 |
UDocumentation UE5.7 10.02.2026 (Source)
API documentation for Unreal Engine 5.7
|
|
UDocumentation UE5.7 10.02.2026 (Source)
API documentation for Unreal Engine 5.7
|
#include "CoreMinimal.h"#include "UObject/ObjectMacros.h"#include "BoneContainer.h"#include "BonePose.h"#include "BoneControllers/AnimNode_SkeletalControlBase.h"#include "AnimNode_ModifyBone.generated.h"Go to the source code of this file.
Classes | |
| struct | FAnimNode_ModifyBone |
Enumerations | |
| enum | EBoneModificationMode : int { UMETA =(DisplayName = "Ignore") , UMETA =(DisplayName = "Ignore") , UMETA =(DisplayName = "Ignore") } |
| enum EBoneModificationMode : int |
| Enumerator | |
|---|---|
| UMETA | The modifier ignores this channel (keeps the existing bone translation, rotation, or scale). The modifier replaces the existing translation, rotation, or scale. The modifier adds to the existing translation, rotation, or scale. Use the Skeleton's ref pose as base. Use a whole animation as a base pose. BasePoseSeq must be set. Use one frame of an animation as a base pose. BasePoseSeq and RefFrameIndex must be set (RefFrameIndex will be clamped). Use one frame of this animation. RefFrameIndex must be set (RefFrameIndex will be clamped). Character can step up onto this Component. Owning actor determines whether character can step up onto this Component (default true unless overridden in code).
Location within the HMD tracking space Location within the view space Quad layer Cylinder layer Cubemap layer Equirect layer Use a constant value. Represents stepped values. Cubic interpolation. See TangentMode for different cubic interpolation options. No interpolation. Repeat the curve without an offset. Repeat the curve with an offset relative to the first or last key's value. Sinusoidally extrapolate. Use a linearly increasing value for extrapolation. Use a constant value for extrapolation No Extrapolation User specifies the tangent as a unified tangent where the two tangents are locked to each other, presenting a consistent curve before and after. User specifies the tangent as two separate broken tangents on each side of the key which can allow a sharp change in evaluation before or after. No tangents. New Auto tangent that creates smoother curves than Auto. Don't take tangent weights into account. Only take the arrival tangent weight into account for evaluation. Only take the leaving tangent weight into account for evaluation. Take both the arrival and leaving tangent weights into account for evaluation. No keys are present All keys use constant interpolation All keys use linear interpolation All keys use cubic interpolation Keys use mixed interpolation modes Keys use weighted interpolation modes Key time is quantized to 16 bits Key time uses full precision Blueprint that is const during execution (no state graph and methods cannot modify member variables). Blueprint that serves as a container for macros to be used in other blueprints. Blueprint that serves as an interface to be implemented by other blueprints. Blueprint that handles level scripting. Blueprint that serves as a container for functions to be used in other blueprints. Add to world. Subtract from world. Special tick group that starts physics simulation. Any item that can be run in parallel with our physics simulation work. Special tick group that ends physics simulation. Any item that needs rigid body and cloth simulation to be complete before being executed. Any item that needs the update work to be done before being ticked. Catchall for anything demoted to the end. Special tick group that is not actually a tick group. After every tick group this is repeatedly re-run until there are no more newly spawned items to run. Uses project based precision mode setting Force full-precision for MaterialFloat only, no effect on shader codes in .ush/.usf All the floats are full-precision Half precision, except explict 'float' in .ush/.usf Get the sampler from the texture. Every unique texture will consume a sampler slot, which are limited in number. Shared sampler source that does not consume a sampler slot. Uses wrap addressing and gets filter mode from the world texture group. Shared sampler source that does not consume a sampler slot. Uses clamp addressing and gets filter mode from the world texture group. Shared sampler source that does not consume a sampler slot, used to sample the terrain weightmap. Gets filter mode from the terrain weightmap texture group. Lighting will be calculated for a volume, without directionality. Use this on particle effects like smoke and dust. This is the cheapest per-pixel lighting method, however the material normal is not taken into account. Lighting will be calculated for a volume, with directionality so that the normal of the material is taken into account. Note that the default particle tangent space is facing the camera, so enable bGenerateSphericalParticleNormals to get a more useful tangent space. Same as Volumetric Non Directional, but lighting is only evaluated at vertices so the pixel shader cost is significantly less. Note that lighting still comes from a volume texture, so it is limited in range. Directional lights become unshadowed in the distance. Same as Volumetric Directional, but lighting is only evaluated at vertices so the pixel shader cost is significantly less. Note that lighting still comes from a volume texture, so it is limited in range. Directional lights become unshadowed in the distance. Lighting will be calculated for a surface. The light is accumulated in a volume so the result is blurry, limited distance but the per pixel cost is very low. Use this on translucent surfaces like glass and water. Only diffuse lighting is supported. Lighting will be calculated for a surface. Use this on translucent surfaces like glass and water. This is implemented with forward shading so specular highlights from local lights are supported, however many deferred-only features are not. This is the most expensive translucency lighting method as each light's contribution is computed per-pixel. Refraction is computed based on the camera vector entering a medium whose index of refraction is defined by the Refraction material input. By default, when the root node refraction pin is unplugged, no refraction will appear. The refraction offset into Scene Color is computed based on the difference between the per-pixel normal and the per-vertex normal. By default, when the root node refraction pin is unplugged, no refraction will appear. Explicit 2D screen offset. This offset is independent of screen resolution and aspect ratio. The user is in charge of any strength and fading. Refraction is disabled. Refraction is computed based on the camera vector entering a medium whose index of refraction is defined by the material IOR evaluated from F0. The new medium's surface is defined by the material's normal. With this mode, a flat plane seen from the side will have a constant refraction offset. This is a physical model of refraction but causes reading outside the scene color texture so is a poor fit for large refractive surfaces like water. This is the pre-Substrate behavior: coverage is ignored and always 1. When rough refraction is disabled, this is behavior is forced ON. This is a new behavior available with Substrate when rough refraction are enabled: account for roughness, coverage and depth. This is a more physically based behavior: the background scene will be visible untouched according to (1-coverage), while the blurred version will be visible according to coverage. This is the legacy mode where PDO is applied differently for Depth (along View Forward) and world position (along Camera Vector). PDO is applied along the Camera Vector for Depth and World Position altogether. Number of unique shading models. Shading model will be determined by the Material Expression Graph, by utilizing the 'Shading Model' MaterialAttribute output pin. None (movement is disabled). Walking on a surface. Simplified walking on navigation data (e.g. navmesh). If GetGenerateOverlapEvents() is true, then we will perform sweeps with each navmesh move. If GetGenerateOverlapEvents() is false then movement is cheaper but characters can overlap other objects without some extra process to repel/resolve their collisions. Falling under the effects of gravity, such as after jumping or walking off the edge of a surface. Swimming through a fluid volume, under the effects of gravity and buoyancy. Flying, ignoring the effects of gravity. Affected by the current physics volume's fluid friction. User-defined custom movement mode, including many possible sub-modes. Reserved for gizmo collision Add new serializeable channels above here (i.e. entries that exist in FCollisionResponseContainer) Add only nonserialized/transient flags below Returns both overlaps with both dynamic and static components returns only overlaps with dynamic actors (far fewer