EN：
| Name | Help |
|---|---|
| r.AllowCachedUniformExpressions | Allow uniform expressions to be cached. |
| r.AllowDepthBoundsTest | If true, use enable depth bounds test when rendering defered lights. |
| r.AllowLandscapeShadows | Allow Landscape Shadows |
| r.AllowOcclusionQueries | If zero, occlusion queries will not be used to cull primitives. |
| r.AllowPointLightCubemapShadows | When 0, will prevent point light cube map shadows from being used and the light will be unshadowed. |
| r.AllowPrecomputedVisibility | If zero, precomputed visibility will not be used to cull primitives. |
| r.AllowStaticLighting | Whether to allow any static lighting to be generated and used, like lightmaps and shadowmaps. Games that only use dynamic lighting should set this to 0 to save some static lighting overhead. |
| r.AmbientOcclusion.FadeRadiusScale | Allows to scale the ambient occlusion fade radius (SSAO). 0.01:smallest .. 1.0:normal (default), <1:smaller, >1:larger |
| r.AmbientOcclusionLevels | Defines how many mip levels are using during the ambient occlusion calculation. This is useful when tweaking the algorithm. 0:none, 1:one, 2:two, 3:three(default), 4:four(larger at little cost but can flicker) |
| r.AmbientOcclusionRadiusScale | Allows to scale the ambient occlusion radius (SSAO). 0:off, 1.0:normal, <1:smaller, >1:larger |
| r.AmbientOcclusionSampleSetQuality | Defines how many samples we use for ScreenSpace Ambient Occlusion -1: sample count depends on post process settings (default) 0: low sample count (defined in shader, 3 * 2) 1: high sample count (defined in shader, 6 * 2) |
| r.AmbientOcclusionStaticFraction | Allows to override the Ambient Occlusion Static Fraction (see post process volume). Fractions are between 0 and 1. <0: use default setting (default -1) 0: no effect on static lighting, 0 is free meaning no extra rendering pass 1: AO affects the stat lighting |
| r.AOAsyncBuildQueue | Whether to asynchronously build distance field volume data from meshes. |
| r.AOComputeShaderNormalCalculation | Whether to use the compute shader version of the distance field normal computation. |
| r.AOFillGaps | Whether to fill in pixels using a screen space filter that had no valid world space interpolation weight from surface cache samples. This is needed whenever r.AOMinLevel is not 0. |
| r.AOHistoryWeight | Amount of last frame's AO to lerp into the final result. Higher values increase stability, lower values have less streaking under occluder movement. |
| r.AOInterpolationAngleScale | Scale applied to angle error during the final interpolation pass. Values larger than 1 result in smoothing. |
| r.AOInterpolationDepthTesting | Whether to use depth testing during the interpolation splat pass, useful for debugging |
| r.AOInterpolationMaxAngle | Largest angle allowed between the shading point's normal and a nearby record's normal. |
| r.AOInterpolationRadiusScale | Scale applied to record radii during the final interpolation pass. Values larger than 1 result in world space smoothing. |
| r.AOInterpolationStencilTesting | Whether to stencil out distant pixels from the interpolation splat pass, useful for debugging |
| r.AOMaxLevel | Largest downsample power of 4 to use for surface cache population. |
| r.AOMaxViewDistance | The maximum distance that AO will be computed at. |
| r.AOMinLevel | Smallest downsample power of 4 to use for surface cache population. The default is 1, which means every 8 full resolution pixels (BaseDownsampleFactor(2) * 4^1) will be checked for a valid interpolation from the cache or shaded. Going down to 0 gives less aliasing, and removes the need for gap filling, but costs a lot. |
| r.AOMinPointBehindPlaneAngle | Minimum angle that a point can lie behind a record and still be considered valid. This threshold helps reduce leaking that happens when interpolating records in front of the shading point, ignoring occluders in between. |
| r.AOOverwriteSceneColor | |
| r.AOPowerOfTwoBetweenLevels | Power of two in resolution between refinement levels of the surface cache |
| r.AORecordRadiusScale | Scale applied to the minimum occluder distance to produce the record radius. This effectively controls how dense shading samples are. |
| r.AOReuseAcrossFrames | Whether to allow reusing surface cache results across frames. |
| r.AOSampleSet | 0 = Original set, 1 = Relaxed set |
| r.AOScatterInterpolation | |
| r.AOScatterTileCulling | Whether to use the rasterizer for binning occluder objects into screenspace tiles. |
| r.AOStepExponentScale | Exponent used to distribute AO samples along a cone direction. |
| r.AOTrimOldRecordsFraction | When r.AOReuseAcrossFrames is enabled, this is the fraction of the last frame's surface cache records that will not be reused. Low settings provide better performance, while values closer to 1 give faster lighting updates when dynamic scene changes are happening. |
| r.AOUseHistory | Whether to apply a temporal filter to the distance field AO, which reduces flickering but also adds trails when occluders are moving. |
| r.Atmosphere | Defines atmosphere will render or not. Only changed by r.Atmosphere console command. Enable/Disable Atmosphere, Load/Unload related data. 0: off 1: on (default) |
| r.BlackBorders | To draw black borders around the rendered image (prevents artifacts from post processing passes that read outside of the image e.g. PostProcessAA) in pixels, 0:off |
| r.BloomQuality | 0: off, no performance impact. 1: average quality, least performance impact. 2: average quality, least performance impact. 3: good quality. 4: good quality. 5: Best quality, most significant performance impact. (default) >5: force experimental higher quality on mobile (can be quite slow on some hardware) |
| r.BokehDOFIndexStyle | Controls whether to use a packed or unwound index buffer for Bokeh DOF. 0: Use packed index buffer (faster) (default) 1: Use unwound index buffer (slower) |
| r.BufferVisualizationDumpFrames | When screenshots or movies dumps are requested, also save out dumps of the current buffer visualization materials 0:off (default) 1:on |
| r.BufferVisualizationDumpFramesAsHDR | When saving out buffer visualization materials in a HDR capable format 0: Do not override default save format. 1: Force HDR format for buffer visualization materials. |
