TNC CVar Catalog Rendering

TNC Cvars: Rendering

Generated by tools/build_cvar_wiki.py from the engine's own cvar registry. Edits made here are overwritten by the next run.

953 cvars of Wolfenstein II: The New Colossus — renderer, textures and video. Grouped by name prefix; prefixes with fewer than 10 cvars share the last table. Part of the TNC cvar catalogue, which explains how to set a cvar in this game and where the data comes from.

Exposed = accepted by the retail console and +name value launch arguments; every other cvar needs the resourceExec route.

image_

15 cvars.

Name	Type	Default	Exposed	Description
image_anisotropy	float	1	yes	set the maximum texture anisotropy if available
image_BC7DissableMode02	int	1	no	Dissable mode 02 for bc7 as it produces unstable output
image_BCCompressionQuality	int	2	no	Compression quality for BC* 0-5 ( 0 = best -- 4 = fastest ) ship with 0, develop with 2
image_debugSkipKraken	bool	0	no	don't compress mips with kraken for easier debugging
image_dropMips	bool	0	yes	drop mips on images marked as dropMip
image_filter	string	GL_LINEAR_MIPMAP_LINEAR	no	changes texture filtering on mipmapped images
image_HACK_generatingExplicit	bool	0	no	Don't assert during explicit generation
image_lodbias	string	0	no	change lod bias on mipmapped images
image_materialEncoding	int	1	yes	0 = BC+Oodle, 1 = WBP
image_mipFilter	int	2	no	0 = Old filter, 1 = Cubic, 2 = Mitchell, 3 = Lanczos
image_screenshotQuality	int	75	no	sets the compression quality for jpeg screenshots, 100 is best quality
image_streamingSeparatelyCompMip	int	32	no	smallest mip that is compressed individually in streamed images
image_useDxtEncodeJobs	bool	1	no	Set to use jobs for DXT encoding
image_useResizeFilterJobs	bool	1	no	use jobs for idFiler::Resize
image_watermark	string	(empty)	no	places a watermark each image with a watermark

r_

794 cvars.

Name	Type	Default	Exposed	Description
r_allowAsyncCompute	bool	1	yes	Enables async computing. No effect until game is restarted
r_allowParallelViewProcessing	bool	0	no	when enabled job chains for multiple render views will be allowed to run simultaneously.
r_ambientAtlasDepth	int	6	no	Ambient Octree atlas depth. This should not go over 256.
r_ambientAtlasHeight	int	256	no	Ambient Octree atlas height. This should not go over 256.
r_ambientAtlasWidth	int	256	no	Ambient Octree atlas width. This should not go over 256.
r_ambientBottomScale	float	0.1	no	Scales the ambient light contribution from below
r_ambientChannelScale	float	1	no	Scales the ambient light contribution
r_ambientDebugGPU	int	0	no	1 - Shows the octree leafs bounding boxes at center distance of r_ambientDebugGPUDistance from player. 2 - same as 0 plus draws probe ids in each cube corner to the closest probes. 3 - same as 1 and 2 + add the volume id of the cube. 4 - budgets
r_ambientDebugGPUDistance	int	4	no	Distance from the player to the center of a leaf bounding box.
r_ambientSHAtlasDepth	int	24	no	Ambient SH atlas depth.
r_ambientSHAtlasHeight	int	256	no	Ambient SH atlas height.
r_ambientSHAtlasWidth	int	256	no	Ambient SH atlas width.
r_amortizedSunCascades	int	1	no	Set to 1 to enable updating near-cascade every frame and alternating far cascade update.
r_analyticalLightFalloff	int	0	no	turn on analytical falloff for light rendering
r_antialiasing	int	6	yes	0 = Disabled, 1 = FXAA, 2 = SMAA, 3 = TAA (1TX), 4 = FXAA (1TX), 5 = SMAA (1TX), 6 = TSSAA (8TX), 7 = TSSAA (2X_8TX), 8 = TSSAA (4X_8TX), 9 = TSSAA (8X_8TX) 10 = MSAA (2X), 11 = MSAA (4X), 12 = MSAA (8X). Enables Anti-Aliasing modes
r_antialiasingEditorFXAA	bool	1	yes	0 = Disabled, 1 = enable override to use FXAA_1TX for editor
r_apiDump	bool	0	no	Write vulkan API dump
r_applyMorphAnimations	int	1	no	userchannel data drives blendshape/morph animation: 0 disabled, 1 enabled
r_applyMorphCount	int	0	no	max blendshapes to apply to a mesh, 0 = unlimited
r_asyncClears	bool	0	no	Use async clear for non-esram/uncompressed color targets (currently: object motion vectors / caustics targets )
r_asyncDeferredPassesEarlyMaxWaves	float	0.2	no	Max number of waves for post processes up until scattering
r_asyncDeferredPassesLateMaxWaves	float	0.4	no	Max number of waves for after and including scattering
r_asyncDeferredPassesPrepare	bool	0	no	1: Enable async deferred passes prepare
r_asyncGuiClears	bool	0	no	—
r_asyncGuiClearsMaxWaves	float	0.2	no	Max number of waves for async GUI clears
r_asyncLightScattering	bool	0	no	Unsuported atm.
r_asyncPostProcess	bool	1	no	Use an async compute shaders for post process
r_asyncPostProcessEarlyMaxWaves	float	0.3	no	Max number of waves for post processes up until tone mapping
r_asyncPostProcessLateMaxWaves	float	0.7	no	Max number of waves for post processes after and including tone mapping
r_asyncWater	bool	0	no	Enable/disable async computer water processing
r_asyncWaterMaxWaves	float	0.2	no	Max number of wavefronts for water stages
r_atlasClearTextureOnResourceDelete	bool	1	no	clear atlas area when the corresponding resource is deleted
r_augmentSilhouette	bool	1	no	0 = disabled, 1 = enabled (default)
r_beamsOrientTravelDir	bool	1	no	use the beam direction to compute each segment's tangent vector
r_blurGaussian	int	1	no	0 = Disabled, 1 = Enabled, 2 = Force
r_blurRadial	bool	1	no	0 = Disabled, 1 = Enabled
r_brightness	float	1.0	no	changes the brightness value
r_buffersForceSyncMap	bool	0	no	—
r_calibrateGamma	float	0	no	if > 0 draw a swatch at a specific luminance for calibrating gamma
r_calibrateGammaLum	float	0	no	luminance to display for calibrating gamma [0,255]
r_calibrateGammaMode	float	0	no	0=measure total luminance, 1=measure red, 2=measure green, 3=measure blue
r_checkBounds	bool	0	no	compare all surface bounds with precalculated ones
r_chromaticAberration	int	1	yes	enable chromatic aberration
r_chromaticAberrationLimit	float	1.0	no	enable chromatic aberration
r_clampToEdgeRenderView	bool	1	no	Clamp edge of framebuffer to emulate fullscreen sampling while using resolution scaling. 0 = Disabled, 1 = Enabled
r_clear	string	2	no	force screen clear every frame, 1 = purple, 2 = black, 'r g b' = custom
r_clearColorTarget	int	0	no	used to enforce a color clear e.g. in editor views with no full color target overdraw
r_clearEditorAlpha	float	0.0	no	background overlay amount for the editor model-viewer
r_clearTargetsAfterScaleDown	bool	1	no	Clear target that are affected by resolution scaling to prevent artifacts. 0 = Disabled, 1 = Enabled
r_clusterDataUseTransferQueue	bool	1	no	Use transfer queue to upload cluster data
r_clusteredDecalsClippedBounds	bool	0	no	1: clips decal volume to frustum. 0 no clipping
r_clusteredDecalsRegisterMode	int	0	no	0: default, OBB screen projection and zmin/zmax, 1: enable volumes registration refinement step
r_clusteredDeferredRendering	bool	0	yes	Use clustered deferred rendering path, else uses forward. CVar can be changed at runtime to verify visual, but for performance check, please enable from cmdline.
r_clusteredDeferredRenderingPreZ	int	2	no	1: preZ alpha test. 2 PreZ alpha test + static opaques
r_clusteredLightingDebug	int	0	no	0: disabled, 1: cluster parent frustums in world space 2: clusters frustums in world space 3: clusters frustums in screen space
r_clusteredLightingFarZ	float	256.0f	no	Set pseudo far z to be used for zslices distribution
r_clusteredLightingNearZ	float	2.0f	no	Set pseudo near z to be used for zslices distribution
r_clusteredLightingRegisterMode	int	0	no	0: default, OBB screen projection and zmin/zmax, 1: enable volumes registration refinement step, 2: compute based, 3: async compute based
r_clusteredLightingShowStall	bool	0	no	Print duration of wait on cluster jobs, if any.