results in practice, much more efficient) returns only overlaps with static actors (fewer results, more efficient) No role at all. Locally simulated proxy of this actor. Locally autonomous proxy of this actor. Authoritative control over the actor. This actor can never go network dormant. This actor can go dormant, but is not currently dormant. Game code will tell it when it go dormant. This actor wants to go fully dormant for all connections. This actor may want to go dormant for some connections, GetNetDormancy() will be called to find out which. This actor is initially dormant for all connection if it was placed in map. Don't affect the walkable slope. Walkable slope angle will be ignored. Increase walkable slope. Makes it easier to walk up a surface, by allowing traversal over higher-than-usual angles. Decrease walkable slope. Makes it harder to walk up a surface, by restricting traversal to lower-than-usual angles. Make surface unwalkable. Note: WalkableSlopeAngle will be ignored. requires compute shader to construct 64 bin histogram faster method that computes single value by downsampling Uses camera settings. Sum of Gaussian formulation Fast Fourier Transform Image based convolution, intended for cinematics (too expensive for games) Obsolete - kept for backwards compatibility. No compression, no mips. Using this TextureGroup triggers special mip map generation code only useful for the BokehDOF post process. No compression, created on import of a .IES file. Non-filtered, useful for 2D rendering. Hierarchical LOD generated textures Impostor Color Textures Impostor Normal and Depth, use default compression 8 bit data stored in textures 16 bit data stored in textures Project specific group, rename in Engine.ini, [EnumRemap] TEXTUREGROUP_Project**.DisplayName=My Fun Group Default for the "texture". 2x2 average, default for the "texture group". 8x8 with sharpening: 0=no sharpening but better quality which is softer, 1=little, 5=medium, 10=extreme. Do not touch existing mip chain as it contains generated data. Blur further (useful for image based reflections). Use the first texel of each 2x2 (or 2x2x2) group. Introduce significant amount of blur using angular filtering (only applies to cubemaps, useful for ambient lighting). CompositingTexture needs to be a normal map with the same or larger size. Get the tiling setting from the texture's group CookPlatformTilingDisabled setting. By default it's to tile during cook, unless it has been changed in the texture group The texture will be tiled during the cook process if the platform supports it. The texture will not be tiled during the cook process, and will be tiled when uploaded to the GPU if the platform supports it. Use setting from the Texture Group. Lighting samples are computed in an adaptive grid which covers the entire Lightmass Importance Volume. Higher density grids are used near geometry. The Volumetric Lightmap is interpolated efficiently on the GPU per-pixel, allowing accurate indirect lighting for dynamic objects and volumetric fog. Positions outside of the Importance Volume reuse the border texels of the Volumetric Lightmap (clamp addressing). On mobile, interpolation is done on the CPU at the center of each object's bounds. Volume lighting samples are placed on top of static surfaces at medium density, and everywhere else in the Lightmass Importance Volume at low density. Positions outside of the Importance Volume will have no indirect lighting. This method requires CPU interpolation so the Indirect Lighting Cache is used to interpolate results for each dynamic object, adding Rendering Thread overhead. UserSet - use the user set value. Primarily intended for weapon effects. Emitter - use the emitter position as the source/target. Particle - use the particles from a given emitter in the system. Actor - use the actor as the source/target. The name of the actor should be set in <Source/Target>Name. Direct - a direct line between source and target. UserSet - use the user set value. Distribution - use the distribution. Emitter - use the emitter direction. Modify the target of the beam. Freeze the particle in place Stop collision checks, but keep updating Stop translations of the particle Stop rotations of the particle Stop all movement of the particle Multiply the module values by the previous results 'Break' the chain and apply the values from the previous results Lock the sprite facing towards the positive X-axis Lock the sprite facing towards the positive Y-axis Lock the sprite facing towards the positive Z-axis Lock the sprite facing towards the negative X-axis Lock the sprite facing towards the negative Y-axis Lock the sprite facing towards the negative Z-axis Lock the sprite rotation on the X-axis Lock the sprite rotation on the Y-axis Lock the sprite rotation on the Z-axis TypeData - TypeData modules Beam - only applied to beam emitters Trail - only applied to trail emitters Spawn - all emitter types REQUIRE it Required - all emitter types REQUIRE it Event - event related modules Light related modules SubUV related modules Random - select a particle at random Sequential - select a particle in order FixedTime - update via a fixed time step Don't determine occlusion on this particle system Use the bounds of the particle system component when determining occlusion Use the custom occlusion bounds when determining occlusion Replay system is disabled. Particles are simulated and rendered normally. Capture all particle data to the clip specified by ReplayClipIDNumber. The frame to capture must be specified using the ReplayFrameIndex Replay captured particle state from the clip specified by ReplayClipIDNumber. The frame to play must be specified using the ReplayFrameIndex Any - allow any event Spawn - a particle spawn event Death - a particle death event Collision - a particle collision event Burst - a particle burst event Blueprint - an event generated by level script Set absolute position of bone in world space. Set position of bone in SkeletalMeshComponent's reference frame. Set position of bone relative to parent bone. Set position of bone in its own reference frame. Don't change rotation at all. Keep forward direction vector relative to the parent bone. Copy rotation of target to bone. No additive. The material's attributes describe a deferred decal, and will be mapped onto the decal's frustum. The material's attributes describe a light's distribution. The material's attributes describe a 3d volume. The material will be used in a custom post process pass. The material will be used for UMG or Slate UI The material will be used for runtime virtual texture (Deprecated). Modulate BaseColor, blend rest, updating the GBuffer, does not work for baked lighting. Does not work in DBuffer mode (approximated as Translucent). Only blend normal, updating the GBuffer, does not work for baked lighting. Additive emissive only. Put into DBuffer to work for baked lighting as well (becomes DBM_TranslucentNormal if normal is not hooked up). Put into DBuffer to work for baked lighting as well. Put into DBuffer to work for baked lighting as well (becomes DBM_DBuffer_Color if normal is not hooked up). Put into DBuffer to work for baked lighting as well (becomes DBM_DBuffer_Roughness if normal is not hooked up). Internal DBffer decal blend modes used for auto-converted decals Output signed distance in Opacity depending on LightVector. Note: Can be costly, no shadow casting but receiving, no per pixel normal yet, no quality settings yet Blend with existing scene color. Decal color is already pre-multiplied by alpha. Ambient occlusion. Do not receive decals (Later we still can read the DBuffer channels to customize the effect, this frees up some interpolators). Receive Decals, applies all DBuffer channels. Receive Decals, applies color DBuffer channels. Receive Decals, applies color and normal DBuffer channels. Receive Decals, applies color, roughness, specular and metallic DBuffer channels. Receive Decals, applies normal DBuffer channels. Receive Decals, applies normal, roughness, specular and metallic DBuffer channels. Receive Decals, applies roughness, specular and metallic DBuffer channels. Render before depth of field. Render after depth of field. Render after motion blur. Disables depth test (the reconstruction post MB and TSR would otherwise flicker due to the TSR camera jittering making the depth buffer unstable). Because of that Lumen high quality reflections is also disabled to avoid visual discrepancy at depth intersection. Object local Skinned local DBuffer B (World Normal). RGB is pre-multiplied by alpha. A is one minus alpha. DBuffer C (Roughness). R is Roughness, G is Metallic, B is Specular, all pre-multiplied by alpha. A is one minus alpha. Buffer index UV in LayerId (batch) x Channels x Height x Wdith High quality for direct use and bumps Non-tiled: ~61 instructions per level, 8 texture lookups Tiling: ~74 