| r.BufferVisualizationOverviewTargets | Specify the list of post process materials that can be used in the buffer visualization overview. Put nothing between the commas to leave a gap. Choose from: BaseColor CustomDepth DecalMask FinalImage ShadingModel MaterialAO Metallic Opacity Roughness SceneColor SceneDepth SeparateTranslucencyRGB SeparateTranslucencyA Specular SubsurfaceColor WorldNormal AmbientOcclusion CustomDepthWorldUnits SceneDepthWorldUnits |
| r.BufferVisualizationTarget | When the viewport view-mode is set to 'Buffer Visualization', this command specifies which of the various channels to display. Values entered other than the allowed values shown below will be ignored. BaseColor CustomDepth DecalMask FinalImage ShadingModel MaterialAO Metallic Opacity Roughness SceneColor SceneDepth SeparateTranslucencyRGB SeparateTranslucencyA Specular SubsurfaceColor WorldNormal AmbientOcclusion CustomDepthWorldUnits SceneDepthWorldUnits |
| r.Cache.DrawDirectionalShadowing | Whether to draw direct shadowing sample points as generated by Lightmass. 0 is off (default), 1 is on |
| r.Cache.DrawInterpolationPoints | Whether to draw positions that indirect lighting is interpolated at when they are updated, which are stored in the cache. Probably need 'r.CacheUpdateEveryFrame 1' as well to be useful, otherwise points will flicker as they update. 0 is off (default), 1 is on |
| r.Cache.DrawLightingSamples | Whether to draw indirect lighting sample points as generated by Lightmass. 0 is off (default), 1 is on |
| r.Cache.LightingCacheDimension | Dimensions of the lighting cache. This should be a multiple of r.LightingCacheMovableObjectAllocationSize for least waste. |
| r.Cache.LightingCacheMovableObjectAllocationSize | Resolution of the interpolation sample volume used to light a dynamic object. Values of 1 or 2 will result in a single interpolation sample per object which does not provide continuous lighting under movement, so interpolation is done over time. Values of 3 or more support the necessary padding to provide continuous results under movement. |
| r.Cache.LimitQuerySize | 0 is off, 1 is on (default) |
| r.Cache.QueryNodeLevel | Level of the lighting sample octree whose node's extents should be the target size for queries into the octree. Primitive blocks will be broken up into multiple octree queries if they are larger than this.0 is the root, 12 is the leaf level |
| r.Cache.ReduceSHRinging | Whether to modify indirect lighting cache SH samples to reduce ringing. 0 is off, 1 is on (default) |
| r.Cache.SampleTransitionSpeed | When using single sample lighting, controls the speed of the transition between two point samples (fade over time). |
| r.Cache.UpdateEveryFrame | Whether to update indirect lighting cache allocations every frame, even if they would have been cached. 0 is off (default), 1 is on |
| r.CatmullRomEndParamOffset | The parameter offset for catmul rom end points. |
| r.ClearSceneMethod | Select how scene rendertarget clears are handled 0: No clear 1: RHIClear (default) 2: Quad at max z |
| r.ClearWithExcludeRects | Control the use of exclude rects when using RHIClear 0: Force off (can be faster on hardware that has fast clears) 1: Use exclude rect if supplied 2: Auto (default is 2, pick what is considered best on this hardware) |
| r.Color.Max | Allows to define where the value 1.0 in the color channels is mapped to after color grading. Value should be around 1, smaller values darken the highlights, larger values move more colors towards white, Default: 1 |
| r.Color.Mid | Allows to define where the value 0.5 in the color channels is mapped to after color grading (This is similar to a gamma correction). Value should be around 0.5, smaller values darken the mid tones, larger values brighten the mid tones, Default: 0.5 |
| r.Color.Min | Allows to define where the value 0 in the color channels is mapped to after color grading. The value should be around 0, positive: a gray scale is added to the darks, negative: more dark values become black, Default: 0 |
| r.CompileMaterialsForShaderFormat | When enabled, compile materials for this shader format in addition to those for the running platform. Note that these shaders are compiled and immediately tossed. This is only useful when directly inspecting output via r.DebugDumpShaderInfo. |
| r.CompileShadersForDevelopment | Setting this to 0 allows to ship a game with more optimized shaders as some editor and development features are not longer compiled into the shaders. Note: This should be done when shipping but it's not done automatically yet (feature need to mature and shaders will compile slower as shader caching from development isn't shared). Cannot be changed at runtime - can be put into BaseEngine.ini 0: off, shader can run a bit faster 1: on (Default) |
| r.CreateShadersOnLoad | Whether to create shaders on load, which can reduce hitching, but use more memory. Otherwise they will be created as needed. |
| r.CustomDepth | 0: feature is disabled 1: feature is enabled, texture is created on demand 2: feature is enabled, texture is not released until required (should be the project setting if the feature should not stall) |
| r.D3DCompilerPath | Allows to specify a HLSL compiler version that is different from the one the code was compiled. No path ("") means the default one is used. If the compiler cannot be found an error is reported and it will compile further with the default one. This console variable works with ShaderCompileWorker (with multi threading) and without multi threading. This variable can be set in ConsoleVariables.ini to be defined at startup. e.g. c:/temp/d3dcompiler_44.dll or "" |
| r.DBuffer | Experimental DBuffer feature: Generating deferred decals before the BasePass. Allows decals to be correctly lit by baked lighting. Receivers need to be rendered in the early zpass. At the moment only can be ensures by full enablng this pass: r.EarlyZPassMovable=1 r.EarlyZPass=2 0: off 1: on (needs early pass rendering on all decal receivers and base pass lookups into the DBuffer, costs GPU memory, allows GBuffer compression) |