r_clusteredLightingTilesX	int	16	no	Clustered lighting tile count in frustum X direction
r_clusteredLightingTilesY	int	8	no	Clustered lighting tile count in frustum Y direction
r_clusteredLightingTilesZ	int	24	no	Clustered lighting tile count in frustum Z direction
r_clusteredLightingUseClipping	bool	1	no	Use light clipping plane/box during cluster assignment.
r_clusteredLightingUseJob	bool	1	no	Use jobs to register lights
r_colorBlindMode	int	0	yes	Enable color blind modes. 1: Protanopia, 2: Deuteranopia, 3: Tritanopia. ( tbd: Protanomaly, Deuteranomaly, Tritanomaly ). Disabled by default = 0
r_colorCorrection	int	1	no	enable color correction
r_commitSurfacesRenderProgDataJobGranularity	int	8	no	Render model granularity for commit surface render prog data jobs
r_contrast	float	1.0	no	changes the contrast value
r_currentMaterial	string	(empty)	no	current material name from r_showSurfaceInfo
r_cutsceneRender	bool	0	no	if set we are rendering a cutscene and the renderer should attempt no monkey business with changing cvars
r_debugArrowStep	int	120	no	step size of arrow cone line rotation in degrees
r_debugBounds	float	0	no	How long largest axis needs to be to show in commit
r_debugContextVerbosity	int	2	no	0: No messages, 1: Errors, 2: Errors & Warnings, 3: Errors, Warnings and Informational
r_debugDrawAlpha	float	0.5	no	alpha for r_show* debug tools, where implemented and applicable
r_debugFlareOcclusionQuad	int	0	no	display occlusion quad debug
r_debugHitTraceModels	int	0	no	+/-1: draws hit test spheres on test, +/-2: draws every frame, < 0 for depth test
r_debugImageQuality	int	0	no	Enable image quality debug view, >= 1 antialiasing debug views, increasing debug view number increases zoom factor, >= 10 pixel density, >= 20 gamma debug
r_debugInvert2ndView	bool	0	no	Invert the second render view in order to debug multiview
r_debugJobDAG	int	0	no	For specified number of frames writes out a GraphViz dot file of the render job DAG
r_debugLightingVtRestart	bool	0	no	If 1, issue vt_restart command on changing debugLighting mode
r_debugLineDepthTest	bool	0	no	perform depth test on debug lines
r_debugLineWidth	int	1	no	width of debug lines
r_debugNewGPUParticles	bool	0	no	—
r_debugPolyDepthTest	bool	0	no	perform depth test on debug polys
r_debugPolygonFilled	bool	1	no	draw a filled polygon
r_debugSphereDist	float	90	no	Distance from camera of testCubeImage sphere
r_debugToolsHDR	bool	0	no	Enable drawing debug tools in HDR for debugging HDR content ( e.g. Ambient Probes )
r_debugTransientDecals	bool	0	no	Print how many decals are active and in what range
r_decalClusteringNumDecalsPerJob	int	32	no	Ideal number of decals per job
r_decalClusteringUseJob	bool	1	no	Use jobs to compute decal bounds
r_decalDecalBoundsJobGranularity	int	32	no	Granularity of decal bounds jobs
r_decalDistanceExtendedFadeStart	int	10	no	Start fading out extended decals at this range.
r_decalDistanceFadeMultiplier	float	1.0	yes	Decal fade distance multiplier.
r_decalDistanceFadeStart	int	60	no	Start fading out all decals at this range.
r_decalFadeCulling	int	2	yes	1. Starts culling decals based on area of visibility on screen after the distance specified on r_decalDistanceFadeStart. Culling starts once decal has r_decalPixelThresholdFadeStart on screen and ends when it reaches r_decalPixelThresholdFadeEnd pixels. 2 - Also cull decals that face normals facing away from player.
r_decalFadeCullingAngle	float	-0.3	no	Threshold value for the dot product between decal normal and player view direction. Range is [-1,1].
r_decalFadeFilterId	int	0	no	Displays just the decal id wanted.
r_decalFilteringQuality	int	2	yes	0 - Low (trilinear). 1 - Medium (aniso 2x). 2 - High (aniso 4x). 3 - Ultra (aniso 8x). 4 - Nightmare (aniso 16x)
r_decalLifetimeMultiplier	float	1.0	yes	Global lifetime multiplier for timed decals.
r_decalPixelThresholdFadeEnd	int	115	no	Number of width pixels at which the decal will be completely faded out.
r_decalPixelThresholdFadeStart	int	125	no	Number of width pixels at which the decal will start fading out.
r_decalRenderedDebug	int	0	no	1. Show how many decals are rendered. 2. Draws the OBB and normal. 3. Draws the id.
r_decals	bool	1	no	enable decal rendering
r_defaultDepthBias	float	1.0	no	default polygonOffset units
r_defaultSlopeScaleDepthBias	float	2.0	no	default polygonOffset factor
r_deferDescriptorSetUpdates	bool	1	no	Defer descriptor set updates if set to true
r_deferedRenderModelMode	int	-1	no	Set before loading a map -1 == do nothing, 0 == force defered off, 1 == force defered on.
r_depthTestCollisionSpheres	bool	0	no	1 = use depth test to render collision spheres
r_detailDistance	float	300	no	Distance at which to start culling detail geometry
r_dialogIconX	float	0.875	no	horizontal screen fraction at which the loading icon is placed
r_dialogIconY	float	0.795	no	vertical screen fraction at which the loading icon is placed
r_directChannelScale	float	1	no	Scales the direct light contribution
r_disableDebugRender	bool	0	no	if true disable all debug render lines, boxes, etc
r_discardLargeStatShadowsAtLoadHack	int	-1	no	throw away large static shadows at load to temporarily get around out of videoram on consoles
r_displayMidnightStoryBoards	bool	0	no	display midnight storyboards in-game
r_displayRefresh	int	0	yes	optional display refresh rate option for vid mode
r_dof	int	1	yes	0 = Disabled, 1 = Enabled, 2 == Enable DOF near field
r_dofDebug	int	0	no	0 = Disabled, 1 = Show DoF debug info
r_dofHalfRes	bool	1	no	0 = full res, 1 = half res ( default )
r_dofNumTapsSideBlur	int	8	no	—
r_dofNumTapsSideFloodFill	int	4	no	—
r_dofTAA	bool	1	yes	0 = Disabled, 1 = Enabled
r_dofTAABlendRatio	float	0.15	no	DOF Temporal AA blend ratio
r_drawFCATBar	bool	0	yes	Enable drawing of FCAT bar
r_drawLoadedImages	int	0	no	Draw loaded images scaled on the screen
r_drawLoadedImagesScale	float	64	no	Shrink factor for r_drawLoadedImages
r_driverDebugLayerVerbosity	int	3	no	0 = disabled, 1 = print low severity (info, all warnings, errors), 2 = print medium severity (performance warnings and errors), 3 = print high severity (errors or dangerous undefined behavior)
r_dumpShaderText	bool	0	no	dump the shadertext on load to generated/renderprogs/textdump/
r_durangoUseInstalledHlslCompiler	bool	0	no	use the durango hlsl compiler from the installed sdk instead the checked in one
r_dynamicLightingScale	float	1	no	Scales all approximated lighting on dynamic models
r_enableApproximateLightingForPoint	int	1	no	0 = off, 1 = amb + lights, 2 = amb only, 3 = lights only
r_enableBlendShapes	bool	1	no	0 to disable, 1 to enable
r_enableVideoOverlay	bool	1	no	enable (subtitle) video overlay
r_envBlendPowerFilterSize	float	1	no	Radius of specular power bleeding filter
r_envBlendShow	int	0	no	report the dynamic blend factors
r_envBlendTime	float	-1.0	no	if >= 0, the seconds to completely fade out an old environment
r_environmentProbes	int	1	no	0 - off, 1 - enable specular environment probes. 2- also enable diffuse environment probes. 3 - probe generation preview mode
r_excludeModel	int	-1	no	exclude only this model
r_explosivesDebug	bool	0	yes	draws debug geometry for idProp_Exposive based entities
r_fastPathReInsert	bool	0	no	Use RelinkModelIntoTree
r_filmGrainRatio	float	1.0	yes	Film grain - allow overriding art settings
r_finishEveryDraw	bool	0	no	call glFinish() or equivalent after each draw call. Does not work with r_useSMP = 2.