instructions per level, 8 texture lookups Even "non-tiled" mode has a repeat of 128. Useful Repeat Size range <= 128 Formerly labeled as Perlin noise High quality for direct use, BAD for bumps; doesn't work on Mobile ~16 instructions per level, 1 texture lookup Always tiles with a repeat of 16, "Tiling" mode is not an option for Fast Gradient noise High quality for direct use and bumps Non-tiled: ~80 instructions per level, no textures Tiling: ~143 instructions per level, no textures Low quality, but pure computation Non-tiled: ~53 instructions per level, no textures Tiling: ~118 instructions per level, no textures Formerly mis-labeled as Gradient noise Also known as Worley or Cellular noise Quality=1 searches 8 cells, Quality=2 searches 16 cells Quality=3 searches 27 cells, Quality=4 searches 32 cells All are about 20 instructions per cell searched Normal (Path Tracing). Stores the denoised radiance if denoising is turned on and complete for the current frame, otherwise, black. Albedo (Path Tracing). Average albedo at the current sample count. Normal (Path Tracing). Average normal at the current sample count. Variance (Path Tracing). Path tracing variance stored as standard derivation. Variance can be per channel variance or variance of luminance, albedo, and normal based on the path tracing configuration. Hooking up this buffer can increase additional cost. Local space (relative to the rendered object, = object space) World space, a unit is 1cm View space (relative to the camera/eye, = camera space, differs from camera space in the shadow passes) Camera space Particle space, deprecated value will be removed in a future release use instance space. Instance space (used to provide per instance transform, i.e. for Instanced Static Mesh / Particles). Tangent space (relative to the surface) Absolute world space Like absolute world space, but the world origin is moved to the center of the tile the camera is in. Logically similar to Translated world space, i.e. world space rotation and scale but with a position relative to the camera First person "space", which can be thought of as a transform that is applied to a position in translated world space. View space (differs from camera space in the shadow passes) Perlin-style noise with 3D vector/color output. RGB output range -1 to 1 R only = ~83 instructions, RGB = ~125 instructions Gradient of Perlin noise, useful for bumps. RGB = Gradient of scalar noise (signed 3D vector) A = Base scalar noise with range -1 to 1 A only = ~83 instructions, RGBA = ~106 instructions Curl of Perlin noise, useful for 3D flow directions. RGB = signed curl vector ~162 instructions Also known as Worley or Cellular noise. RGB = position of closest point at center of Voronoi cell A = distance to closest point with range 0 to about 4 Quality levels 1-4 search 8, 16, 27 & 32 cells All ~20 instructions per cell searched Horizontal and vertical field of view angles in radian Tan(FieldOfView * 0.5) Horizontal and vertical size of the view in pixels Absolute world space view position (differs from the camera position in the shadow passes) Absolute world space camera position Horizontal and vertical position of the viewport in pixels within the buffer. Number of temporal AA sample used across multiple to converge to anti aliased output. Index of the Temporal AA jitter for this frame. Offset of the temporal sample for this frame in pixel size. Mip Level that Runtime Virtual Texture Output is rendering to. World space derivatives for Runtime Virtual Texture Output. Pre Exposure Maximum mip level of Runtime Virtual Texture that Runtime Virtual Texture Output is rendering to. Screen percentage at which the rendering resolution happens, to allow tech-art to remain consistent with dynamic resolution. Post process volume user flags, useful for varying the behavior of a material per view Horizontal and vertical first person field of view angles in radian Tan(FirstPersonFieldOfView * 0.5) Scaling factor applied to first person primitives to keep them from intersecting with the scene. Near plane. Absolute world position with no material shader offsets applied Camera relative world position with all material shader offsets applied Camera relative world position with no material shader offsets applied Scene depth, single channel, contains the linear depth of the opaque objects Material diffuse, RGB color (computed from GBuffer) Material specular, RGB color (computed from GBuffer) Material subsurface, RGB color (GBuffer, only for some ShadingModels) Material base, RGB color (GBuffer), can be modified on read by the ShadingModel, consider StoredBasedColor Material specular, single channel (GBuffer), can be modified on read by the ShadingModel, consider StoredSpecular Material metallic, single channel (GBuffer) Normal, RGB in -1..1 range, not normalized (GBuffer) Not yet supported Material opacity, single channel (GBuffer) Material roughness, single channel (GBuffer) Material ambient occlusion, single channel (GBuffer) Scene depth, single channel, contains the linear depth of the opaque objects rendered with CustomDepth (mesh property) Input #0 of this postprocess pass, usually the only one hooked up Input #1 of this postprocess pass, usually not used Input #2 of this postprocess pass, usually not used Input #3 of this postprocess pass, usually not used Input #4 of this postprocess pass, usually not used Input #5 of this postprocess pass, usually not used Input #6 of this postprocess pass, usually not used Decal Mask, single bit (was moved to stencil for better performance, not accessible at the moment) Shading model Shading model ID Ambient Occlusion, single channel Scene stencil, contains CustomStencil mesh property of the opaque objects rendered with CustomDepth Material base, RGB color (GBuffer) Material specular, single channel (GBuffer) Scene Velocity Tangent, RGB in -1..1 range, not normalized (GBuffer) Material anisotropy, single channel (GBuffer) Whether the pixel is marked as belonging to an opaque first person primitive, single channel (GBuffer) Only supported with forward shading. MSAA sample count is controlled by r.MSAACount. Looped playback mode. Play all movies in the play list in order then start over until manually canceled Alternate Looped mode. Play all of the movies in the play list and loop just the last movie until loading is finished. Create both simple and complex shapes. Simple shapes are used for regular scene queries and collision tests. Complex shape (per poly) is used for complex scene queries. Create only simple shapes. Use simple shapes for all scene queries and collision tests. Create only complex shapes (per poly). Use complex shapes for all scene queries and collision tests. Can be used in simulation for static shapes only (i.e can be collided against but not moved through forces or velocity.) Follow owner. Do not follow owner, but make kinematic. Do not follow owner, but simulate. No constraint against this axis. Limited freedom along this axis. Fully constraint against this axis. Left-align. Center-align. Right-align. Fill the entire height. Top-align. Bottom-align. Place the menu immediately below the anchor Place the menu immediately centered below the anchor Place the menu immediately below the anchor aligned to the right of the content Place the menu immediately below the anchor and match is width to the anchor's content Place the menu immediately below the anchor and match is width to the anchor's content. If the width overflows, align with the right edge of the anchor. Place the menu to the right of the anchor Place the menu immediately above the anchor, no transition effect Place the menu immediately centered above the anchor, no transition effect Place the menu immediately above the anchor aligned to the right of the content Place the menu to the left of the anchor Place the menu's center on top of the menu anchor's center point Place the menu's vertical center on the left side at the menu anchor's vertical center on the right side Place the menu's bottom left corner directly on top of the menu anchor's bottom left corner Orient horizontally, i.e. left to right. Orient vertically, i.e. top to bottom. Scroll down. Scroll up. |
| UMETA | The modifier ignores this channel (keeps the existing bone translation, rotation, or scale). The modifier replaces the existing translation, rotation, or scale. The modifier adds to the existing translation, rotation, or scale. Use the Skeleton's ref pose as base. Use a whole animation as a base pose. BasePoseSeq must be set. Use one frame of an animation as a base pose. BasePoseSeq and RefFrameIndex must be set (RefFrameIndex will be clamped). Use one frame of this animation. RefFrameIndex must be set (RefFrameIndex will be clamped). Character can step up onto this Component. Owning actor determines whether character can step up onto this Component (default true unless overridden in code).
