AutoMovie API
    Preparing search index...

    Interface IAutoMovieDirectionalLight

    Infinitely distant parallel source whose transform orientation defines direction and whose intensity has no distance falloff.

    interface IAutoMovieDirectionalLight {
        castShadow?: boolean;
        color: IAutoMovieColor;
        id: string;
        intensity: number;
        shadow?: IAutoMovieLightShadow;
        transform: IAutoMovieTransform;
        type: "directional";
    }

    Hierarchy (View Summary)

    Index
    castShadow?: boolean

    Whether this source casts a shadow map. Omitted preserves legacy output.

    What this light would cast, not what the frame renders: a scene's environment.shadows.enabled is the master switch and turning it off renders no shadow map for any light, which is also how the render budget prices it. A scene declaring no environment leaves the decision to the host renderer, exactly as it did before environments existed.

    Only the three punctual kinds can cast one. A rectangular area source is analytically integrated rather than rasterized from a light-space camera, so three.js renders no shadow map for it and the engine refuses castShadow on an IAutoMovieAreaLight instead of accepting a flag no frame would honor.

    Light color (linear).

    id: string

    Stable id.

    intensity: number

    Finite nonnegative photometric magnitude: illuminance in lux for directional, luminous intensity in candela for point/spot and luminance in nits (candela per square metre) for an area panel. Three's rectangular-area power convention is luminous flux in lumens equal to intensity times width times height times PI.

    Deterministic shadow-camera tuning, required exactly when castShadow is true.

    World placement. For directional light only the orientation matters.

    Light motion supports position for point/spot/area and rotation for directional/spot/area, through /lights/<id>/position and /lights/<id>/rotation. Other supported axes are intensity, RGB colour, point/spot range and spot cone angle. Scale, identity, shadow settings and area dimensions have no light pointer channel; the viewer still applies the authored TRS to its native light object.

    type: "directional"

    Discriminator.