r_firstCutsceneFrame	bool	0	no	Marks the next frame as the first frame in a cutscene rendering session
r_fixObjMaterialPath	bool	0	no	temp hack fix for material paths being jumbled up for objs coming out of modo
r_flaresQuality	int	1	no	Flares quality (resolution & format precision). 0: 1/2 res (64bit), 1: 1/4 res (32bit), 2: 1/8 res (32bit)
r_flatShadeOnMapLoad	bool	0	no	set to flat shade on all map loads
r_flatShadeOnMissingMegas	bool	1	no	draw missing megatextures as flat shaded upon loading map
r_foliageBigFadeDistMax	float	7000	no	all foliage will be completely faded out at this range
r_foliageSmallFadeDistMax	float	5000	no	small foliage will be completely faded out at this range
r_foliageStartFadeDist	float	3000	no	start fading foliage out at this range
r_forceFullVirtualTextureLoad	bool	0	no	force each frame to do a full virtual texture load for capturing video, etc * SLOW *
r_forceLod	int	-1	no	render a specific lod only
r_forceNonDeferredAnimBlend	string	0	no	1 = force non-deferred animation blending
r_forceOcclusionBoxQueries	bool	0	no	force the use of occlusion query boxes
r_forceReferencesOnCommit	bool	0	no	Never skip the reference update
r_forceUpdateAllSunCascades	int	0	no	Set to 1 to force all sun cascades to be updated each frame ( used for cutscene rendering ).
r_freeTriangleCPUData	bool	1	no	dump the CPU data after building vertex buffers
r_fuelTrailEffects	bool	1	no	fuel trail particle effects
r_fullscreen	int	1	yes	0 = windowed, 1 = full screen, 2 = borderless window
r_gamma	float	1.0	yes	changes the user defined gamma value of the display device - the default value 1.0 has no effect on the brightness of the image
r_generatedRenderprogPath	string	(empty)	no	if set to a folder it will be added as an alternative source for shaders to load from
r_generateMipMaps	int	1	no	0 = overlapped downsample, 1 = glGenerateMipMap
r_gpuParticleStats	bool	0	no	—
r_gpuParticleUseAsyncCompute	bool	1	no	—
r_gpuTriangleCullingAsyncCompute	bool	1	no	use async compute for GPU triangle culling
r_gpuTriangleCullingDebug	int	0	no	1 Get the number of triangles culled. 2 Render only GPU culled geometry
r_gpuTriangleCullingGranularity	int	1	no	Granularity of the compute waves. 0: 64 tris/wave - 1: 256 tris/wave; 2: 512 tris/wave
r_gpuTriangleCullingMaxJobs	int	0	no	Maximum number of concurrent gputriangle culling jobs to allow
r_gpuTriangleCullingMaxWaves	int	2	no	—
r_gpuTriangleCullingOptions	int	4	no	0. No culling. 1. Frustum culling 2. Backface culling 3. Small triangles culling. 4. All
r_guiDebug	bool	0	no	Provides debug output in idRenderModelGui code
r_guiFocusBlur	float	0	no	Debug, set gui focus blur
r_guiMotionBlurX	float	0	no	Debug, set gui motion blur X axis
r_guiMotionBlurY	float	0	no	Debug, set gui motion blur Y axis
r_hairCullingDistance	float	27	no	Cull all hair after this distance in meters
r_hdrAutoExposureBase	float	0.01	no	0.01 (default). Set base exposure adjustment. Helpful for screenshots and fine tunning automated exposure behavior.
r_hdrAutoExposureInstant	bool	0	no	0 = use r_hdrAutoExposureSpeed, 1 = Auto exposure happens instantly ( as if r_hdrAutoExposureSpeed == infinity )
r_hdrAutoExposureRatio	float	1.0	no	1.0 = use full adjustment. 0 = use base exposure adjustment. Helpful for screenshots and fine tunning automated exposure behavior.
r_hdrAutoExposureSpeed	float	1.0	no	0 = disable, 1.0 = 16.5 ms (default), 2.0 = 8.25 ms, etc
r_hdrBloom	bool	1	yes	0 = disabled, 1 = enabled (default)
r_hdrBloomRatio	float	0.03	no	0 = disable, 0.03 = 3.0% (default)
r_hdrBloomThreshold	float	2.0	no	Luminance threshold for bloom. -1 = all pixels contribute for bloom
r_hdrDebug	int	0	no	0 = Disabled, 1 = Non-tonemapped gamma corrected output, 2 = debug view (red = nans, green = negative, grayscale = valid), 3 = range debug (red = above max range, blue = bellow min range), 4 = Auto Exposure + Bloom debug
r_hdrDebugRangeMax	float	1.0	no	Max range used for r_hdrDebug mode 3
r_hdrDebugRangeMin	float	0.01	no	Min range used for r_hdrDebug mode 3
r_hdrPostProcess	bool	1	no	0 = disabled, 1 = enabled (default)
r_hdrShutterSpeed	float	10.0	yes	Set camera shutter speed. Default = 10 (1/10 th of a second)
r_highlightFrequency	float	0.75	no	Hz of highlight pulse
r_highlightFrequencyFast	float	2	no	Hz of highlight pulse
r_highlightSynchronized	bool	0	no	All highlights in phase
r_hotPatchWarningsAreErrors	bool	0	no	set to 1 to throw a resource ERROR if hot patching materials fails.
r_imageAtlasDecalHeight	int	8	yes	Hight of decal image atlas (in k, must be pot)
r_imageAtlasDecalWidth	int	8	yes	Width of decal image atlas (in k, must be pot)
r_imageAtlasForceDropMip	bool	0	yes	force dropmip for image atlas resources > 128px
r_imageAtlasLightHeight	int	1	yes	Hight of light image atlas (in k, must be pot)
r_imageAtlasLightWidth	int	2	yes	Width of light image atlas (in k, must be pot)
r_imageAtlasMaxAniso	float	4	yes	Max aniso for atlas textures.
r_imageAtlasTranSortHeight	int	8	yes	Hight trans sort of image atlas (in k, must be pot)
r_imageAtlasTranSortWidth	int	8	yes	Width of trans sort image atlas (in k, must be pot)
r_imageCompareMultiplier	float	1.0	no	This is a multiplier for the difference between the two images in the ImageCompare command
r_incrementalUpdate	bool	1	no	—
r_initialModeHeight	int	1080	yes	Default Vertical Resolution on first start ever - Only gets used if r_mode = -2
r_initialModeWidth	int	1920	yes	Default Horizontal Resolution on first start ever - Only gets used if r_mode = -2
r_initialMonitor	int	-1	yes	-1 = auto detect default monitor, >= 0 specific monitor index used to initialize the render system
r_jitter	bool	0	no	randomly subpixel jitter the projection matrix
r_jointAxisScale	float	0.02	no	scale of joint axes
r_jointName	string	(empty)	no	name of model joint to render
r_jointNameOffset	float	0.01	no	offset of joint names when r_showskel is set to 1
r_jointNameScale	float	0.02	no	size of joint names when r_showskel is set to 1
r_jointSubString	string	(empty)	no	render all joints with this substring in the name
r_lazyBindBuffers	bool	1	no	lazy buffer binds
r_lazyBindImageBuffers	bool	1	no	lazy buffer image binds
r_lazyBindParms	bool	1	no	lazy update of render parms
r_lazyBindPrograms	bool	1	no	lazy program binds
r_lazyBindTextures	bool	1	no	lazy texture binds
r_lensDirtRatio	float	1.0	no	Set ratio for lens dirt composition
r_lensFlaresRatio	float	1.0	no	Set ratio for lens flare composition
r_lightClassThreshold	int	2	yes	Set which light classes are rendered: <= 1 - only key lights, 2 - key + some additional lights, >= 3 - all lights
r_lightDistanceFadeMultiplier	float	1.0	yes	Light fade distance multiplier.