Location within the HMD tracking space Location within the view space Quad layer Cylinder layer Cubemap layer Equirect layer Use a constant value. Represents stepped values. Cubic interpolation. See TangentMode for different cubic interpolation options. No interpolation. Repeat the curve without an offset. Repeat the curve with an offset relative to the first or last key's value. Sinusoidally extrapolate. Use a linearly increasing value for extrapolation. Use a constant value for extrapolation No Extrapolation User specifies the tangent as a unified tangent where the two tangents are locked to each other, presenting a consistent curve before and after. User specifies the tangent as two separate broken tangents on each side of the key which can allow a sharp change in evaluation before or after. No tangents. New Auto tangent that creates smoother curves than Auto. Don't take tangent weights into account. Only take the arrival tangent weight into account for evaluation. Only take the leaving tangent weight into account for evaluation. Take both the arrival and leaving tangent weights into account for evaluation. No keys are present All keys use constant interpolation All keys use linear interpolation All keys use cubic interpolation Keys use mixed interpolation modes Keys use weighted interpolation modes Key time is quantized to 16 bits Key time uses full precision Blueprint that is const during execution (no state graph and methods cannot modify member variables). Blueprint that serves as a container for macros to be used in other blueprints. Blueprint that serves as an interface to be implemented by other blueprints. Blueprint that handles level scripting. Blueprint that serves as a container for functions to be used in other blueprints. Add to world. Subtract from world. Special tick group that starts physics simulation. Any item that can be run in parallel with our physics simulation work. Special tick group that ends physics simulation. Any item that needs rigid body and cloth simulation to be complete before being executed. Any item that needs the update work to be done before being ticked. Catchall for anything demoted to the end. Special tick group that is not actually a tick group. After every tick group this is repeatedly re-run until there are no more newly spawned items to run. Uses project based precision mode setting Force full-precision for MaterialFloat only, no effect on shader codes in .ush/.usf All the floats are full-precision Half precision, except explict 'float' in .ush/.usf Get the sampler from the texture. Every unique texture will consume a sampler slot, which are limited in number. Shared sampler source that does not consume a sampler slot. Uses wrap addressing and gets filter mode from the world texture group. Shared sampler source that does not consume a sampler slot. Uses clamp addressing and gets filter mode from the world texture group. Shared sampler source that does not consume a sampler slot, used to sample the terrain weightmap. Gets filter mode from the terrain weightmap texture group. Lighting will be calculated for a volume, without directionality. Use this on particle effects like smoke and dust. This is the cheapest per-pixel lighting method, however the material normal is not taken into account. Lighting will be calculated for a volume, with directionality so that the normal of the material is taken into account. Note that the default particle tangent space is facing the camera, so enable bGenerateSphericalParticleNormals to get a more useful tangent space. Same as Volumetric Non Directional, but lighting is only evaluated at vertices so the pixel shader cost is significantly less. Note that lighting still comes from a volume texture, so it is limited in range. Directional lights become unshadowed in the distance. Same as Volumetric Directional, but lighting is only evaluated at vertices so the pixel shader cost is significantly less. Note that lighting still comes from a volume texture, so it is limited in range. Directional lights become unshadowed in the distance. Lighting will be calculated for a surface. The light is accumulated in a volume so the result is blurry, limited distance but the per pixel cost is very low. Use this on translucent surfaces like glass and water. Only diffuse lighting is supported. Lighting will be calculated for a surface. Use this on translucent surfaces like glass and water. This is implemented with forward shading so specular highlights from local lights are supported, however many deferred-only features are not. This is the most expensive translucency lighting method as each light's contribution is computed per-pixel. Refraction is computed based on the camera vector entering a medium whose index of refraction is defined by the Refraction material input. By default, when the root node refraction pin is unplugged, no refraction will appear. The refraction offset into Scene Color is computed based on the difference between the per-pixel normal and the per-vertex normal. By default, when the root node refraction pin is unplugged, no refraction will appear. Explicit 2D screen offset. This offset is independent of screen resolution and aspect ratio. The user is in charge of any strength and fading. Refraction is disabled. Refraction is computed based on the camera vector entering a medium whose index of refraction is defined by the material IOR evaluated from F0. The new medium's surface is defined by the material's normal. With this mode, a flat plane seen from the side will have a constant refraction offset. This is a physical model of refraction but causes reading outside the scene color texture so is a poor fit for large refractive surfaces like water. This is the pre-Substrate behavior: coverage is ignored and always 1. When rough refraction is disabled, this is behavior is forced ON. This is a new behavior available with Substrate when rough refraction are enabled: account for roughness, coverage and depth. This is a more physically based behavior: the background scene will be visible untouched according to (1-coverage), while the blurred version will be visible according to coverage. This is the legacy mode where PDO is applied differently for Depth (along View Forward) and world position (along Camera Vector). PDO is applied along the Camera Vector for Depth and World Position altogether. Number of unique shading models. Shading model will be determined by the Material Expression Graph, by utilizing the 'Shading Model' MaterialAttribute output pin. None (movement is disabled). Walking on a surface. Simplified walking on navigation data (e.g. navmesh). If GetGenerateOverlapEvents() is true, then we will perform sweeps with each navmesh move. If GetGenerateOverlapEvents() is false then movement is cheaper but characters can overlap other objects without some extra process to repel/resolve their collisions. Falling under the effects of gravity, such as after jumping or walking off the edge of a surface. Swimming through a fluid volume, under the effects of gravity and buoyancy. Flying, ignoring the effects of gravity. Affected by the current physics volume's fluid friction. User-defined custom movement mode, including many possible sub-modes. Reserved for gizmo collision Add new serializeable channels above here (i.e. entries that exist in FCollisionResponseContainer) Add only nonserialized/transient flags below Returns both overlaps with both dynamic and static components returns only overlaps with dynamic actors (far fewer results in practice, much more efficient) returns only overlaps with static actors (fewer results, more efficient) No role at all. Locally simulated proxy of this actor. Locally autonomous proxy of this actor. Authoritative control over the actor. This actor can never go network dormant. This actor can go dormant, but is not currently dormant. Game code will tell it when it go dormant. This actor wants to go fully dormant for all connections. This actor may want to go dormant for some connections, GetNetDormancy() will be called to find out which. This actor is initially dormant for all connection if it was placed in map. Don't affect the walkable slope. Walkable slope angle will be ignored. Increase walkable slope. Makes it easier to walk up a surface, by allowing traversal over higher-than-usual angles. Decrease walkable slope. Makes it harder to walk up a surface, by restricting traversal to lower-than-usual angles. Make surface unwalkable. Note: WalkableSlopeAngle will be ignored. requires compute shader to construct 64 bin histogram faster method that computes single value by downsampling Uses camera settings. Sum of Gaussian formulation Fast Fourier Transform Image based convolution, intended for cinematics (too expensive for games) Obsolete - kept for backwards compatibility. No compression, no mips. Using this TextureGroup triggers special mip map generation code only useful for the BokehDOF post process. No compression, created on import of a .IES file. Non-filtered, useful for 2D rendering. Hierarchical LOD generated textures Impostor Color Textures Impostor Normal and Depth, use default compression 8 bit data stored in textures 16 bit data stored in textures Project specific group, rename in Engine.ini, [EnumRemap] TEXTUREGROUP_Project**.DisplayName=My Fun Group Default for the "texture". 2x2 average, default for the "texture group". 