| r.Decal.StencilSizeThreshold | Control a per decal stencil pass that allows to large (screen space) decals faster. It adds more overhead per decals so this <0: optimization is disabled 0: optimization is enabled no matter how small (screen space) the decal is 0..1: optimization is enabled, value defines the minimum size (screen space) to trigger the optimization (default 0.1) |
| r.DefaultFeature.AmbientOcclusion | Engine default (project setting) for AmbientOcclusion is (postprocess volume/camera/game setting still can override) 0: off, sets AmbientOcclusionIntensity to 0 1: on (default) |
| r.DefaultFeature.AmbientOcclusionStaticFraction | Engine default (project setting) for AmbientOcclusion is (postprocess volume/camera/game setting still can override) 0: off, sets AmbientOcclusionStaticFraction to 0 1: on (default, costs extra pass, only useful if there is some baked lighting) |
| r.DefaultFeature.AntiAliasing | Engine default (project setting) for AntiAliasingMethod is (postprocess volume/camera/game setting still can override) 0: off (no anti-aliasing) 1: FXAA (faster than TemporalAA but much more shimmering for non static cases) 2: TemporalAA (default) |
| r.DefaultFeature.AutoExposure | Engine default (project setting) for AutoExposure is (postprocess volume/camera/game setting still can override) 0: off, sets AutoExposureMinBrightness and AutoExposureMaxBrightness to 1 1: on (default) |
| r.DefaultFeature.Bloom | Engine default (project setting) for Bloom is (postprocess volume/camera/game setting still can override) 0: off, set BloomIntensity to 0 1: on (default) |
| r.DefaultFeature.LensFlare | Engine default (project setting) for LensFlare is (postprocess volume/camera/game setting still can override) 0: off, sets LensFlareIntensity to 0 1: on (default) |
| r.DefaultFeature.MotionBlur | Engine default (project setting) for MotionBlur is (postprocess volume/camera/game setting still can override) 0: off, sets MotionBlurAmount to 0 1: on (default) |
| r.DemosaicVposOffset | This offset is added to the rasterized position used for demosaic in the ES2 tonemapping shader. It exists to workaround driver bugs on some Android devices that have a half-pixel offset. |
| r.DepthOfField.MaxSize | Allows to clamp the gaussian depth of field radius (for better performance), default: 100 |
| r.DepthOfFieldNearBlurSizeThreshold | Sets the minimum near blur size before the effect is forcably disabled. Currently only affects Gaussian DOF. (default = 0.01f) |
| r.DepthOfFieldQuality | Allows to adjust the depth of field quality. Currently only fully affects BokehDOF. GaussianDOF is either 0 for off, otherwise on. 0: Off 1: Low 2: high quality (default, adaptive, can be 4x slower) |
| r.DetailMode | Current detail mode; determines whether components of actors should be updated/ ticked. 0: low, show only object with DetailMode low or higher 1: medium, show all object with DetailMode medium or higher 2: high, show all objects (default) |
| r.DFShadowScatterTileCulling | Whether to use the rasterizer to scatter objects onto the tile grid for culling. |
| r.DFShadowWorldTileSize | World space size of a tile used for culling for directional lights. |
| r.DiffuseFromCaptures | Apply indirect diffuse lighting from captures instead of lightmaps. 0 is off (default), 1 is on |
| r.DisableLODFade | Disable fading for distance culling |
| r.DistanceFadeMaxTravel | Max distance that the player can travel during the fade time. |
| r.DistanceFieldAO | Whether the distance field AO feature is allowed, which is used to implement shadows of Movable sky lights from static meshes. |
| r.DistanceFieldGI | |
| r.DistanceFieldShadowing | Whether the distance field shadowing feature is allowed. |
| r.DontLimitOnBattery | 0: Limit performance on devices with a battery.(default) 1: Do not limit performance due to device having a battery. |
| r.DownsampledOcclusionQueries | Whether to issue occlusion queries to a downsampled depth buffer |
| r.DrawRectangleOptimization | Controls an optimization for DrawRectangle(). When enabled a triangle can be used to draw a quad in certain situations (viewport sized quad). Using a triangle allows for slightly faster post processing in lower resolutions but can not always be used. 0: Optimization is disabled, DrawDenormalizedQuad always render with quad 1: Optimization is enabled, a triangle can be rendered where specified (default) |
| r.DumpingMovie | Allows to dump each rendered frame to disk (slow fps, names MovieFrame..). <=0:off (default), <0:remains on, >0:remains on for n frames (n is the number specified) |
| r.DumpShaderDebugInfo | When set to 1, will cause any shaders that are then compiled to dump debug info to GameName/Saved/ShaderDebugInfo The debug info is platform dependent, but usually includes a preprocessed version of the shader source. On iOS, if the PowerVR graphics SDK is installed to the default path, the PowerVR shader compiler will be called and errors will be reported during the cook. |
| r.EarlyZPass | Whether to use a depth only pass to initialize Z culling for the base pass. Cannot be changed at runtime. Note: also look at r.EarlyZPassMovable 0: off 1: only if not masked, and only if large on the screen 2: all opaque (including masked) x: use built in heuristic (default is 3) |
| r.EarlyZPassMovable | Whether to render movable objects into the depth only pass. Movable objects are typically not good occluders so this defaults to off. Note: also look at r.EarlyZPass |
| r.Editor.2DGridFade | Tweak to define the grid rendering in 2D viewports. |
| r.Editor.2DSnapFade | Tweak to define the grid rendering in 2D viewports. |
| r.Editor.2DSnapMin | Tweak to define the grid rendering in 2D viewports. |
| r.Editor.2DSnapScale | Tweak to define the grid rendering in 2D viewports. |
| r.Editor.3DGridFade | Tweak to define the grid rendering in 3D viewports. |
| r.Editor.3DSnapFade | Tweak to define the grid rendering in 3D viewports. |
| r.Editor.AlignedOrthoZoom | Only affects the editor ortho viewports. 0: Each ortho viewport zoom in defined by the viewport width 1: All ortho viewport zoom are locked to each other to allow axis lines to be aligned with each other. |
| r.Editor.MovingPattern | 0:animation over time is off (default is 1) |
