module Input Lighting

Operations relating to input lighting generation for the Enlighten runtime.

This is the original input lighting api. Note that there is a new 'extended' input lighting api available on most platforms through EnlightenInputLighting.h.

There is a considerable amount of freedom in how you choose to provide input lighting to Enlighten and the design trade-offs involved can have a huge impact on the success of Enlighten in your product. Most games are unique in some respect, so please feel free to contact Enlighten support for help in choosing the best option for your game.

There are also now two apis for providing input lighting to Enlighten. This one (the original api) and a newer 'extended' input lighting api, available through EnlightenInputLighting.h. They both use the same InputWorkspace structure and have other elements in common, but differ in how the calculation is performed, and the extended api supports additional features.

Enlighten uses a physically realistic algorithm for bouncing indirect light around a scene (subject to the tweakable albedo and bounce scale values), but makes no particular assumptions about the direct lighting model.

In practice, it's often more convenient to use a non-physical fall-off model for spot and point lights. They can be simpler to work with, reduce artefacts, and given the limitations of output displays it's often the only way to produce the desired effect.

Because of this, Enlighten allows users to specify a custom distance fall-off for point and spot light using the Enlighten::InputLightFalloffTable. Now, it is worth emphasising that there needn't be an exact correspondence between the model used for direct, on-screen lighting and the model used for producing input lighting for Enlighten. The closer the two models correspond, the more self-consistent the lighting model as a whole; on the other hand, if the Enlighten input lighting uses a completely non-physical model, it can produce counter-intuitive effects. (One example would be that the amount of light in the scene could vary wildly as a light source of constant intensity is moved closer to or further away from geometry; as soon as the model is non-physical, energy is not necessarily conserved.)

For convenience, Enlighten provides two API functions for generating suitable fall-off tables: Enlighten::GenerateUnrealCompatibleLightFalloffTable(), Enlighten::GenerateUnityCompatibleLightFalloffTable() and Enlighten::GenerateInverseSquaredLightFalloffTable(). See also the Enlighten::InputLight class.

public: bool GEO_CALL AddCachedValuesToInputWorkspace
(
    const InputWorkspace * workspaceMemory,
    InputLightingBuffer * output,
    const InputLightingBuffer * cachedValues
)

Adds light values from one input lighting buffer to another.

public: bool GEO_CALL AddDusterValuesToInputWorkspace
(
    const InputWorkspace * workspaceMemory,
    InputLightingBuffer * output,
    const Geo::v128 * dusterValues
)

Adds light values laid out in a duster block to an Enlighten input workspace.

public: Geo::u32GEO_CALL CalcInputLightingBufferSize
(
    const InputWorkspace * inputWorkspace,
    PrecisionHint::Value precisionHint
)

Tells you the amount of memory required to hold input lighting for a given system.

Required memory in bytes, 0xFFFFFFFF upon error.

public: Geo::u32GEO_CALL CalcLightVisibilitySize
(
    const InputWorkspace * workspaceMemory,
    eLightType lightType
)

Determine how much memory is required for a light visibility block.

Directional lights store a byte per cluster, all other lights store a bit per input sample (duster).

Required memory in bytes, 0xFFFFFFFF upon error.

public: Geo::u32GEO_CALL CalcLightVisibilitySize
(
    const InputWorkspace * workspaceMemory,
    Enlighten::VisibilityFormat::Type visibilityType
)

Determine how much memory is required for a light visibility block.

Directional lights store a byte per cluster, all other lights store a bit per input sample (duster).

Required memory in bytes, 0xFFFFFFFF upon error.

public: Geo::u32GEO_CALL CalcPrecomputedVisibilityWorkspaceSize
(
    const InputWorkspace * workspaceMemory,
    const PrecomputedVisibilityData * precomputedVisibilityData
)

Determine how much memory is required for a precomputed visibility workspace.

Required memory in bytes, 0xFFFFFFFF upon error.

public: bool GEO_CALL ClearInputWorkspace
(
    const InputWorkspace * workspaceMemory,
    InputLightingBuffer * buffer
)

Clear the input lighting to black.

This wipes out all bounced light for all previous iterations. Used when "starting over".

public: bool GEO_CALL CompareInputLightingBuffers
(
    Geo::Statistics & res,
    const InputLightingBuffer * bufA,
    const InputLightingBuffer * bufB
)

Compares Returns true if the lighting buffers could be compared.

public: InputLightingBuffer *GEO_CALL CreateInputLightingBuffer
(
    void * memory,
    const InputWorkspace * inputWorkspace,
    PrecisionHint::Value precisionHint
)

Construct an InputLightingBuffer inside the memory provided.

public: PrecomputedVisibilityWorkspace *GEO_CALL CreatePrecomputedVisibilityWorkspace
(
    void * memory,
    const InputWorkspace * workspaceMemory,
    const PrecomputedVisibilityData * precomputedVisibilityData
)

Construct an PrecomputedVisibilityWorkspace inside the memory provided.

public: bool GEO_CALL GenerateInverseSquaredLightFalloffTable
(
    InputLightFalloffTable * lightTable,    float lightFalloffDistance,    float lightRadius)