8x8 with sharpening: 0=no sharpening but better quality which is softer, 1=little, 5=medium, 10=extreme. Do not touch existing mip chain as it contains generated data. Blur further (useful for image based reflections). Use the first texel of each 2x2 (or 2x2x2) group. Introduce significant amount of blur using angular filtering (only applies to cubemaps, useful for ambient lighting). CompositingTexture needs to be a normal map with the same or larger size. Get the tiling setting from the texture's group CookPlatformTilingDisabled setting. By default it's to tile during cook, unless it has been changed in the texture group The texture will be tiled during the cook process if the platform supports it. The texture will not be tiled during the cook process, and will be tiled when uploaded to the GPU if the platform supports it. Use setting from the Texture Group. Lighting samples are computed in an adaptive grid which covers the entire Lightmass Importance Volume. Higher density grids are used near geometry. The Volumetric Lightmap is interpolated efficiently on the GPU per-pixel, allowing accurate indirect lighting for dynamic objects and volumetric fog. Positions outside of the Importance Volume reuse the border texels of the Volumetric Lightmap (clamp addressing). On mobile, interpolation is done on the CPU at the center of each object's bounds. Volume lighting samples are placed on top of static surfaces at medium density, and everywhere else in the Lightmass Importance Volume at low density. Positions outside of the Importance Volume will have no indirect lighting. This method requires CPU interpolation so the Indirect Lighting Cache is used to interpolate results for each dynamic object, adding Rendering Thread overhead. UserSet - use the user set value. Primarily intended for weapon effects. Emitter - use the emitter position as the source/target. Particle - use the particles from a given emitter in the system. Actor - use the actor as the source/target. The name of the actor should be set in <Source/Target>Name. Direct - a direct line between source and target. UserSet - use the user set value. Distribution - use the distribution. Emitter - use the emitter direction. Modify the target of the beam. Freeze the particle in place Stop collision checks, but keep updating Stop translations of the particle Stop rotations of the particle Stop all movement of the particle Multiply the module values by the previous results 'Break' the chain and apply the values from the previous results Lock the sprite facing towards the positive X-axis Lock the sprite facing towards the positive Y-axis Lock the sprite facing towards the positive Z-axis Lock the sprite facing towards the negative X-axis Lock the sprite facing towards the negative Y-axis Lock the sprite facing towards the negative Z-axis Lock the sprite rotation on the X-axis Lock the sprite rotation on the Y-axis Lock the sprite rotation on the Z-axis TypeData - TypeData modules Beam - only applied to beam emitters Trail - only applied to trail emitters Spawn - all emitter types REQUIRE it Required - all emitter types REQUIRE it Event - event related modules Light related modules SubUV related modules Random - select a particle at random Sequential - select a particle in order FixedTime - update via a fixed time step Don't determine occlusion on this particle system Use the bounds of the particle system component when determining occlusion Use the custom occlusion bounds when determining occlusion Replay system is disabled. Particles are simulated and rendered normally. Capture all particle data to the clip specified by ReplayClipIDNumber. The frame to capture must be specified using the ReplayFrameIndex Replay captured particle state from the clip specified by ReplayClipIDNumber. The frame to play must be specified using the ReplayFrameIndex Any - allow any event Spawn - a particle spawn event Death - a particle death event Collision - a particle collision event Burst - a particle burst event Blueprint - an event generated by level script Set absolute position of bone in world space. Set position of bone in SkeletalMeshComponent's reference frame. Set position of bone relative to parent bone. Set position of bone in its own reference frame. Don't change rotation at all. Keep forward direction vector relative to the parent bone. Copy rotation of target to bone. No additive. The material's attributes describe a deferred decal, and will be mapped onto the decal's frustum. The material's attributes describe a light's distribution. The material's attributes describe a 3d volume. The material will be used in a custom post process pass. The material will be used for UMG or Slate UI The material will be used for runtime virtual texture (Deprecated). Modulate BaseColor, blend rest, updating the GBuffer, does not work for baked lighting. Does not work in DBuffer mode (approximated as Translucent). Only blend normal, updating the GBuffer, does not work for baked lighting. Additive emissive only. Put into DBuffer to work for baked lighting as well (becomes DBM_TranslucentNormal if normal is not hooked up). Put into DBuffer to work for baked lighting as well. Put into DBuffer to work for baked lighting as well (becomes DBM_DBuffer_Color if normal is not hooked up). Put into DBuffer to work for baked lighting as well (becomes DBM_DBuffer_Roughness if normal is not hooked up). Internal DBffer decal blend modes used for auto-converted decals Output signed distance in Opacity depending on LightVector. Note: Can be costly, no shadow casting but receiving, no per pixel normal yet, no quality settings yet Blend with existing scene color. Decal color is already pre-multiplied by alpha. Ambient occlusion. Do not receive decals (Later we still can read the DBuffer channels to customize the effect, this frees up some interpolators). Receive Decals, applies all DBuffer channels. Receive Decals, applies color DBuffer channels. Receive Decals, applies color and normal DBuffer channels. Receive Decals, applies color, roughness, specular and metallic DBuffer channels. Receive Decals, applies normal DBuffer channels. Receive Decals, applies normal, roughness, specular and metallic DBuffer channels. Receive Decals, applies roughness, specular and metallic DBuffer channels. Render before depth of field. Render after depth of field. Render after motion blur. Disables depth test (the reconstruction post MB and TSR would otherwise flicker due to the TSR camera jittering making the depth buffer unstable). Because of that Lumen high quality reflections is also disabled to avoid visual discrepancy at depth intersection. Object local Skinned local DBuffer B (World Normal). RGB is pre-multiplied by alpha. A is one minus alpha. DBuffer C (Roughness). R is Roughness, G is Metallic, B is Specular, all pre-multiplied by alpha. A is one minus alpha. Buffer index UV in LayerId (batch) x Channels x Height x Wdith High quality for direct use and bumps Non-tiled: ~61 instructions per level, 8 texture lookups Tiling: ~74 instructions per level, 8 texture lookups Even "non-tiled" mode has a repeat of 128. Useful Repeat Size range <= 128 Formerly labeled as Perlin noise High quality for direct use, BAD for bumps; doesn't work on Mobile ~16 instructions per level, 1 texture lookup Always tiles with a repeat of 16, "Tiling" mode is not an option for Fast Gradient noise High quality for direct use and bumps Non-tiled: ~80 instructions per level, no textures Tiling: ~143 instructions per level, no textures Low quality, but pure computation Non-tiled: ~53 instructions per level, no textures Tiling: ~118 instructions per level, no textures Formerly mis-labeled as Gradient noise Also known as Worley or Cellular noise Quality=1 searches 8 cells, Quality=2 searches 16 cells Quality=3 searches 27 cells, Quality=4 searches 32 cells All are about 20 instructions per cell searched Normal (Path Tracing). Stores the denoised radiance if denoising is turned on and complete for the current frame, otherwise, black. Albedo (Path Tracing). Average albedo at the current sample count. Normal (Path Tracing). Average normal at the current sample count. Variance (Path Tracing). Path tracing variance stored as standard derivation. Variance can be per channel variance or variance of luminance, albedo, and normal based on the path tracing configuration. Hooking up this buffer can increase additional cost. Local space (relative to the rendered object, = object space) World space, a unit is 1cm View space (relative to the camera/eye, = camera space, differs from camera space in the shadow passes) Camera space Particle space, deprecated value will be removed in a future release use instance space. Instance space (used to provide per instance transform, i.e. for Instanced Static Mesh / Particles). Tangent space (relative to the surface) Absolute world space Like absolute world space, but the world origin is moved to the center of the tile the camera is in. Logically similar to Translated world space, i.e. world space rotation and scale but with a position relative to the camera First person "space", which can be thought of as a transform that is applied to a position in translated world space. View space (differs from camera space in the shadow passes) Perlin-style noise with 3D vector/color output. RGB output range -1 to 1 R only = ~83 instructions, RGB = ~125 instructions Gradient of Perlin noise, useful for bumps. RGB = Gradient of scalar noise (signed 3D vector) A = Base scalar noise with range -1 to 1 A only = ~83 instructions, RGBA = ~106 instructions Curl of Perlin noise, useful for 3D flow directions. RGB = signed curl vector ~162 instructions Also known as Worley or Cellular noise. RGB = position of closest point at center of Voronoi cell A = distance to closest point with range 0 to about 4 Quality levels 1-4 search 8, 16, 27 & 32 cells All ~20 instructions per cell searched Horizontal and vertical field of view angles in radian Tan(FieldOfView * 0.5) Horizontal and vertical size of the view in pixels Absolute world space view position (differs from the camera position in the shadow passes) Absolute world space camera position Horizontal and vertical position of the viewport in pixels within the buffer. Number of temporal AA sample used across multiple to converge to anti aliased output. Index of the Temporal AA jitter for this frame. Offset of the temporal sample for this frame in pixel size. Mip Level that Runtime Virtual Texture Output is rendering to. World space derivatives for Runtime Virtual Texture Output. Pre Exposure Maximum mip level