| r.Editor.NewLevelGrid | Wether to show the new editor level grid 0: off 1: Analytical Antialiasing 2: Texture based(default) |
| r.Editor.OpaqueGizmo | 0..1 0: occluded gizmo is partly transparent (default), 1:gizmo is never occluded |
| r.EmitMeshDrawEvents | Emits a GPU event around each drawing policy draw call. /nUseful for seeing stats about each draw call, however it greatly distorts total time and time per draw call. |
| r.EmitterSpawnRateScale | A global scale upon the spawn rate of emitters. Emitters can choose to apply or ignore it via their bApplyGlobalSpawnRateScale property. |
| r.ExposureOffset | For adjusting the exposure on top of post process settings and eye adaptation. For developers only. 0:default |
| r.EyeAdaptationQuality | Defines the eye adaptation quality which allows to adjust for quality or performance. <=0: off (fastest) 1: low quality (e.g. non histogram based, not yet implemented) 2: normal quality (default) 3: high quality (e.g. screen position localized, not yet implemented) |
| r.FastBlurThreshold | Defines at what radius the Gaussian blur optimization kicks in (estimated 25% - 40% faster). The optimzation uses slightly less memory and has a quality loss on smallblur radius. 0: use the optimization always (fastest, lowest quality) 3: use the optimization starting at a 3 pixel radius (quite fast) 7: use the optimization starting at a 7 pixel radius (default) >15: barely ever use the optimization (high quality) |
| r.FeatureLevelPreview | If 1 the quick settings menu will contain an option to enable feature level preview modes |
| r.FinishCurrentFrame | If on, the current frame will be forced to finish and render to the screen instead of being buffered. This will improve latency, but slow down overall performance. |
| r.FogDensity | Allows to override the FogDensity setting (needs ExponentialFog in the level). Using a strong value allows to quickly see which pixel are affected by fog. Using a start distance allows to cull pixels are can speed up rendering. <0: use default settings (default: -1) >=0: override settings by the given value (0:off, 1=very dense fog) |
| r.FogStartDistance | Allows to override the FogStartDistance setting (needs ExponentialFog in the level). <0: use default settings (default: -1) >=0: override settings by the given value (in world units) |
| r.ForceDebugViewModes | 0: Setting has no effect. 1: Forces debug view modes to be available, even on cooked builds.2: Forces debug view modes to be unavailable, even on editor builds. Removes many shader permutations for faster shader iteration. |
| r.ForceLOD | LOD level to force, -1 is off. |
| r.ForwardBasePassSort | How to sort the forward shading base pass: 0: Decide automatically based on the hardware. 1: Sort drawing policies. 2: Sort drawing policies and the meshes within them. |
| r.FreeSkeletalMeshBuffers | Controls whether skeletal mesh buffers are kept in CPU memory to support merging of skeletal meshes. 0: Keep buffers(default) 1: Free buffers |
| r.FullScreenMode | Defines how we do full screen when requested (e.g. command line option -fullscreen or in ini [SystemSettings] fullscreen=true) 0: normal full screen (renders faster, more control over vsync, less GPU memory, 10bit color if possible) 1: windowed full screen (quick switch between applications and window mode, slight performance loss) any other number behaves like 0 |
| r.GBuffer | 0=Do not use GBuffer (fast minimal rendering path), 1=Use GBuffer [default]. |
| r.GBufferFormat | Defines the memory layout used for the GBuffer. (affects performance, mostly through bandwidth, quality of normals and material attributes). 0: lower precision (8bit per component, for profiling) 1: low precision (default) 5: high precision |
| r.GeneralPurposeTweak | Useful for low level shader development to get quick iteration time without having to change any c++ code. Value maps to View.GeneralPurposeTweak inside the shaders. Example usage: Multiplier on some value to tweak, toggle to switch between different algorithms (Default: 1.0) DON'T USE THIS FOR ANYTHING THAT IS CHECKED IN. Compiled out in SHIPPING to make cheating a bit harder. |
| r.GenerateMeshDistanceFields | Whether to build distance fields of static meshes, needed for distance field AO, which is used to implement Movable SkyLight shadows. Enabling will increase mesh build times and memory usage. Changing this value will cause a rebuild of all static meshes. |
| r.GPUBusyWait | <=0:off, >0: keep the GPU busy with n units of some fixed amount of work, independent on the resolution This can be useful to make GPU timing experiments. The value should roughly represent milliseconds. Clamped at 500. |
| r.GraphicsAdapter | User request to pick a specific graphics adapter (e.g. when using a integrated graphics card with a descrete one) At the moment this only works on Direct3D 11. -2: Take the first one that fulfills the criteria -1: Favour non integrated because there are usually faster 0: Adpater #0 1: Adpater #1, ... |
| r.HalfResReflections | Compute ReflectionEnvironment samples at half resolution. 0 is off (default), 1 is on |
| r.HighQualityLightMaps | If set to 1, allow high quality lightmaps which don't bake in direct lighting of stationary lights |
| r.HZBOcclusion | Defines which occlusion system is used. 0: Hardware occlusion queries 1: Use HZB occlusion system (default, less GPU and CPU cost, more conservative results) 2: Force HZB occlusion system (overrides rendering platform preferences) |
| r.IndirectLightingCache | Whether to use the indirect lighting cache on dynamic objects. 0 is off, 1 is on (default) |
| r.LegacySingleThreadedRelevance | Toggles the legacy codepath for view relevance. |
| r.LensFlareQuality | 0: off but best for performance 1: low quality with good performance 2: good quality (default) 3: very good quality but bad performance |
| r.LightFunctionQuality | Defines the light function quality which allows to adjust for quality or performance. <=0: off (fastest) 1: low quality (e.g. half res with blurring, not yet implemented) 2: normal quality (default) 3: high quality (e.g. super-sampled or colored, not yet implemented) |
| r.LightPropagationVolume | Project setting of the work in progress feature LightPropgationVolume. Cannot be changed at runtime. 0: off (default) 1: on |