r_lightScattering	bool	1	no	1: Enable light scattering, 0: disable
r_lightScatteringAnisotropy	float	0.5	no	Light scattering anisotropy
r_lightScatteringClearBuffers	bool	0	no	1: Enable light scattering clearing all buffers
r_lightScatteringFarPlane	float	100.0	no	Light scattering far clipping plane: 100 meters default ( higher view distances will require increase r_lightScatteringVolumeDepth )
r_lightScatteringNearPlane	float	0.1	no	Light scattering near clipping plane
r_lightScatteringQuality	int	1	yes	Light scattering volume width: 3: screenres / 8 x 128 depth (high) 2: screenres / 16 x 128 depth (high UDH) 1: screenres / 8 x 64 (medium), 0: screenres / 16 x 64 (low). >=1440p uses 0 by default
r_lightScatteringQualityCinematic	int	0	no	Light scattering uxing x8 samples. This is reserved for cinematics
r_lightScatteringQualityResClamp	int	2560	no	Light scattering resolution clamping. Anything above r_lightScatteringQualityResClamp uses ( screenres / 16 ) tile size
r_lightScatteringTAA	bool	1	no	1: Enable light scattering TAA, 0: disable
r_lightScatteringTAABlendRatio	float	0.05	no	Light scattering temporal blend ratio
r_loadingIconSize	float	0.07	no	size of the loading icon
r_loadingIconSpeed	float	0.5	no	rotation speed of the loading icon
r_lockView	bool	0	no	lock the current view
r_LODAutoGenerate	bool	1	no	enables auto generation of LODs
r_LODAutoGenerateThreshold	int	0	no	If the number of vertices between LODs are less or equal than this, no LODs are generated.
r_LODMinVertCount	int	50	no	minimum number of vertices needed so that LODs get created
r_lodScaleDeviation	float	0.5	yes	Scale LOD deviation override
r_lodScaleDeviationNeo	float	1.0	yes	Scale LOD deviation override for Neo
r_lodScaleDeviationScorpio	float	1.0	yes	Scale LOD deviation override for Scorpio
r_LODSkipGen	int	2	no	0: normal, 2: skip, skip generating lods or generate at low quality to improve load times on workstation when cooker lags behind
r_LODSkipOnLargeModelsMaxVerts	int	500000	no	skip creating lods on models with more than this many vertices
r_LODSkipOnLargeModelsVersionDepCVar	int	1	no	a cvar that all skiped models depends on, so we easily can increment it when we make the test more sophisticated
r_logFile	int	0	no	number of frames to emit render logs
r_logLightTris	int	0	no	if 0 do nothing, if 1 print all light triangle interaction numbers, if > 1 print light interactions more than nK triangles. r_logLightTris 100 will print interactions touching more than 100K triangles
r_logRenderWorldLinkFailures	bool	0	no	Log a warning for every model and light that fail to link in the render world
r_manifoldMaxDetailWidth	float	0.64f	no	—
r_MaterialStream	bool	1	no	—
r_mergeModelSurfaces	bool	1	no	combine model surfaces with the same material
r_MidnightStoryBoardsScale	float	0.15	no	midnight storyboards in-game placeholder size (in percent of screen size)
r_mip3Border	bool	1	no	use clampToBorder for the mip3 buffers
r_mode	int	-2	yes	Select the resolution and monitor to use in fullscreen mode (hit TAB to view available modes)
r_modelToIndexGranularity	int	128	no	The number of models to process in a single coop job pass
r_morphDisplacementNormalWeight	float	0	no	1 = Weight morph normals heavily based on displacement, 0 = weight normals the same as displacement
r_morphsCullingDistance	float	10	no	Cull all morphs after this distance in meters.
r_motionblur	int	1	yes	0 = Disabled, 1 = Enabled
r_motionblurMotionPixelThreshold	float	1	no	1 = 1 pixel threshold motion, 2 = 2 pixel, etc
r_motionBlurQuality	int	2	yes	Set motion blur quality. 2: high, 1; medium, 0: low
r_multiSamples	int	0	yes	number of anti-aliasing samples (0 = no multi-sampled FBO)
r_newCutsceneFrame	bool	1	no	Marks the next frame as the start of a new cutscene frame
r_newCutsceneFrameBlend	bool	0	no	Marks the next frame as the first blend frame of a new cutscene frame
r_newGpuParticleCullingOptions	int	0	no	0. No culling. 1. Frustum culling 2. Small triangles culling. 3. All Modes 4. Cull Always
r_newGpuParticleUseAsyncCompute	bool	1	no	—
r_opaqueSurfacesSortMode	int	3	no	0 = render program; 1 = depth; 2 = model; 3 = pipeline
r_openglUseValidator	bool	0	no	make sure opengl shaders pass validation before allowing usage (disabled because validator usually doesn't support all extensions we use)
r_opticsMaxPixelBlurLimit	float	14	no	Value in pixels. ( 14 default )
r_orbisUseInstalledPsslCompiler	bool	0	no	use the orbis pssl compiler from the installed sdk instead the checked in one
r_parallelCommitSurfacesRenderProgData	bool	1	no	Run the commit surfaces render prog data jobs in parallel with cull & parts of gather
r_parallelTextureStreamingUpdate	bool	1	no	Run the UpdateTextureStreaming job in parallel with cull & gather
r_particleCoopJobCount	int	8	no	The number of jobs to break the particle coop work up into
r_particleFadeQualityMultiplier	float	1.0	yes	—
r_particleFadeQualityMultiplierDurango	float	0.8	yes	—
r_particlesCullByVertexAlpha	bool	1	no	if particle alpha in vertex shader is below epsilon, the particle will be culled
r_particlesEmissionScale_FarDist	float	70	no	Far emitter distance triggering emission rate scaling
r_particlesEmissionScale_FarScale	float	0	no	Particles min alpha threshold for discarding (0-255 sRGB range )
r_particlesEmissionScale_ForceDefaultRate	bool	0	no	Force default rate on all emitters
r_particlesEmissionScale_NearDist	float	68	no	Near emitter distance triggering emission rate scaling
r_particlesEmissionScale_NearScale	float	1	no	Near emission rate scaling factor
r_particlesLightAtlas	bool	1	no	Particles light atlas update. 0 disables update
r_particlesLightAtlasDebug	int	0	no	Draw particle quads. 1 - Green - LOD0, yellow - LOD1, brown - LOD2, red - LOD3. 2 shows tile occupancy inside the bands.
r_particlesLightAtlasHeight	int	2048	yes	Particles light atlas height
r_particlesLightAtlasQuality	int	1	yes	Particles light atlas quality. 0 - low, 1 - medium, 2 - high, 3 - ultra.
r_particlesLightAtlasVTSThreashold	float	0.02	no	Tile size for particle light atlas decreases after this threshold.
r_particlesLightAtlasWidth	int	4096	yes	Particles light atlas width
r_particlesMinAlpha	float	16	yes	Particles min alpha threshold for discarding (0-255 sRGB range )
r_particlesMinAlphaDurango	float	32	yes	Particles min alpha threshold for discarding (0-255 sRGB range )
r_particlesSurfaceSpawn	bool	1	no	Particles surface spawn update. 0 disables update
r_pbrDebug	int	0	no	0 = Disabled, 1 = albedo, 2 = tbd...
r_pbrDebugAlbedoMax	float	0	no	Albedo max luminance (sRGB) to compare against
r_pbrDebugAlbedoMetalThreshold	float	89	no	Albedo max luminance (sRGB) to compare against, for metals
r_pbrDebugAlbedoMin	float	0	no	Albedo min luminance (sRGB) to compare against
r_pbrDebugAlbedoMultiplier	float	1.0	no	For debugging purposes, apply multiplier to albedo
r_pbrDebugAlbedoNonMetalThreshold	float	89	no	Albedo max luminance (sRGB) to compare against, for non-metals
r_pbrDebugFogDensityScale	float	1	no	Apply a scale to fog density - for debugging purposes
r_pbrDebugLightingSaturation	float	1.0	no	For debugging purposes, apply lighting saturation. Bellow 1.0 desaturates.
r_pbrDebugSmoothnessGamma	float	1.0	no	For debugging purposes, apply gamma curve to smoothness
r_pbrDebugSmoothnessMultiplier	float	1.0	no	For debugging purposes, apply multiplier to smoothness
r_perfSaveFile	string	PerfStats.txt	no	Name of output file for r_perfSaveStatsToFile, if empty, PerfStats.txt will be used
r_perfSaveStatsToFile	int	0	no	Writes the current perf stats to a file
r_physicalPagesAspectRatio	float	2.0	yes	Set to change the aspect ratio of physical pages for the uniquemap.
r_playerCullSurfacesCPU	int	0	no	Test and cull marked player surfaces (legs/boots)
r_primeLightScale	float	1	no	Scales prime light contribution
r_printHotPatchWarnings	bool	0	no	set to 1 to spam hotpatch warnings.