of Runtime Virtual Texture that Runtime Virtual Texture Output is rendering to. Screen percentage at which the rendering resolution happens, to allow tech-art to remain consistent with dynamic resolution. Post process volume user flags, useful for varying the behavior of a material per view Horizontal and vertical first person field of view angles in radian Tan(FirstPersonFieldOfView * 0.5) Scaling factor applied to first person primitives to keep them from intersecting with the scene. Near plane. Absolute world position with no material shader offsets applied Camera relative world position with all material shader offsets applied Camera relative world position with no material shader offsets applied Scene depth, single channel, contains the linear depth of the opaque objects Material diffuse, RGB color (computed from GBuffer) Material specular, RGB color (computed from GBuffer) Material subsurface, RGB color (GBuffer, only for some ShadingModels) Material base, RGB color (GBuffer), can be modified on read by the ShadingModel, consider StoredBasedColor Material specular, single channel (GBuffer), can be modified on read by the ShadingModel, consider StoredSpecular Material metallic, single channel (GBuffer) Normal, RGB in -1..1 range, not normalized (GBuffer) Not yet supported Material opacity, single channel (GBuffer) Material roughness, single channel (GBuffer) Material ambient occlusion, single channel (GBuffer) Scene depth, single channel, contains the linear depth of the opaque objects rendered with CustomDepth (mesh property) Input #0 of this postprocess pass, usually the only one hooked up Input #1 of this postprocess pass, usually not used Input #2 of this postprocess pass, usually not used Input #3 of this postprocess pass, usually not used Input #4 of this postprocess pass, usually not used Input #5 of this postprocess pass, usually not used Input #6 of this postprocess pass, usually not used Decal Mask, single bit (was moved to stencil for better performance, not accessible at the moment) Shading model Shading model ID Ambient Occlusion, single channel Scene stencil, contains CustomStencil mesh property of the opaque objects rendered with CustomDepth Material base, RGB color (GBuffer) Material specular, single channel (GBuffer) Scene Velocity Tangent, RGB in -1..1 range, not normalized (GBuffer) Material anisotropy, single channel (GBuffer) Whether the pixel is marked as belonging to an opaque first person primitive, single channel (GBuffer) Only supported with forward shading. MSAA sample count is controlled by r.MSAACount. Looped playback mode. Play all movies in the play list in order then start over until manually canceled Alternate Looped mode. Play all of the movies in the play list and loop just the last movie until loading is finished. Create both simple and complex shapes. Simple shapes are used for regular scene queries and collision tests. Complex shape (per poly) is used for complex scene queries. Create only simple shapes. Use simple shapes for all scene queries and collision tests. Create only complex shapes (per poly). Use complex shapes for all scene queries and collision tests. Can be used in simulation for static shapes only (i.e can be collided against but not moved through forces or velocity.) Follow owner. Do not follow owner, but make kinematic. Do not follow owner, but simulate. No constraint against this axis. Limited freedom along this axis. Fully constraint against this axis. Left-align. Center-align. Right-align. Fill the entire height. Top-align. Bottom-align. Place the menu immediately below the anchor Place the menu immediately centered below the anchor Place the menu immediately below the anchor aligned to the right of the content Place the menu immediately below the anchor and match is width to the anchor's content Place the menu immediately below the anchor and match is width to the anchor's content. If the width overflows, align with the right edge of the anchor. Place the menu to the right of the anchor Place the menu immediately above the anchor, no transition effect Place the menu immediately centered above the anchor, no transition effect Place the menu immediately above the anchor aligned to the right of the content Place the menu to the left of the anchor Place the menu's center on top of the menu anchor's center point Place the menu's vertical center on the left side at the menu anchor's vertical center on the right side Place the menu's bottom left corner directly on top of the menu anchor's bottom left corner Orient horizontally, i.e. left to right. Orient vertically, i.e. top to bottom. Scroll down. Scroll up. |
| UMETA | The modifier ignores this channel (keeps the existing bone translation, rotation, or scale). The modifier replaces the existing translation, rotation, or scale. The modifier adds to the existing translation, rotation, or scale. Use the Skeleton's ref pose as base. Use a whole animation as a base pose. BasePoseSeq must be set. Use one frame of an animation as a base pose. BasePoseSeq and RefFrameIndex must be set (RefFrameIndex will be clamped). Use one frame of this animation. RefFrameIndex must be set (RefFrameIndex will be clamped). Character can step up onto this Component. Owning actor determines whether character can step up onto this Component (default true unless overridden in code).
Location within the HMD tracking space Location within the view space Quad layer Cylinder layer Cubemap layer Equirect layer Use a constant value. Represents stepped values. Cubic interpolation. See TangentMode for different cubic interpolation options. No interpolation. Repeat the curve without an offset. Repeat the curve with an offset relative to the first or last key's value. Sinusoidally extrapolate. Use a linearly increasing value for extrapolation. Use a constant value for extrapolation No Extrapolation User specifies the tangent as a unified tangent where the two tangents are locked to each other, presenting a consistent curve before and after. User specifies the tangent as two separate broken tangents on each side of the key which can allow a sharp change in evaluation before or after. No tangents. New Auto tangent that creates smoother curves than Auto. Don't take tangent weights into account. Only take the arrival tangent weight into account for evaluation. Only take the leaving tangent weight into account for evaluation. Take both the arrival and leaving tangent weights into account for evaluation. No keys are present All keys use constant interpolation All keys use linear interpolation All keys use cubic interpolation Keys use mixed interpolation modes Keys use weighted interpolation modes Key time is quantized to 16 bits Key time uses full precision Blueprint that is const during execution (no state graph and methods cannot modify member variables). Blueprint that serves as a container for macros to be used in other blueprints. Blueprint that serves as an interface to be implemented by other blueprints. Blueprint that handles level scripting. Blueprint that serves as a container for functions to be used in other blueprints. Add to world. Subtract from world. Special tick group that starts physics simulation. Any item that can be run in parallel with our physics simulation work. Special tick group that ends physics simulation. Any item that needs rigid body and cloth simulation to be complete before being executed. Any item that needs the update work to be done before being ticked. Catchall for anything demoted to the end. Special tick group that is not actually a tick group. After every tick group this is repeatedly re-run until there are no more newly spawned items to run. Uses project based precision mode setting Force full-precision for MaterialFloat only, no effect on shader codes in .ush/.usf All the floats are full-precision Half precision, except explict 'float' in .ush/.usf Get the sampler from the texture. Every unique texture will consume a sampler slot, which are limited in number. Shared sampler source that does not consume a sampler slot. Uses wrap addressing and gets filter mode from the world texture group. Shared sampler source that does not consume a sampler slot. Uses clamp addressing and gets filter mode from the world texture group. Shared sampler source that does not consume a sampler slot, used to sample the terrain weightmap. Gets filter mode from the terrain weightmap texture group. Lighting will be calculated for a volume, without directionality. Use this on particle effects like smoke and dust. This is the cheapest per-pixel lighting method, however the material normal is not taken into account. Lighting will be calculated for a volume, with directionality so that the normal of the material is taken into account. Note that the default particle tangent space is facing the camera, so enable bGenerateSphericalParticleNormals to get a more useful tangent space. Same as Volumetric Non Directional, but lighting is only evaluated at vertices so the pixel shader cost is significantly less. Note that lighting still comes from a volume texture, so it is limited in range. Directional lights become unshadowed in the distance. Same as Volumetric Directional, but lighting is only evaluated at vertices so the pixel shader cost is significantly less. Note that lighting still comes from a volume texture, so it is limited in range. Directional lights become unshadowed in the distance. Lighting will be calculated for a surface. The light is accumulated in a volume so the result is blurry, limited distance but the per pixel cost is very low. Use this on translucent surfaces like glass and water. Only diffuse lighting is supported. Lighting will be calculated for a surface. Use this on translucent surfaces like glass and water. This is implemented with forward shading so specular highlights from local lights are supported, however many deferred-only features are not. This is the most expensive translucency lighting method as each light's contribution is