| r.LightShaftBlurPasses | Number of light shaft blur passes. |
| r.LightShaftDownSampleFactor | Downsample factor for light shafts. range: 1..8 |
| r.LightShaftFirstPassDistance | Fraction of the distance to the light to blur on the first radial blur pass. |
| r.LightShaftNumSamples | Number of samples per light shaft radial blur pass. Also affects how quickly the blur distance increases with each pass. |
| r.LightShaftQuality | Defines the light shaft quality. 0: off 1: on (default) |
| r.LightShafts | Whether light shafts are allowed to be rendered, defaults to 1. |
| r.LimitRenderingFeatures | Allows to quickly reduce render feature to increase render performance. This is just a quick way to alter multiple show flags and console variables in the game Disabled more feature the higher the number <=0:off, order is defined in code (can be documented here when we settled on an order) |
| r.LODFadeTime | How long LOD takes to fade (in seconds). |
| r.MaterialEditor.UseDevShaders | Toggles whether the material editor will use shaders that include extra overhead incurred by the editor. Material editor must be re-opened if changed at runtime. |
| r.MaterialQualityLevel | 0 corresponds to low quality materials, as defined by quality switches in materials, 1 corresponds to high. |
| r.MaxAnisotropy | MaxAnisotropy should range from 1 to 16. Higher values mean better texure quality when using anisotropic filtering but at a cost to performance. |
| r.MaxCSMRadiusToAllowPerObjectShadows | Only stationary lights with a CSM radius smaller than this will create per object shadows for dynamic objects. |
| r.MaxForwardBasePassDraws | Stops rendering static forward base pass draws after the specified number of times. Useful for seeing the order in which meshes render when optimizing. |
| r.MaxGPUSkinCacheElementsPerFrame | The maximum compute processed skin cache elements per frame. |
| r.MaxQualityMode | If set to 1, override certain system settings to highest quality regardless of performance impact |
| r.MinScreenRadiusForCSMDepth | Threshold below which meshes will be culled from CSM depth pass. |
| r.MinScreenRadiusForDepthPrepass | Threshold below which meshes will be culled from depth only pass. |
| r.MinScreenRadiusForLights | Threshold below which lights will be culled. |
| r.MipMapLODBias | Apply additional mip map bias for all 2D textures, range of -15.0 to 15.0 |
| r.MobileContentScaleFactor | Content scale multiplier (equates to iOS's contentScaleFactor to support Retina displays |
| r.MobileHDR | 0: Mobile renders in LDR gamma space. (suggested for unlit games targeting low-end phones) 1: Mobile renders in HDR linear space. (default) |
| r.MobileHDR32bpp | 0: Mobile HDR renders to an FP16 render target. 1: Mobile HDR renders to an RGBA8 target. |
| r.MobileMSAA | Use MSAA instead of Temporal AA on mobile: 1: Use Temporal AA (MSAA disabled) 2: Use 2x MSAA (Temporal AA disabled) 4: Use 4x MSAA (Temporal AA disabled) |
| r.MotionBlur.Max | Allows to clamp the postprocess setting (max distortion caused by motion blur, in percent of the screen width) -1: don't clamp (default) |
| r.MotionBlur.Scale | Allows to scale the postprocess intensity/amount setting in the postprocess. 1: don't do any scaling (default) |
| r.MotionBlurFiltering | Useful developer variable 0: off (default, expected by the shader for better quality) 1: on |
| r.MotionBlurQuality | Defines the motion blur method which allows to adjust for quality or performance. 0:off, 1:low, 2:medium, 3:high (default), 4: very high |
| r.MotionBlurSoftEdgeSize | Defines how wide the object motion blur is blurred (percent of screen width) to allow soft edge motion blur. This scales linearly with the size (up to a maximum of 32 samples, 2.5 is about 18 samples) and with screen resolution Smaller values are better for performance and provide more accurate motion vectors but the blurring outside the object is reduced. If needed this can be exposed like the other motionblur settings. 0:off (not free and does never completely disable), >0, 1.0 (default) |
| r.MSAA.CompositingSampleCount | Affects the render quality of the editor 3d objects. 1: no MSAA, lowest quality 2: 2x MSAA, medium quality (medium GPU memory consumption) 4: 4x MSAA, high quality (high GPU memory consumption) 8: 8x MSAA, very high quality (insane GPU memory consumption) |
| r.NormalMapsForStaticLighting | Whether to allow any static lighting to use normal maps for lighting computations. |
| r.NumBufferedOcclusionQueries | Number of frames to buffer occlusion queries (including the current renderthread frame). More frames reduces the chance of stalling the CPU waiting for results, but increases out of date query artifacts. |
| r.OneFrameThreadLag | Whether to allow the rendering thread to lag one frame behind the game thread (0: disabled, otherwise enabled) |
| r.OnlyStreamInTextures | If set to 1, texture will only be streamed in, not out |
| r.OptimizeForUAVPerformance | Allows to profile if hardware has a performance cost due to render target reuse (more info: search for GCNPerformanceTweets.pdf Tip 37) If we see a noticeable difference on some hardware we can add another option like -1 (meaning auto) and make it the new default. 0: Optimize for GPU memory savings and reuse render targets (default) 1: Optimize for GPU performance (might render faster but can require more GPU memory) |
| r.ParallelInitViews | Toggles parallel init views. |
| r.ParallelShadows | Toggles parallel shadow rendering. Parallel rendering must be enabled for this to have an effect. |
| r.ParallelTranslucency | Toggles parallel translucency rendering. Parallel rendering must be enabled for this to have an effect. |
| r.ParallelVelocity | Toggles parallel velocity rendering. Parallel rendering must be enabled for this to have an effect. |
| r.ParticleLODBias | LOD bias for particle systems. Development feature, default is 0 |
| r.PostProcessAAQuality | Defines the postprocess anti aliasing method which allows to adjust for quality or performance. 0:off, 1:very low (faster FXAA), 2:low (FXAA), 3:medium (faster TemporalAA), 4:high (default TemporalAA), 5:very high, 6:max |
