64-bit floats (24 bytes per pixel, 549 MiB at 24 MP), held as
 sampled every 50 ms). The extra pass therefore costs about 5 s on one worker after the parts, against about 10 s to render
 the whole frame again on one worker.
 
+### Exposure build-up (#534)
+
+`simulateExposureBuildUp(options, { fractions })` renders fractions of one exposure as one accumulating process, the way a
+camera's live readout of a long exposure shows the light building up. Each fraction is read out and developed as its own
+photo, and the fraction at 1 is `simulateCapture(options)`, byte for byte (owner decision, 2026-11-06).
+
+- **Signal.** Each fraction fⱼ below 1 is a capture of the same request in manual mode at the whole exposure's resolved
+  f-number and ISO and the shutter fⱼ·t. So its expected signal at every site is integrated from the spectrum over the window
+  [0, fⱼ·t] by the capture itself. That includes camera rotation, `motion.subject` (#551), mixed lighting (#550, which does not
+  change over time) and, with a rolling shutter, each row's own window. No per-channel signal is precomputed (owner rule).
+- **Charge.** Photon and dark-charge arrivals are Poisson processes with any rate over time. Given a count n over [0, t], the
+  count over [0, s] is Binomial(n, Λ(s)/Λ(t)), with Λ the expected count (R. G. Gallager, *Discrete Stochastic Processes*,
+  chapter 2, sections 2.4 and 2.5, MIT OpenCourseWare 6.262, free). The whole exposure keeps its ordinary draws. Going down
+  the fractions, each site's photo and dark counts are split: N(fⱼ) ~ Binomial(N(fⱼ₊₁), λ(fⱼ)/λ(fⱼ₊₁)), with λ the site's
+  expected mean in each render (after PRNU and DSNU). As a result:
+  - each fraction's counts are Poisson with that fraction's own mean, which is the mean and variance of a capture at fⱼ·t;
+  - the counts never decrease from one fraction to the next;
+  - two fractions' counts covary by the earlier one's mean.
+
+  The binomial draws are exact: inversion, and Hörmann's BTRS rejection (`src/sensor/binomial-sample.ts`).
+- **Quadrature.** The windows are nested, so the true means never decrease. The temporal midpoint quadrature can still give an
+  earlier fraction a larger estimate than a later one at a site under motion. There the count is kept (probability 1), and the
+  `exposure.buildUp` assumption counts those sites (`nonMonotoneSites`). Raise `sensor.temporalSampleCount` to remove them.
+- **Sensor.** Fixed-pattern noise (PRNU, DSNU, hot pixels, offsets) belongs to the sensor and is the same in every fraction.
+  Each fraction's charge is clipped at the full well (`sensor.fullWell`), then read out once: its own read noise (the regime's
+  components; dual gain as the ISO selects), pre-ADC and ADC saturation, and quantization at the ADC bit depth (#555; a #554
+  sensor profile's readout feeds the same stage).
+- **Cost and settings.** One capture per fraction, plus about 64 bytes per native site for the chain. The request's `fast`
+  and the format default apply to every fraction, and the build-up adds no fast method: with `fast: false` (the calculator's
+  setting) every fraction is an exact capture. `"auto"` white balance is measured on each fraction, as its own photo.
+- **Limits.** On screen only: formats `"srgb-8"` and `"srgb-16"` (a fraction's JPEG, PNG or DNG is refused). 1 to 64
+  fractions, strictly increasing, above 0 and at most 1. Each fraction's shutter fⱼ·t must be one the body can set. A
+  fraction's realization depends on the whole list, because each step of the chain has its own seeded streams
+  (`EXPOSURE_BUILD_UP_STREAM_BASE`). The same request and list always give the same frames.
+- **Result.** One frame per fraction: `{ fraction, elapsedSeconds, result }`. Below 1, `result` is the fraction capture's
+  result (`image`, `raw`, `plan` and `request` at fⱼ·t), with the build-up's counts and the assumption `exposure.buildUp`.
+  Its `exposure` and `sensor` are the whole exposure's reports: the camera's settings and noise figure, not the elapsed part's.
+
+```ts
+import { simulateExposureBuildUp } from "@photivra/engine";
+
+const build = await simulateExposureBuildUp({ scene: { preset: "slanted-edge" }, exposure: { mode: "manual", shutterSeconds: 1 / 15, iso: 400 },
+  motion: { angularVelocityRadPerSec: { yaw: 0.2 } }, output: { width: 64, height: 48 } }, { fractions: [0.2, 0.4, 0.6, 0.8, 1] });
+for (const frame of build.frames) console.log(frame.fraction, frame.elapsedSeconds, frame.result.image.samples.length);
+```
+
+The tests are in `test/exposure-build-up.test.ts`:
+
+- The last fraction equals `simulateCapture` byte for byte: tabulated and lean paths, fixed pattern, dual gain, auto white
+  balance and Auto ISO, camera rotation, rolling shutter, a moving subject and mixed lighting.
+- Each fraction's mean and variance match a single capture at f·t. Under motion this is checked column by column, and the
+  fraction is shown to differ from the whole exposure scaled by f.
+- Without read noise the codes never decrease, and Cov(c₁, c₂) = Var(c₁).
+- Dark frames' read noise is independent between readouts.
+- A fraction past the full well clips.
+
+`test/binomial-sample.test.ts` checks the binomial draws with chi-square tests of the probability mass function.
+
 ### A Print JPEG written in strips
 
 At 1:1 (output size equal to the crop), a Print JPEG's rows can be coded on separate workers.
@@ -2164,10 +2318,10 @@ the exact speed-ups, in [DISPLAY_VIEW.md](https://github.com/Photivra/photivra/b
 | 12 MP phone | Quality | 49.7 s (20×) | 106 s (3.8×) | 272 s (2.2×) |
 | 100 MP medium format | Quality | 18.2 s (3.7×) | 49.7 s (1.8×) | 126 s (1.0×) |
 
-Quality costs more on small pixels. The diffraction kernel spans several real sites, and at 1.4 µm even ISO 100 is
-noisy enough for the AHD correction to run on every sampled pixel; its correction states are kept for later views of
-the same sensor, exposure and development in the same JavaScript realm (2.6.0: a second 384-wide phone view of the same
-request took 73 s instead of 103 s). In 2.6.0 the results were byte-identical to those of the display view as first
+Quality costs more on small pixels. The diffraction kernel spans several real sites, and (2.6.0 to 2.21.0) at 1.4 µm even
+ISO 100 was noisy enough for the AHD low-signal correction to run on every sampled pixel (2.6.0: a second 384-wide phone view
+of the same request took 73 s instead of 103 s, reusing its states). 2.22.0 removed that correction, so these timings
+include work the view no longer does. In 2.6.0 the results were byte-identical to those of the display view as first
 merged, which took 222 s, 450 s and 837 s on the phone; later releases changed some views' values (for example the
 exact area mean, 2.8.0; AHD's choice, 2.13.0).