r_printOnRunTimePipelineGeneration	bool	0	no	print a message when pipelines are generated runtime
r_printRenderFlagsOnce	bool	0	no	prints the active renderflags once when triggered
r_purgeShaderTextOnLoad	bool	0	no	Free memory used by shader text when loading shaders
r_queryClustersForFrustum	bool	0	no	Use more (spatially) accurate query for lights
r_queryClustersInBatches	bool	0	no	Query clusters in batches for better performance
r_queryClustersInTree	bool	0	no	Use AABB tree to gather Umbra clusters
r_queryThreshold	int	4	no	minimum return value from a query
r_refractionForceRefreshThreshold	float	400.0f	no	Surfaces whose depth delta is smaller than this value will share the same refraction
r_refractionMaxUpdates	int	2	no	—
r_refractionMinSurfaceWeight	float	0.001	no	Min refractive surface size on screen to trigger refraction updates
r_refractionUpdatesDebug	bool	0	no	—
r_remapStaticModelSurfaces	bool	0	no	Try to remap surfaces into genericvmtr if possible
r_renderGatherDecalsGranularity	int	16	no	The number of decals a worker thread will consume on each pass. Too low and memory contention becomes an issue. Too high and completion times spike.
r_renderGatherMode	int	0	no	render gather mode: 0 = Parallelized, 1 = Parallel code, but called serially
r_renderGatherModelsGranularity	int	4	no	The number of areas a worker thread will consume on each pass. Too low and memory contention becomes an issue. Too high and completion times spike.
r_renderGatherWorldAreaGranularity	int	8	no	The number of areas a worker thread will consume on each pass. Too low and memory contention becomes an issue. Too high and completion times spike.
r_renderLods	int	1	no	renderer uses lods
r_renderProgResourceUsageWarnings	int	2	no	0 = no output, 1 = warnings, 2 = error
r_renderThreadIdealProcessor	int	-1	no	HACK for NVIDIA only - Forces an ideal thread for render thread to mitigate driver issues
r_renderThreadStackSizeKB	int	4194304	no	Render thread stack size
r_saturation	float	1.0	no	changes the saturation value
r_sb_deferredNormalBiasScale	float	0.04	no	Deferred scale for normal depth bias
r_sb_deferredNormalBiasScaleMax	float	10	no	Deferred max depth normal bias
r_sb_depthBias	float	1	no	depth bias units for drawing shadow buffer
r_sb_depthOffset	float	0.002	no	polygon depth offst for drawing shadow buffer
r_sb_occluderFacing	int	0	no	0 = front faces, 1 = back faces, 2 = front and back sides separately with polygon offset
r_sb_parallelDepthBias	float	0.01	no	depth bias for drawing parallel light shadow buffer
r_sb_parallelSlopeScaleDepthBias	float	2.5	no	slope scale depth bias for drawing parallel light shadow buffer
r_sb_playerShadowShrink	float	0.5	no	scale of projected player shadow map
r_sb_sliceOffset	float	0	no	Offset the cascade start point along view direction
r_sb_sliceRampScale	float	10	no	larger values perform a faster dithered lerp between slices
r_sb_sliceStep	float	512	no	initial ortho size for first slice
r_sb_slopeScaleDepthBias	float	2.5	no	slope scale depth bias for drawing shadow buffer
r_screenshotBatchSize	int	8	no	Number of jobs issue when batching screenshots (min 1, max 16)
r_screenshotTag	string	(empty)	no	String to display on captured screenshots
r_shaderLimits_blendedSurfaces_numLateAlloc	int	0	no	—
r_shaderLimits_blendedSurfaces_numPS	int	-1	no	—
r_shaderLimits_blendedSurfaces_numVS	int	-1	no	—
r_shaderLimits_blendedSurfaces_VSLockThreshold	int	-1	no	—
r_shaderLimits_renderOpaque_dynamicModels_numLateAlloc	int	32	no	—
r_shaderLimits_renderOpaque_dynamicModels_numPS	int	-1	no	—
r_shaderLimits_renderOpaque_dynamicModels_numVS	int	3	no	—
r_shaderLimits_renderOpaque_dynamicModels_PSLockThreshold	int	-1	no	—
r_shaderLimits_renderOpaque_dynamicModels_VSLockThreshold	int	-1	no	—
r_shaderLimits_renderOpaque_viewModels_numLateAlloc	int	32	no	—
r_shaderLimits_renderOpaque_viewModels_numPS	int	-1	no	—
r_shaderLimits_renderOpaque_viewModels_numVS	int	3	no	—
r_shaderLimits_renderOpaque_viewModels_PSLockThreshold	int	-1	no	—
r_shaderLimits_renderOpaque_viewModels_VSLockThreshold	int	-1	no	—
r_shaderLimits_renderOpaque_world_numLateAlloc	int	10	no	—
r_shaderLimits_renderOpaque_world_numPS	int	-1	no	—
r_shaderLimits_renderOpaque_world_numVS	int	-1	no	—
r_shaderLimits_renderOpaque_world_PSLockThreshold	int	-1	no	—
r_shaderLimits_renderOpaque_world_VSLockThreshold	int	-1	no	—
r_shadingRate	bool	1	yes	Use NVIDIA Variable Shading Rate (VSR)
r_shadingRateAntiFlickerMotionNorm	float	5.0	yes	Normalization in pixels
r_shadingRateAntiFlickerRoughMaterial	float	0.8	yes	1/x factor to increase luma sensitivity under slow motion and avoid shimmering
r_shadingRateAntiFlickerSmoothMaterial	float	0.33	yes	1/x factor to increase luma sensitivity under slow motion and avoid shimmering
r_shadingRateDebug	bool	0	yes	Visualize shading rate in colors: blue - 1x2/2x1; green - 2x2; yellow - 2x4/4x2; red - 4x4; white dot = transposed
r_shadingRateFinalDither	float	0.0	yes	Dithering of shading rates
r_shadingRateLevelMax	int	4	yes	Maximum allowed shading rate (0 - 4x4; 1 - 4x2; 2 - 2x2; 3 - 2x1; 4 - 1x1)
r_shadingRateLevelMin	int	1	yes	Minimum allowed shading rate (0 - 4x4; 1 - 4x2; 2 - 2x2; 3 - 2x1; 4 - 1x1)
r_shadingRateMotionInfluence	float	40.0	yes	Motion influence to shading rate in %
r_shadingRateMotionSensitivityToLuma	float	5.0	yes	Reduce motion influence in areas with divergent colors
r_shadingRateNormalInfluence	float	1.0	yes	Sensitivity to divergent normals
r_shadingRatePrevAmount	float	0.001	yes	Amount of previous shading rate
r_shadingRateQualityDistribution	float	10.0	yes	Quality distribution in % (-100 = super coarse, 0 = balanced, 100 = super fine)
r_shadingRateQualityOverride	int	5	yes	0 - 4x4; 1 - 4x2; 2 - 2x2; 3 - 2x1; 4 - 1x1; 5 - adaptive
r_shadingRateSensitivityBrightness	float	60.0	yes	Sensitivity to the screen brightness in %
r_shadingRateSensitivityLuminance	float	63.0	yes	Sensitivity to per-pixel luminance changes in %
r_shadingRateSmooth	bool	1	yes	Smoothing of shading rates
r_shadowAssumeNumTriDynamic	int	256	no	kilo-triangles to assume when a shadow has no known triangle count.
r_shadowAssumeNumTriStatic	int	256	no	kilo-triangles to assume when a shadow has no known triangle count.
r_shadowAtlasHeight	int	8192	yes	Shadow map atlas height
r_shadowAtlasTileSize	int	2048	yes	Shadow map atlas tile size
r_shadowAtlasWidth	int	8192	yes	Shadow map atlas width
r_shadowCacheFmt	int	0	no	Shadow map atlas fmt. 0: 32 bit, 1: 16 bit
r_shadowDebugSlice	int	0	no	fits depth ranges of shadowmaps to the included casters to increase stability
r_shadowDirtyWarningTreshold	int	1000000	no	Warn if a model intersects more than this number of lights.
r_shadowDMAStaticSlices	bool	1	no	—
r_shadowDurangoDebug	int	15	no	Debug!