| r.PrecomputedVisibilityWarning | If set to 1, a warning will be displayed when rendering a scene from a view point without precomputed visibility. |
| r.PreTileTextures | If set to 1, textures will be tiled during cook and are expected to be cooked at runtime |
| r.PreventInvalidMaterialConnections | Controls whether users can make connections in the material editor if the system determines that they may cause compile errors 0: Allow all connections 1: Prevent invalid connections |
| r.PreViewTranslation | To limit issues with float world space positions we offset the world by the PreViewTranslation vector. This command allows to disable updating this vector. 0: disable update 1: update the offset is each frame (default) |
| r.ProfileGPU.Pattern | Allows to filter the entries when using ProfileGPU, the pattern match is case sensitive. '*' can be used in the end to get all entries starting with the string. '*' without any leading characters disables the pattern matching and uses a time threshold instead (default). '?' allows to ignore one character. e.g. AmbientOcclusionSetup, AmbientOcclusion*, Ambient???lusion*, * |
| r.ProfileGPU.Root | Allows to filter the tree when using ProfileGPU, the pattern match is case sensitive. |
| r.ProfileGPU.Screenshot | Whether a screenshot should be taken when profiling the GPU. 0:off, 1:on (default) |
| r.ProfileGPU.ShowUI | Whether the user interface profiler should be displayed after profiling the GPU. The results will always go to the log/console 0:off, 1:on (default) |
| r.PS4DumpShaderSDB | Whether to dump shader sdb files used for shader association. 0: Disabled 1: Enabled) |
| r.PS4MixedModeShaderDebugInfo | Whether to compile shaders to allow mixed mode shader debugging. This will currently generate slower code. 0: Normal mode 1: Mixed mode) |
| r.ReflectionEnvironment | 0:off, 1:on and blend with scene, 2:on and overwrite scene. Whether to render the reflection environment feature, which implements local reflections through Reflection Capture actors. |
| r.RefractionQuality | Defines the distorion/refraction quality which allows to adjust for quality or performance. <=0: off (fastest) 1: low quality (not yet implemented) 2: normal quality (default) 3: high quality (e.g. color fringe, not yet implemented) |
| r.RenderLastFrameInStreamingPause | If 1 the previous frame is displayed during streaming pause. If zero the screen is left black. |
| r.RenderTargetPoolMin | If the render target pool size (in MB) is below this number there is no deallocation of rendertargetsDefault is 200 MB. |
| r.RenderTargetPoolTest | Clears the texture returned by the rendertarget pool with a special color so we can see better which passes would need to clear. Doesn't work on volume textures and non rendertargets yet. 0:off (default), 1:on |
| r.RenderTargetSwitchWorkaround | Workaround needed on some mobile platforms to avoid a performance drop related to switching render targets. Only enabled on some hardware. This affects the bloom quality a bit. It runs slower than the normal code path but still faster as it avoids the many render target switches. (Default: 0) We want this enabled (1) on all 32 bit iOS devices (implemented through DeviceProfiles). |
| r.RenderTimeFrozen | Allows to freeze time based effects in order to provide more deterministic render profiling. 0: off 1: on (Note: this also disables occlusion queries) |
| r.RHICmdBasePassDeferredContexts | True to use deferred contexts to parallelize base pass command list execution. |
| r.RHICmdBypass | Whether to bypass the rhi command list and send the rhi commands immediately. 0: Disable, 1: Enable |
| r.RHICmdForceRHIFlush | Force a flush for every task sent to the RHI thread. 0: Disable, 1: Enable |
| r.RHICmdPrePassDeferredContexts | True to use deferred contexts to parallelize prepass command list execution. |
| r.RHICmdShadowDeferredContexts | True to use deferred contexts to parallelize shadow command list execution. |
| r.RHICmdTranslucencyPassDeferredContexts | True to use deferred contexts to parallelize base pass command list execution. |
| r.RHICmdUseParallelAlgorithms | True to use parallel algorithms. Ignored if r.RHICmdBypass is 1. |
| r.RHICmdUseThread | Uses the RHI thread. 0: Disable, 1: Enable |
| r.RHICmdVelocityPassDeferredContexts | True to use deferred contexts to parallelize velocity pass command list execution. |
| r.RHICmdWidth | Number of threads. |
| r.SceneColorFormat | Defines the memory layout (RGBA) used for the scene color (affects performance, mostly through bandwidth, quality especially with translucency). 0: PF_B8G8R8A8 32Bit (mostly for testing, likely to unusable with HDR) 1: PF_A2B10G10R10 32Bit 2: PF_FloatR11G11B10 32Bit 3: PF_FloatRGB 32Bit 4: PF_FloatRGBA 64Bit (default, might be overkill, especially if translucency is mostly using SeparateTranslucency) 5: PF_A32B32G32R32F 128Bit (unreasonable but good for testing) |
| r.SceneColorFringe.Max | Allows to clamp the postprocess setting (in percent, Scene chromatic aberration / color fringe to simulate an artifact that happens in real-world lens, mostly visible in the image corners) -1: don't clamp (default) |
| r.SceneColorFringeQuality | 0: off but best for performance 1: 3 texture samples (default)n |
| r.SceneRenderTargetResizeMethod | Control the scene render target resize method: (This value is only used in game mode and on windowing platforms.) 0: Resize to match requested render size (Default) (Least memory use, can cause stalls when size changes e.g. ScreenPercentage) 1: Fixed to screen resolution. 2: Expands to encompass the largest requested render dimension. (Most memory use, least prone to allocation stalls.) |
| r.ScreenPercentage | To render in lower resolution and upscale for better performance. 70 is a good value for low aliasing and performance, can be verified with 'show TestImage' in percent, >0 and <=100, <0 means the post process volume settings are used |
| r.ScreenPercentage.Editor | To allow to have an effect of ScreenPercentage in the editor. 0: off (default) 1: allow upsample (blurry but faster) and downsample (cripser but slower) |