computed per-pixel. Refraction is computed based on the camera vector entering a medium whose index of refraction is defined by the Refraction material input. By default, when the root node refraction pin is unplugged, no refraction will appear. The refraction offset into Scene Color is computed based on the difference between the per-pixel normal and the per-vertex normal. By default, when the root node refraction pin is unplugged, no refraction will appear. Explicit 2D screen offset. This offset is independent of screen resolution and aspect ratio. The user is in charge of any strength and fading. Refraction is disabled. Refraction is computed based on the camera vector entering a medium whose index of refraction is defined by the material IOR evaluated from F0. The new medium's surface is defined by the material's normal. With this mode, a flat plane seen from the side will have a constant refraction offset. This is a physical model of refraction but causes reading outside the scene color texture so is a poor fit for large refractive surfaces like water. This is the pre-Substrate behavior: coverage is ignored and always 1. When rough refraction is disabled, this is behavior is forced ON. This is a new behavior available with Substrate when rough refraction are enabled: account for roughness, coverage and depth. This is a more physically based behavior: the background scene will be visible untouched according to (1-coverage), while the blurred version will be visible according to coverage. This is the legacy mode where PDO is applied differently for Depth (along View Forward) and world position (along Camera Vector). PDO is applied along the Camera Vector for Depth and World Position altogether. Number of unique shading models. Shading model will be determined by the Material Expression Graph, by utilizing the 'Shading Model' MaterialAttribute output pin. None (movement is disabled). Walking on a surface. Simplified walking on navigation data (e.g. navmesh). If GetGenerateOverlapEvents() is true, then we will perform sweeps with each navmesh move. If GetGenerateOverlapEvents() is false then movement is cheaper but characters can overlap other objects without some extra process to repel/resolve their collisions. Falling under the effects of gravity, such as after jumping or walking off the edge of a surface. Swimming through a fluid volume, under the effects of gravity and buoyancy. Flying, ignoring the effects of gravity. Affected by the current physics volume's fluid friction. User-defined custom movement mode, including many possible sub-modes. Reserved for gizmo collision Add new serializeable channels above here (i.e. entries that exist in FCollisionResponseContainer) Add only nonserialized/transient flags below Returns both overlaps with both dynamic and static components returns only overlaps with dynamic actors (far fewer results in practice, much more efficient) returns only overlaps with static actors (fewer results, more efficient) No role at all. Locally simulated proxy of this actor. Locally autonomous proxy of this actor. Authoritative control over the actor. This actor can never go network dormant. This actor can go dormant, but is not currently dormant. Game code will tell it when it go dormant. This actor wants to go fully dormant for all connections. This actor may want to go dormant for some connections, GetNetDormancy() will be called to find out which. This actor is initially dormant for all connection if it was placed in map. Don't affect the walkable slope. Walkable slope angle will be ignored. Increase walkable slope. Makes it easier to walk up a surface, by allowing traversal over higher-than-usual angles. Decrease walkable slope. Makes it harder to walk up a surface, by restricting traversal to lower-than-usual angles. Make surface unwalkable. Note: WalkableSlopeAngle will be ignored. requires compute shader to construct 64 bin histogram faster method that computes single value by downsampling Uses camera settings. Sum of Gaussian formulation Fast Fourier Transform Image based convolution, intended for cinematics (too expensive for games) Obsolete - kept for backwards compatibility. No compression, no mips. Using this TextureGroup triggers special mip map generation code only useful for the BokehDOF post process. No compression, created on import of a .IES file. Non-filtered, useful for 2D rendering. Hierarchical LOD generated textures Impostor Color Textures Impostor Normal and Depth, use default compression 8 bit data stored in textures 16 bit data stored in textures Project specific group, rename in Engine.ini, [EnumRemap] TEXTUREGROUP_Project**.DisplayName=My Fun Group Default for the "texture". 2x2 average, default for the "texture group". 8x8 with sharpening: 0=no sharpening but better quality which is softer, 1=little, 5=medium, 10=extreme. Do not touch existing mip chain as it contains generated data. Blur further (useful for image based reflections). Use the first texel of each 2x2 (or 2x2x2) group. Introduce significant amount of blur using angular filtering (only applies to cubemaps, useful for ambient lighting). CompositingTexture needs to be a normal map with the same or larger size. Get the tiling setting from the texture's group CookPlatformTilingDisabled setting. By default it's to tile during cook, unless it has been changed in the texture group The texture will be tiled during the cook process if the platform supports it. The texture will not be tiled during the cook process, and will be tiled when uploaded to the GPU if the platform supports it. Use setting from the Texture Group. Lighting samples are computed in an adaptive grid which covers the entire Lightmass Importance Volume. Higher density grids are used near geometry. The Volumetric Lightmap is interpolated efficiently on the GPU per-pixel, allowing accurate indirect lighting for dynamic objects and volumetric fog. Positions outside of the Importance Volume reuse the border texels of the Volumetric Lightmap (clamp addressing). On mobile, interpolation is done on the CPU at the center of each object's bounds. Volume lighting samples are placed on top of static surfaces at medium density, and everywhere else in the Lightmass Importance Volume at low density. Positions outside of the Importance Volume will have no indirect lighting. This method requires CPU interpolation so the Indirect Lighting Cache is used to interpolate results for each dynamic object, adding Rendering Thread overhead. UserSet - use the user set value. Primarily intended for weapon effects. Emitter - use the emitter position as the source/target. Particle - use the particles from a given emitter in the system. Actor - use the actor as the source/target. The name of the actor should be set in <Source/Target>Name. Direct - a direct line between source and target. UserSet - use the user set value. Distribution - use the distribution. Emitter - use the emitter direction. Modify the target of the beam. Freeze the particle in place Stop collision checks, but keep updating Stop translations of the particle Stop rotations of the particle Stop all movement of the particle Multiply the module values by the previous results 'Break' the chain and apply the values from the previous results Lock the sprite facing towards the positive X-axis Lock the sprite facing towards the positive Y-axis Lock the sprite facing towards the positive Z-axis Lock the sprite facing towards the negative X-axis Lock the sprite facing towards the negative Y-axis Lock the sprite facing towards the negative Z-axis Lock the sprite rotation on the X-axis Lock the sprite rotation on the Y-axis Lock the sprite rotation on the Z-axis TypeData - TypeData modules Beam - only applied to beam emitters Trail - only applied to trail emitters Spawn - all emitter types REQUIRE it Required - all emitter types REQUIRE it Event - event related modules Light related modules SubUV related modules Random - select a particle at random Sequential - select a particle in order FixedTime - update via a fixed time step Don't determine occlusion on this particle system Use the bounds of the particle system component when determining occlusion Use the custom occlusion bounds when determining occlusion Replay system is disabled. Particles are simulated and rendered normally. Capture all particle data to the clip specified by ReplayClipIDNumber. The frame to capture must be specified using the ReplayFrameIndex Replay captured particle state from the clip specified by ReplayClipIDNumber. The frame to play must be specified using the ReplayFrameIndex Any - allow any event Spawn - a particle spawn event Death - a particle death event Collision - a particle collision event Burst - a particle burst event Blueprint - an event generated by level script Set absolute position of bone in world space. Set position of bone in SkeletalMeshComponent's reference frame. Set position of bone relative to parent bone. Set position of bone in its own reference frame. Don't change rotation at all. Keep forward direction vector relative to the parent bone. Copy rotation of target to bone. No additive. The material's attributes describe a deferred decal, and will be mapped onto the decal's frustum. The material's attributes describe a light's distribution. The material's attributes describe a 3d volume. The material will be used in a custom post process pass. The material will be used for UMG or Slate UI The material will be used for runtime virtual texture (Deprecated). Modulate BaseColor, blend rest, updating the GBuffer, does not work for baked lighting. Does not work in DBuffer mode (approximated as Translucent). Only blend normal, updating the GBuffer, does not work for baked lighting. Additive emissive only. Put into DBuffer to work for baked lighting as well (becomes DBM_TranslucentNormal if normal is not hooked up). Put into DBuffer to work for baked lighting as well. Put into DBuffer to work for baked lighting as well (becomes