r_shadowDurangoESRAM	bool	1	no	Render to ESRAM if true, otherwise directly to atlas
r_shadowDurangoESRAMComposite	bool	0	no	Render composite in ESRAM if true, otherwise directly to atlas
r_shadowDurangoESRAMMaxMip	int	3	no	Render directly to atlas if mip > ESRAMMaxMip
r_shadowFadeRangeAdd	float	0.0	no	Global add for the user defined per-light shadow fade range
r_shadowFadeRangeScale	float	1.0	no	Global scale for the user defined per-light shadow fade range
r_shadowFitDepthToCasterBounds	bool	0	no	fits depth ranges of shadowmaps to the included casters to increase stability
r_shadowForceDynLights	bool	1	no	always update for dynamic lights
r_shadowHardLimitPixels	int	262144	yes	Shadow per-frame pixels hard cap in kilo-pixels.
r_shadowHardLimitTriangles	int	65536	yes	Shadow per-frame triangles soft cap in kilo-triangles.
r_shadowJitter	bool	0	no	Enables subpixel jitter for shadow projection matrix for cutscenes
r_shadowJitterDebugScale	float	1.0	no	Scales subpixel offset of shadows for debug purposes
r_shadowKeepStaticMipsMask	int	24	no	Bit mask that represent mip levels of static cached shadow maps that won't get deleted when the current mip is lowered.
r_shadowLODAdjustThreshold	float	0.2	no	Cached shadows are discarded if the wanted float shadow level differs this many LOD levels from the cached shadow.
r_shadowLodBias	int	0	no	—
r_shadowLodForceValue	int	-1	no	Force a specific lod level for shadow rendering of dynamic models. -1: Use the same LoD as opaque.
r_shadowLODProjectionSizeFactor	float	0.08	no	Projection size lod scale
r_shadowLogBase	float	3.0	no	The size of shadow cascades calculated by pow(r_shadowLogBase, cascadeID)
r_shadowMaxStaleFrames	string	0,0,1,2,2	yes	Max stale frames for spots/point lights per mip level. (0=60hz, 1=30hz, 2=20hz, etc..)
r_shadowMinStaticMip	int	3	no	Mip at which the static caching is applied, idea is that the more shadows -> lower mips -> more vertex pressure -> static caching
r_shadowOcclusionMinRadius	float	10	no	Lights with radius greater than this value will use occlusion culling for shadow buffer rendering. (set before loading map)
r_shadowParallelCompositeAllEntitiesMaxSlice	int	-1	no	All non-worldspawn shadows are rendered as dynamics into the composite for slices <= this value. (set to -1 to disable)
r_shadowParallelLightFastPath	bool	1	no	—
r_shadowParallelMaxStaleFrames	string	0,0,1,2	no	Max stale frames for parallel lights per cascade. (0=60hz, 1=30hz, 2=20hz, etc..)
r_shadowParallelMoveMargin	float	0.5	no	Parallel light: static cache move margin. (in meters)
r_shadowParallelSkipDynamicModelsFromSlice	int	-1	no	—
r_shadowParallelSliceOffsetScale	float	0	no	offsets the first parallel cascade in direction of the player
r_shadowPlayerCullRadius	float	32	no	Approximate radius for the player first person models used for culling shadows.
r_shadowPlayerCullSlice	bool	1	no	Cull player shadows per slice.
r_shadowPlayerLOD	int	1	no	Shadow LOD level for player/weapon shadows.
r_shadowPlayerMaxStaleFrames	int	0	no	Max throttling for weapon shadows.
r_shadowRenderDepthCopy	bool	1	no	Copy depth using rasterizer.
r_shadowsDistanceFadeMultiplier	float	1.0	yes	Shadows fade distance multiplier.
r_shadowSkipSliceViewCull	bool	0	no	Skip culling point light slices against main view.
r_shadowSoftLimitPixels	int	262144	yes	Shadow per-frame pixels soft cap in kilo-pixels.
r_shadowSoftLimitTriangles	int	6144	yes	Shadow per-frame triangles soft cap in kilo-triangles.
r_shadowsSunCascadesRadius	float	1000.0	no	Sun shadow cascades radius
r_shadowStaticMipTriThresholds	string	-1,10,8,4,2	no	kTri thresholds for using static caching for each mip level. (-1=off)
r_shadowStaticMode	int	1	no	Select heuristic for static caching. 0 = Use r_shadowMinStaticMip, 1 = Use r_shadowStaticMipTriThresholds
r_shadowStats	int	0	no	0=Off, 1=Trace stats, 2=Record stats, 3=Trace recorded data and set to 0, 4=only print if stats change, 5=only print if shadow alloc failed
r_shadowTopOffenders	int	0	no	List the top N most expensive shadow casting lights. (disables soft caps)
r_shadowUpdateDynamicOnly	bool	1	no	just shadowmaps that include moving objects or from moving lightsources are updated
r_shadowUseClearAttachment	bool	1	no	Use GL_ClearDepthAttachment to clear shadow slices
r_shadowUseStaticGeometry	bool	1	no	Use preprocessed static shadow geometry for shadow maps. (set before loading map)
r_shadowUseTriangleAndPixelLimits	bool	1	no	enforce r_shadowSoftLimit* and r_shadowHardLimit
r_shadowVSMSmoothness	float	0.5	no	biases the smoothness of VSM shadows
r_sharpening	float	2.0	yes	Sharpening filter, applied after all rendering
r_showAmbientProbes	int	0	no	Draw boxes for all the ambient samples, and lines from all dynamic models to their ambient. Note: due to how debug views are rendered at this point, these boxes are not affected by tone mapping - for more accurate ambient color view, use r_debugToolsHDR 1
r_showAnimatedBounds	int	0	no	displays the bounding boxes of all animated models (1 = frame, 2 = normalized, 4 = translated)
r_showBlendedSurfOverdraw	int	0	no	if > 0 show overdraw from all blended surfaces
r_showBlendedSurfOverdrawMax	int	10	no	overdraw >= this value is white
r_showBuffers	int	0	no	Show buffer info logging, bitfields: 0 - none, 1 - all, 2 - vertex, 4 - index, 8 - uniform, 16 - structured, 32 - pixel pack/unpack
r_showCommits	int	0	no	report model and light commits and ref counts
r_showCracks	int	0	no	display cracks less that the given number of pixels wide
r_showCurvature	bool	0	no	Show the deviation between interpolated normals and facet normals
r_showDebugTools	bool	1	no	Show debug lines/text/etc
r_showDecalsByExpense	int	0	no	prints a list of currently processed decals, sorted by their impact on rendering time
r_showDecalsByExpenseCount	int	8	no	shows the slowest n decals r_showDecalsByExpense
r_showDecalsByExpenseSensitivity	int	16	no	higher values reduce the variance of r_showDecalsByExpense
r_showDecalsByExpenseTextScale	float	2.5	no	scales the text showing the timings of r_showDecalsByExpense
r_showDeltaStats	int	0	no	report render stat deltas
r_showDepth	bool	0	no	display the contents of the depth buffer and the depth range
r_showDestinationAlpha	bool	0	no	draw destination alpha
r_showDisaster	int	0	no	1 shadow/motion blur updating objects 2 alpha-test 4 two sided
r_showDrawColor	string	1 1 1 0.4	no	specifies the draw color for r_showTris mode 4
r_showDuplicateTriangles	int	0	no	show duplicate triangles, 2 = reload map genmodel
r_showDynamic	int	0	no	report stats on dynamic surface generation
r_showDynamicLighting	int	0	no	Draw lines to lights for each dynamic model, 3 = draw bounds of problem models and lights affecting them
r_showDynamicLightingSingleModel	int	-1	no	suppress show dynamic lighting on all but one model specified with this model index
r_showEdges	bool	0	no	draw the sil edges
r_showEnvironmentProbes	int	0	no	Draw boxes for all the environment probes. Note: due to how debug views are rendered at this point, these boxes are not affected by tone mapping - for more accurate probes color view, use r_debugToolsHDR 1
r_showEnvs	bool	0	no	displays the bounding boxes of all the dynamic environments
r_showExpandedWorld	int	0	no	—
r_showFoliageNodeBounds	int	0	no	show the individual foliage bounds - 0 = off, 1 = just the bounds, 2 = bounds + position and size
r_showFoliageSubTreeBounds	int	-1	no	show the bounds of the specified subtree
r_showFoliageSubTreeNodes	int	-1	no	show the nodes of the specified subtree
r_showGuiOverdraw	bool	0	no	—
r_showGuiTris	bool	0	no	—
r_showHotPatchModelMaterials	bool	0	no	set to 1 to show details on materials in the patch table
r_showImageCreation	bool	0	no	Print image name, size and api object on creation.