| r.SeparateTranslucency | Allows to disable the separate translucency feature (all translucency is rendered in separate RT and composited after DOF, if not specified otherwise in the material). 0: off (translucency is affected by depth of field) 1: on costs GPU performance and memory but keeps translucency unaffected by Depth of Fieled. (default) |
| r.SetRes | Set the display resolution for the current game view. Has no effect in the editor. e.g. 1280x720w for windowed 1920x1080f for fullscreen 1920x1080wf for windowed fullscreen 1920x1080wm for windowed mirror |
| r.ShaderDevelopmentMode | 0: Default, 1: Enable various shader development utilities, such as the ability to retry on failed shader compile, and extra logging as shaders are compiled. |
| r.Shaders.KeepDebugInfo | Whether to keep shader reflection and debug data from shader bytecode, default is to strip. When using graphical debuggers like Nsight it can be useful to enable this on startup. |
| r.Shaders.Optimize | Whether to optimize shaders. When using graphical debuggers like Nsight it can be useful to disable this on startup. |
| r.Shadow.CachePreshadow | Whether preshadows can be cached as an optimization |
| r.Shadow.ConservativeBounds | Whether to use safe and conservative shadow frustum creation that wastes some shadowmap space |
| r.Shadow.CSM.MaxCascades | The maximum number of cascades with which to render dynamic directional light shadows. |
| r.Shadow.CSM.TransitionScale | Allows to scale the cascaded shadow map transition region. Clamped within 0..2. 0: no transition (fastest) 1: as specific in the light settings (default) 2: 2x larger than what was specified in the light |
| r.Shadow.CSMDepthBias | Constant depth bias used by CSM |
| r.Shadow.CSMDepthBoundsTest | Whether to use depth bounds tests rather than stencil tests for the CSM bounds |
| r.Shadow.CSMSplitPenumbraScale | Scale applied to the penumbra size of Cascaded Shadow Map splits, useful for minimizing the transition between splits |
| r.Shadow.DistanceFieldPenumbraSize | Controls the size of the uniform penumbra produced by static shadowing. This is a fraction of MaxTransitionDistanceWorldSpace in BaseLightmass.ini. |
| r.Shadow.DistanceScale | Scalability option to trade shadow distance versus performance for directional lights (clamped within a reasonable range). <1: shorter distance 1: normal (default) >1: larger distance |
| r.Shadow.DrawPreshadowFrustums | visualize preshadow frustums when the shadowfrustums show flag is enabled |
| r.Shadow.FadeExponent | Controls the rate at which shadows are faded out |
| r.Shadow.FadeResolution | Resolution in texels below which shadows are faded out |
| r.Shadow.FreezeCamera | Debug the shadow methods by allowing to observe the system from outside. 0: default 1: freeze camera at current location |
| r.Shadow.MaxResolution | Max square dimensions (in texels) allowed for rendering shadow depths. Range 4 to hardware limit. Higher = better quality shadows but at a performance cost. |
| r.Shadow.MinPreShadowResolution | Minimum dimensions (in texels) allowed for rendering preshadow depths |
| r.Shadow.MinResolution | Minimum dimensions (in texels) allowed for rendering shadow subject depths |
| r.Shadow.PerObject | Whether to render per object shadows (character casting on world) 0: off 1: on (default) |
| r.Shadow.PointLightDepthBias | Depth bias that is applied in the depth pass for shadows from point lights. (0.03 avoids peter paning but has some shadow acne) |
| r.Shadow.PreshadowExpand | How much bounds will be expanded when rendering a cached preshadow (0.15 = 15% larger) |
| r.Shadow.PreShadowFadeResolution | Resolution in texels below which preshadows are faded out |
| r.Shadow.PreShadowResolutionFactor | Mulitplier for preshadow resolution |
| r.Shadow.Preshadows | Whether to allow preshadows (static world casting on character) |
| r.Shadow.RadiusThreshold | Cull shadow casters if they are too small, value is the minimal screen space bounding sphere radius (default 0.03) |
| r.Shadow.RadiusThresholdRSM | Cull shadow casters in the RSM if they are too small, values is the minimal screen space bounding sphere radius (default 0.06) |
| r.Shadow.SpotLightDepthBias | Depth bias that is applied in the depth pass for per object projected shadows from spot lights |
| r.Shadow.SpotLightTransitionScale | Transition scale for spotlights |
| r.Shadow.TexelsPerPixel | The ratio of subject pixels to shadow texels |
| r.Shadow.TransitionScale | This controls the 'fade in' region between a caster and where his shadow shows up. Larger values make a smaller region which will have more self shadowing artifacts |
| r.Shadow.UnbuiltNumWholeSceneDynamicShadowCascades | DynamicShadowCascades to use when using CSM to preview unbuilt lighting from a directional light |
| r.Shadow.UnbuiltWholeSceneDynamicShadowRadius | WholeSceneDynamicShadowRadius to use when using CSM to preview unbuilt lighting from a directional light |
| r.Shadow.WholeSceneShadowUnbuiltInteractionThreshold | How many unbuilt light-primitive interactions there can be for a light before the light switches to whole scene shadows |
| r.ShadowQuality | Defines the shadow method which allows to adjust for quality or performance. 0:off, 1:low(unfiltered), 2:low .. 5:max (default) |
| r.ShowPrecomputedVisibilityCells | If not zero, draw all precomputed visibility cells. |
| r.ShowRelevantPrecomputedVisibilityCells | If not zero, draw relevant precomputed visibility cells only. |
| r.SimpleDynamicLighting | Whether to use simple dynamic lighting, which just renders an unshadowed dynamic directional light and a skylight. All other lighting features are disabled when true. This is useful for supporting very low end hardware. 0:off, 1:on |
| r.SkeletalMeshLODBias | LOD bias for skeletal meshes. |
| r.SkeletalMeshLODRadiusScale | Scale factor for the screen radius used in computing discrete LOD for skeletal meshes. (0.25-1) |
| r.SkinCaching | Whether or not to use the GPU compute skinning cache. This will perform skinning on a compute job and not skin on the vertex shader. GPUSkinVertexFactory.usf needs to be touched to cause a recompile if this changes. 0 is off(default), 1 is on |