DBM_DBuffer_Color if normal is not hooked up). Put into DBuffer to work for baked lighting as well (becomes DBM_DBuffer_Roughness if normal is not hooked up). Internal DBffer decal blend modes used for auto-converted decals Output signed distance in Opacity depending on LightVector. Note: Can be costly, no shadow casting but receiving, no per pixel normal yet, no quality settings yet Blend with existing scene color. Decal color is already pre-multiplied by alpha. Ambient occlusion. Do not receive decals (Later we still can read the DBuffer channels to customize the effect, this frees up some interpolators). Receive Decals, applies all DBuffer channels. Receive Decals, applies color DBuffer channels. Receive Decals, applies color and normal DBuffer channels. Receive Decals, applies color, roughness, specular and metallic DBuffer channels. Receive Decals, applies normal DBuffer channels. Receive Decals, applies normal, roughness, specular and metallic DBuffer channels. Receive Decals, applies roughness, specular and metallic DBuffer channels. Render before depth of field. Render after depth of field. Render after motion blur. Disables depth test (the reconstruction post MB and TSR would otherwise flicker due to the TSR camera jittering making the depth buffer unstable). Because of that Lumen high quality reflections is also disabled to avoid visual discrepancy at depth intersection. Object local Skinned local DBuffer B (World Normal). RGB is pre-multiplied by alpha. A is one minus alpha. DBuffer C (Roughness). R is Roughness, G is Metallic, B is Specular, all pre-multiplied by alpha. A is one minus alpha. Buffer index UV in LayerId (batch) x Channels x Height x Wdith High quality for direct use and bumps Non-tiled: ~61 instructions per level, 8 texture lookups Tiling: ~74 instructions per level, 8 texture lookups Even "non-tiled" mode has a repeat of 128. Useful Repeat Size range <= 128 Formerly labeled as Perlin noise High quality for direct use, BAD for bumps; doesn't work on Mobile ~16 instructions per level, 1 texture lookup Always tiles with a repeat of 16, "Tiling" mode is not an option for Fast Gradient noise High quality for direct use and bumps Non-tiled: ~80 instructions per level, no textures Tiling: ~143 instructions per level, no textures Low quality, but pure computation Non-tiled: ~53 instructions per level, no textures Tiling: ~118 instructions per level, no textures Formerly mis-labeled as Gradient noise Also known as Worley or Cellular noise Quality=1 searches 8 cells, Quality=2 searches 16 cells Quality=3 searches 27 cells, Quality=4 searches 32 cells All are about 20 instructions per cell searched Normal (Path Tracing). Stores the denoised radiance if denoising is turned on and complete for the current frame, otherwise, black. Albedo (Path Tracing). Average albedo at the current sample count. Normal (Path Tracing). Average normal at the current sample count. Variance (Path Tracing). Path tracing variance stored as standard derivation. Variance can be per channel variance or variance of luminance, albedo, and normal based on the path tracing configuration. Hooking up this buffer can increase additional cost. Local space (relative to the rendered object, = object space) World space, a unit is 1cm View space (relative to the camera/eye, = camera space, differs from camera space in the shadow passes) Camera space Particle space, deprecated value will be removed in a future release use instance space. Instance space (used to provide per instance transform, i.e. for Instanced Static Mesh / Particles). Tangent space (relative to the surface) Absolute world space Like absolute world space, but the world origin is moved to the center of the tile the camera is in. Logically similar to Translated world space, i.e. world space rotation and scale but with a position relative to the camera First person "space", which can be thought of as a transform that is applied to a position in translated world space. View space (differs from camera space in the shadow passes) Perlin-style noise with 3D vector/color output. RGB output range -1 to 1 R only = ~83 instructions, RGB = ~125 instructions Gradient of Perlin noise, useful for bumps. RGB = Gradient of scalar noise (signed 3D vector) A = Base scalar noise with range -1 to 1 A only = ~83 instructions, RGBA = ~106 instructions Curl of Perlin noise, useful for 3D flow directions. RGB = signed curl vector ~162 instructions Also known as Worley or Cellular noise. RGB = position of closest point at center of Voronoi cell A = distance to closest point with range 0 to about 4 Quality levels 1-4 search 8, 16, 27 & 32 cells All ~20 instructions per cell searched Horizontal and vertical field of view angles in radian Tan(FieldOfView * 0.5) Horizontal and vertical size of the view in pixels Absolute world space view position (differs from the camera position in the shadow passes) Absolute world space camera position Horizontal and vertical position of the viewport in pixels within the buffer. Number of temporal AA sample used across multiple to converge to anti aliased output. Index of the Temporal AA jitter for this frame. Offset of the temporal sample for this frame in pixel size. Mip Level that Runtime Virtual Texture Output is rendering to. World space derivatives for Runtime Virtual Texture Output. Pre Exposure Maximum mip level of Runtime Virtual Texture that Runtime Virtual Texture Output is rendering to. Screen percentage at which the rendering resolution happens, to allow tech-art to remain consistent with dynamic resolution. Post process volume user flags, useful for varying the behavior of a material per view Horizontal and vertical first person field of view angles in radian Tan(FirstPersonFieldOfView * 0.5) Scaling factor applied to first person primitives to keep them from intersecting with the scene. Near plane. Absolute world position with no material shader offsets applied Camera relative world position with all material shader offsets applied Camera relative world position with no material shader offsets applied Scene depth, single channel, contains the linear depth of the opaque objects Material diffuse, RGB color (computed from GBuffer) Material specular, RGB color (computed from GBuffer) Material subsurface, RGB color (GBuffer, only for some ShadingModels) Material base, RGB color (GBuffer), can be modified on read by the ShadingModel, consider StoredBasedColor Material specular, single channel (GBuffer), can be modified on read by the ShadingModel, consider StoredSpecular Material metallic, single channel (GBuffer) Normal, RGB in -1..1 range, not normalized (GBuffer) Not yet supported Material opacity, single channel (GBuffer) Material roughness, single channel (GBuffer) Material ambient occlusion, single channel (GBuffer) Scene depth, single channel, contains the linear depth of the opaque objects rendered with CustomDepth (mesh property) Input #0 of this postprocess pass, usually the only one hooked up Input #1 of this postprocess pass, usually not used Input #2 of this postprocess pass, usually not used Input #3 of this postprocess pass, usually not used Input #4 of this postprocess pass, usually not used Input #5 of this postprocess pass, usually not used Input #6 of this postprocess pass, usually not used Decal Mask, single bit (was moved to stencil for better performance, not accessible at the moment) Shading model Shading model ID Ambient Occlusion, single channel Scene stencil, contains CustomStencil mesh property of the opaque objects rendered with CustomDepth Material base, RGB color (GBuffer) Material specular, single channel (GBuffer) Scene Velocity Tangent, RGB in -1..1 range, not normalized (GBuffer) Material anisotropy, single channel (GBuffer) Whether the pixel is marked as belonging to an opaque first person primitive, single channel (GBuffer) Only supported with forward shading. MSAA sample count is controlled by r.MSAACount. Looped playback mode. Play all movies in the play list in order then start over until manually canceled Alternate Looped mode. Play all of the movies in the play list and loop just the last movie until loading is finished. Create both simple and complex shapes. Simple shapes are used for regular scene queries and collision tests. Complex shape (per poly) is used for complex scene queries. Create only simple shapes. Use simple shapes for all scene queries and collision tests. Create only complex shapes (per poly). Use complex shapes for all scene queries and collision tests. Can be used in simulation for static shapes only (i.e can be collided against but not moved through forces or velocity.) Follow owner. Do not follow owner, but make kinematic. Do not follow owner, but simulate. No constraint against this axis. Limited freedom along this axis. Fully constraint against this axis. Left-align. Center-align. Right-align. Fill the entire height. Top-align. Bottom-align. Place the menu immediately below the anchor Place the menu immediately centered below the anchor Place the menu immediately below the anchor aligned to the right of the content Place the menu immediately below the anchor and match is width to the anchor's content Place the menu immediately below the anchor and match is width to the anchor's content. If the width overflows, align with the right edge of the anchor. Place the menu to the right of the anchor Place the menu immediately above the anchor, no transition effect Place the menu immediately centered above the anchor, no transition effect Place the menu immediately above the anchor aligned to the right of the content Place the menu to the left of the anchor Place the menu's center on top of the menu anchor's center point Place the menu's vertical center on the left side at the menu anchor's vertical center on the right side Place the menu's bottom left corner directly on top of the menu anchor's bottom left corner Orient horizontally, i.e. left to right. Orient vertically, i.e. top to bottom. Scroll down. Scroll up. |
1.9.8