r_showIntensity	bool	0	no	draw the screen colors based on intensity, red = 0, green = 128, blue = 255
r_showKdTreeBounds	int	0	no	0 = off, 1 = nodes, 2 = areas, 3 = both
r_showLightCenter	bool	0	no	Draw light axis and center.
r_showLightCount	int	0	no	1 = color dkgreen to green for lights below 5 surfaces and color yellow to red for lights with more than 5 surfaces, white is more than 12, 2 = colors surfaces based on light count touching the surface
r_showLightOcclusion	int	0	no	Light index, -1 == Off
r_showLightOcclusionSlice	int	0	no	Light slice index for point lights. [0..5]
r_showLightOverdraw	int	0	no	1 = print light overdraw
r_showLightRelinkDirty	int	0	no	Show light shadow dirty mask due to light relinking.
r_showLights	int	0	no	1 = just print volumes numbers, highlighting ones covering the view, 2 = also draw planes of each volume, 3 = spews a one-time list of the names of active lights
r_showLightsByExpense	int	0	no	prints a list of currently processed lights, sorted by their impact on rendering time
r_showLightsByExpenseSensitivity	int	16	no	higher values reduce the variance of r_showLightsByExpense
r_showLightsMode	int	0	no	0 = Show all, 1 = Show non-static, 2 = Show non-static shadow casters, 3 = Show only in static shadow index range. (r_showLightsSSMin/Max)
r_showLightSpecific	string	(empty)	no	show the volume of a specifically named light
r_showLightsSSMax	int	65536	no	Max static shadow slice for r_showLights & r_showStaticShadows.
r_showLightsSSMin	int	0	no	Min static shadow slice index for r_showLights & r_showStaticShadows.
r_showLods	int	0	no	displays the color bounding boxes for each lod
r_showMagnifier	float	0	no	—
r_showManifolds	int	0	no	0 = off, 1 = all manifolds, 2 = significant manifolds, 3 = omni-significant manifolds, 4 = show detail manifolds, 5 = show pipe manifolds, 6 = subdivided world manifolds
r_showModelGroupMasters	int	0	no	draw lines from from models that use another model's origin for lighting and shadowing (1 = visible models, 2 = all models)
r_showModels	int	0	no	1 = displays the bounding boxes of all view models, 2 = print index numbers
r_showModelsSingle	int	-1	no	if >= 0 then only that specific model will be debugged with r_showModels
r_showNavPortals	int	1	no	—
r_showNoDrawSurfaceInfo	bool	0	no	allow th r_showSurfaceInfo to intersect materials that don't draw ( "isdrawn 0" )
r_showNonTwoManifoldEdges	bool	0	no	draw edges that are not shared by exactly two triangles
r_showNormals	float	0	no	draws wireframe normals
r_showOcclusionBaseModel	int	0	no	show sound BSP base occlusion model
r_showOcclusionBoxes	bool	0	no	—
r_showParms	bool	0	no	print all world / view parmBlock assignments
r_showPIDInTitle	bool	0	yes	1 = show process ID in window title. 0 = don't.
r_showPrimitives	int	0	no	report drawsurf/index/vertex counts
r_showPushFrustumIntoTreeProfiling	bool	0	no	—
r_showPVS	int	0	no	0 - disabled, 1 - draw cells visible from player as green, 2 - draw cells invisible from player as red, 3 - draw both red and green cells
r_showQueries	bool	0	no	reports the number of occlusion queries per frame
r_showReferenceUpdates	bool	0	no	show rendermodels which have their references updated
r_showRefPose	bool	0	no	selects the pose to draw with r_showskel, 0 = draw the final pose, 1 = draw the reference pose
r_showRenderLog	int	0	no	report stats
r_showRuler	bool	0	no	—
r_showShadowActivity	int	0	no	Shows shadow slice dirty tag (red lines from dynamic models, the orange boxes) and color-coded rendering cost. Slice triangles: blue<=16k, green<=64k, yellow<=256k, orange<=512k, red<=1M, purple>1M. Parallel lights are not included since they are always dynamic. Options: 1 = Enable, 2 = Show activity for weapon shadow set, 4 = Disable depth-test for slice rendering.
r_showShadowAtlas	bool	0	no	display the contents of the shadow atlas buffer
r_showShadowCache	int	0	no	Show shadow cache status.
r_showShadows	int	0	no	Show shadow geometry. (1 = Static, 2 = Static bounds, 4 = Shadow view surfaces (#define ID_SHADOWVIEWDEBUG needs to be enabled) )
r_showShadowSliceViewCull	bool	0	no	Log culling point light slices against main view.
r_showShadowsName	string	(empty)	no	Limit shadow display to a specific light name
r_showShadowsSlice	int	-1	no	Limit shadow display to a specific slice. -1 = Show all.
r_showSkel	int	0	no	draw the skeleton when model animates, 1 = draw model with skeleton, 2 = draw skeleton only
r_showSpecularTriangles	bool	0	no	show specular triangles
r_showStats	int	0	no	report render stats
r_showStatsNewGPUParticles	bool	0	no	—
r_showStencil	bool	0	no	display the contents of the stencil buffer
r_showSurfaceBounds	int	0	no	1 = displays the bounding boxes of all model surfaces
r_showSurfaceInfo	int	0	no	show surface material name under crosshair
r_showSurfaceInfoDistance	float	65536	no	distance r_showSurfaceInfo traces
r_showTangentSpace	int	0	no	shade triangles by tangent space, 1 = use 1st tangent vector, 2 = use 2nd tangent vector, 3 = use normal vector
r_showTexelDensity	int	0	no	shade triangles by texel density. Set > 0 to scale range by specified amount. Colors mean: 0: black, 1: purple, 2: blue, 3: green, 4: yellow, 5: orange, 6: red, 7+: white.
r_showTexelDensitySmooth	bool	1	no	Turn on for gradient colored rendering of the texel densities
r_showTextureDistortion	int	0	no	shade triangles based on UV stretching
r_showTextureIslands	int	0	no	0 = off, 1 = all manifolds
r_showTexturePolarity	bool	0	no	shade triangles by texture area polarity
r_showTextureSpace	int	0	no	shade triangles by texture space
r_showTextureValues	bool	0	no	Report the texture channel numbers under the crosshair
r_showTJunctions	int	0	no	0 = off, 1 = all t-junctions, 2 = t-junctions on significant occluders, 3 = j-junctions with vertices between two colinear edges, 4 = = j-junctions with vertices between two colinear edges on significant occluders
r_showTotalSurfaceArea	int	0	no	print the total surface area of all static surfaces (2=include non-statics too)
r_showTrace	int	0	no	show the intersection of an eye trace with the world
r_showTraceWorld	int	0	no	show traceworld info under crosshair
r_showTransparencyStats	int	0	no	1 = print stats
r_showTransSortOverdraw	float	0	no	if > 0 show overdraw from transparency sorted surfaces
r_showTriangleInfo	int	0	no	show vertex numbers and tangents of the triangle the crosshair is over
r_showTris	int	0	no	enables wireframe rendering of the world, 1 = only draw visible ones, 2 = draw all front facing, 3 = draw all, 4 = only visible ones with blended lines, 5 = draw all front facing non-static
r_showTwoSided	int	0	no	outline two sided geometry
r_showUmbraStats	int	0	no	0 = off, 1 = visible world areas, 2 = culling details, 3 = timing
r_showVertexColor	bool	0	no	draws all triangles with the solid vertex color
r_showViewVolumes	int	0	no	Show Umbra view volumes
r_showVisibleInfluenceSpheres	bool	0	no	show the visible influence spheres
r_showVoxels	int	0	no	—
r_showWorldAreaBounds	int	0	no	0 = off, 1 = subspace, 2 = geometry, 3 = both
r_showWorldAreas	int	0	no	0 = 0ff, 1 = show world areas, 2 = lock view and show world areas
r_singleFlareModel	int	-1	no	suppress all but one flare
r_singleGuiSurface	int	-1	no	suppress all but one surface on gui model
r_singleLight	int	-1	no	suppress all but one light
r_singleModel	int	-1	no	suppress all but one model
r_singleSurface	int	-1	no	suppress all but one surface on each model
r_singleTriangle	bool	0	no	only draw a single triangle per primitive
r_singleWorldArea	int	-1	no	suppress all but one world area
r_skinningUseRelativeOffsets	bool	1	no	Use offsets to surface 0 joints offsets.