| r.SSR.MaxRoughness | Allows to override the post process setting ScreenSpaceReflectionMaxRoughness. It defines until what roughness we fade the screen space reflections, 0.8 works well, smaller can run faster. (Useful for testing, no scalability or project setting) 0..1: use specified max roughness (overrride PostprocessVolume setting) -1: no override (default) |
| r.SSR.Quality | Whether to use screen space reflections and at what quality setting. (limits the setting in the post process settings which has a different scale) (costs performance, adds more visual realism but the technique has limits) 0: off (default) 1: low (no glossy) 2: medium (no glossy) 3: high (glossy/using roughness, few samples) 4: very high (likely too slow for real-time) |
| r.SSR.Temporal | Defines if we use the temporal smoothing for the screen space reflection 0 is off (for debugging), 1 is on (default) |
| r.SSS.Filter | 0: point filter (useful for testing, could be cleaner) 1: bilinear filter |
| r.SSS.SampleSet | 0: lowest quality 1: medium quality 2: high quality (default) |
| r.SSS.Scale | Experimental screen space subsurface scattering pass(use shadingmodel SubsurfaceProfile, get near to the object as the default) is human skin which only scatters about 1.2cm) 0: off (if there is no object on the screen using this pass it should automatically disable the post process pass) <1: scale scatter radius down (for testing) 1: use given radius form the Subsurface scattering asset (default) >1: scale scatter radius up (for testing) |
| r.StaticMeshLODDistanceScale | Scale factor for the distance used in computing discrete LOD for static meshes. (0.25-1) |
| r.Streaming.Boost | =1.0: normal <1.0: decrease wanted mip levels >1.0: increase wanted mip levels |
| r.Streaming.MipBias | 0..x reduce texture quality for streaming by a floating point number. 0: use full resolution (default) 1: drop one mip 2: drop two mips |
| r.Streaming.PoolSize | -1: Default texture pool size, otherwise the size in MB |
| r.Streaming.UseFixedPoolSize | If non-zero, do not allow the pool size to change at run time. |
| r.SubsurfaceQuality | Define the quality of the Screenspace subsurface scattering postprocess. 0: low quality for speculars on subsurface materials 1: higher quality as specular is separated before screenspace blurring (Only used if SceneColor has an alpha channel) |
| r.TemporalAASamples | Number of jittered positions for temporal AA (4, 8=default, 16, 32, 64). |
| r.TemporalAASharpness | Sharpness of temporal AA (0.0 = smoother, 1.0 = sharper). |
| r.TessellationAdaptivePixelsPerTriangle | Global tessellation factor multiplier |
| r.TextureStreaming | Allows to define if texture streaming is enabled, can be changed at run time. 0: off 1: on (default) |
| r.TiledDeferredShading | Whether to use tiled deferred shading. 0 is off, 1 is on (default) |
| r.TiledDeferredShading.MinimumCount | Number of applicable lights that must be on screen before switching to tiled deferred. 0 means all lights that qualify (e.g. no shadows, ...) are rendered tiled deferred. Default: 80 |
| r.TonemapperQuality | 0: low (minor performance benefit) 1: high (default, with high frequency pixel pattern to fight 8 bit color quantization) |
| r.TranslucencyLightingVolumeDim | Dimensions of the volume textures used for translucency lighting. Larger textures result in higher resolution but lower performance. |
| r.TranslucencyLightingVolumeInnerDistance | Distance from the camera that the first volume cascade should end |
| r.TranslucencyLightingVolumeOuterDistance | Distance from the camera that the second volume cascade should end |
| r.TranslucencyVolumeBlur | Whether to blur the translucent lighting volumes. 0:off, otherwise on, default is 1 |
| r.TranslucentLightingVolume | Whether to allow updating the translucent lighting volumes. 0:off, otherwise on, default is 1 |
| r.TranslucentVolumeFOVSnapFactor | FOV will be snapped to a factor of this before computing volume bounds. |
| r.TranslucentVolumeMinFOV | Minimum FOV for translucent lighting volume. Prevents popping in lighting when zooming in. |
| r.TriangleOrderOptimization | Controls the algorithm to use when optimizing the triangle order for the post-transform cache. 0: Use NVTriStrip (slower) 1: Use Forsyth algorithm (fastest)(default)2: No triangle order optimization. (least efficient, debugging purposes only) |
| r.UniformBufferPooling | If we pool object in RHICreateUniformBuffer to have less real API calls to creat buffers 0: off (for debugging) 1: on (optimization) |
| r.Upscale.Cylinder | Allows to apply a cylindrical distortion to the rendered image. Values between 0 and 1 allow to fade the effect (lerp). There is a quality loss that can be compensated by adjusting r.ScreenPercentage (>100). 0: off(default) >0: enabled (requires an extra post processing pass if upsampling wasn't used - see r.ScreenPercentage) 1: full effect |
| r.Upscale.Quality | 0: Nearest filtering 1: Simple Bilinear 2: 4 tap bilinear 3: Directional blur with unsharp mask upsample. (default) |
| r.Upscale.Softness | To scale up with higher quality loosing some sharpness 0..1 (0.3 is good for ScreenPercentage 90 |
| r.UseGetMeshElements | Whether to use the GetMeshElements path or the legacy DrawDynamicElements |
| r.UseMobileBloom | HACK: Set to 1 to use mobile bloom. |
| r.ViewDistanceScale | Controls the view distance scale. A primitive's MaxDrawDistance is scaled by this value. Higher values will increase view distance but at a performance cost. Default = 1. Value should be in the range [0.0f, 1.0f]. |
| r.ViewportTest | Allows to test different viewport rectangle configuations (in game only) as they can happen when using Matinee/Editor. 0: off(default) 1..7: Various Configuations |
| r.VirtualTexture | If set to 1, textures will use virtual memory so they can be partially resident. |
| r.VirtualTextureReducedMemory | If set to 1, the cost of virtual textures will be reduced by using a more packed layout. |
| r.VisualizeOccludedPrimitives | Draw boxes for all occluded primitives |
| r.VisualizeTexturePool | Allows to enable the visualize the texture pool (currently only on console). 0: off 1: on |
| r.VSync | 0: VSync is disabled.(default) 1: VSync is enabled. |
| r.WireframeCullThreshold | Threshold below which objects in ortho wireframe views will be culled. |