r_skipAddAlways	bool	0	no	skip rendering add-always models
r_skipAfterPostProcessSurfaces	bool	0	no	skip surfaces that need to be rendered after the post process
r_skipAllModels	bool	0	no	skip rendering all dynamic and static instance models
r_skipAnalyze	int	0	no	skip the feedback analysis
r_skipAreaCPUCulling	bool	0	no	skip CPU MVP culling of world areas
r_skipAreaGPUCulling	bool	0	no	skip GPU occlusion testing of world areas
r_skipAugment	bool	0	no	skip rendering augment models
r_skipBeams	bool	0	no	skip render model beam system
r_skipBlendedSurfaces	bool	0	no	skip blended surfaces
r_skipCommits	bool	0	no	1 = don't accept any entity or light updates, making everything static
r_skipDecalCulling	bool	0	no	skip decal culling
r_skipDecals	bool	0	no	skip decal rendering
r_skipDepthFixupSurfaces	bool	0	no	skip depth fixup surfaces
r_skipDetailTris	bool	0	no	skip all detail triangles
r_skipDistortionSurfaces	bool	0	no	skip distortion surfaces
r_skipFeedbackCapture	bool	0	no	skip capturing feedback
r_skipFlares	bool	0	no	skip rendering flare models
r_skipFog	bool	0	no	skip rendering fog
r_skipFoliage	bool	0	no	Completely disable foliage models
r_skipGamePostProcess	float	0	no	Skip gameplay post-process effects (near death, drowning, burning etc)
r_skipGeoBlendSurfaces	bool	0	no	blended geo surfaces
r_skipGlass	bool	0	no	skip glass
r_skipGPUParticleSort	bool	0	no	—
r_skipGPUTriangleCulling	bool	0	no	skip GPU triangle culling
r_skipGuis	bool	0	no	skip rendering all guis
r_skipHotPatchDuringModelLoad	bool	0	no	set to 1 to skip hot patching of vmtrs
r_skipInGameGuis	bool	0	no	skip rendering in-game guis
r_skipLightCPUCulling	bool	0	no	skip CPU MVP culling of lights
r_skipLightGPUCulling	bool	0	no	skip GPU occlusion testing of lights. Note: This is forced enabled in slowmap mode.
r_skipLightRangeCulling	bool	0	no	allow lights to be removed at distance
r_skipLights	int	0	no	1 - skip all dynamic lights, 2 - points, 4 - spots, 8 - area, 16 - parallel, 32 - probes (combine by adding, e.g. skip spots + points = 2+4 = r_skipLights 6)
r_skipMergeMeshes	bool	0	no	skip combining rendermodel surface to reduce the batch count
r_skipModelCPUCulling	bool	0	no	skip CPU MVP culling of models
r_skipModelGPUCulling	bool	0	no	skip GPU occlusion testing of models
r_skipModelRangeCulling	bool	0	no	allow models to be removed at distance
r_skipModels	bool	0	no	skip rendering all dynamic models
r_skipMultipleMaterialErrors	bool	1	no	set to 0 to spam hotpatch warnings per vertex, 1 to only emit a warning for the first vertex in a surface.
r_skipNewGPUParticles	bool	0	no	—
r_skipNodeCPUCulling	bool	0	no	skip CPU MVP culling of world nodes
r_skipNodeGPUCulling	bool	0	no	skip GPU occlusion testing of world nodes
r_skipOcclusionBaseModel	bool	1	no	use sound BSP derived model to cover up missed occlusion queries
r_skipOcclusionSurfaces	bool	0	no	skip occlusion surfaces
r_skipOpaqueSurfaces	bool	0	no	skip opaque surfaces
r_skipPlayerShadow	bool	1	no	skip casting shadows from player hands+weapons on self
r_skipPostProcess	bool	0	no	skip post processing
r_skipRenderTools	bool	0	no	skip rendering all render tools
r_skipShadowDirtyTag	int	0	no	Skip shadow dirty tagging in render world (1=static, 2=dynamic)
r_skipShadowMipFlickerLimit	bool	0	no	Disable mip-map flicker limiter
r_skipShadowModelCPUCulling	int	0	no	skip CPU culling of shadow casting models. (1=Light bound+Light MVP, 2=Main view MVP, 4=Shadow MVP)
r_skipShadowModelOcclusionCulling	bool	0	no	skip occlusion culling of shadow casting models
r_skipShadowModelSort	bool	0	no	render unsorted models to the shadow buffer
r_skipShadowParallelTickOffset	bool	0	no	Disables update frame offset mechanism for parallel lights.
r_skipShadows	bool	0	no	just use the existing depth texture contents
r_skipShadowSurfaceCPUCulling	bool	0	no	skip CPU MVP culling of shadow casting surfaces
r_skipShadowSurfaceOcclusionCulling	bool	1	no	skip occlusion culling of shadow casting surfaces
r_skipShadowThrottle	bool	0	no	Disable low-frequency updates of shadows when soft limits are exceeded.
r_skipSuppress	bool	0	no	ignore the per-view suppressions
r_skipSurfaceCPUCulling	bool	0	no	skip CPU MVP culling of model surfaces
r_skipSwapAfterRenderModeChange	int	6	no	Skip swap this many frames when switching back to RENDER_MODE_MANUAL
r_skipSwapChainRecreateWhenMinimized	bool	1	yes	Skips recreation of the swapchain and buffer swaps if the window is minimized.
r_skipTracers	bool	0	no	skip tracers
r_skipTransparentDepthSort	bool	0	no	skips setup for depth based sort of drawcalls
r_skipUpdateInView	bool	0	no	skip world update in view call
r_skipViewParms	bool	0	no	don't use the parmBlock in idRenderView
r_skipVmtrs	bool	0	no	1 = skip loading of vmtrs
r_skipWater	int	0	no	1 = skip render update of all water surfaces
r_skipWorld	bool	0	no	skip v all world geometry
r_skipWorldViews	bool	0	no	skip rendering all world views
r_skyLighting	bool	1	no	Enable sky lighting
r_skyLightingLut	int	1	no	1: Enable skylighting LUT precomputing, amortize results over N frames, 2: Enable skylighting LUT precomputing, compute entire LUT. 0: disable
r_skyLightingLutResolutionAngleLatitude	int	128	no	Pre-computed LUT for atmosphere light scattering, latitude angle component
r_skyLightingLutResolutionAngleLongitude	int	32	no	Pre-computed LUT for atmosphere light scattering, longitude angle component
r_skyLightingLutResolutionDistance	int	32	no	Pre-computed LUT for atmosphere light scattering, distance component
r_sleep	int	0	no	Sleep for this many milliseconds to simulate low frame rates.
r_slopNormal	string	0.0001	no	merge normals that dot less than this
r_slopTexCoord	string	0.00001	no	merge texture coordinates this far apart
r_slopVertex	string	0.0002	no	merge xyz coordinates this far apart
r_SSDO	int	1	yes	0 - off, 1 - enables screen space directional occlusion
r_SSDODiffuseOcclusion	float	1.0	no	SSDO diffuse occlusion scale
r_SSDOEdgeSharping	float	2.0	no	SSDO sharpness of corners/edges, 1.0 == linear, 2.0 == squared etc.
r_SSDONormalBias	float	0.0	no	SSDO Normal Bias. To mitigate depth precision artifacts
r_SSDOQuality	int	1	yes	0 - low, 1 - medium, 2 - high
r_SSDORadius	float	0.7	no	SSDO radius
r_SSDOSpecularOcclusion	float	1.0	no	SSDO specular occlusion scale
r_SSDOTemporalAA	bool	1	no	Enable/disable SSDO temporal AA
r_SSDOTemporalAABlendRatio	float	0.15	no	SSDO temporal AA blend ratio
r_SSR	int	1	yes	1: Enable screen space reflections pass
r_SSRMinSmoothness	float	0.3	no	Set minimum smoothness for SSR to blend
r_SSRQuality	int	2	yes	0 - low, 1 - medium, 2 - high
r_staticBloomMaskScale	float	1	no	Set multiplier for static object emissive mask
r_staticModelErrorUnwrapError	int	0	no	fail generation of static instance static models if vertices doesn't match after unwrap
r_staticModelTriStrips	bool	0	no	build tristrips for static models
r_staticShadowsBuildForSingleLightName	string	(empty)	no	only include this light in the static shadows ('sun' for the sun light) (for debugging)
r_staticShadowsCullTrisOutsideRadius	bool	1	no	Cull triangle that are outside the light radius
r_staticShadowsCullWithTraceWorld	bool	1	no	Cull triangles with traceworld
r_staticShadowsCullWithTraceWorldSkipAlphatestSurfaces	bool	0	no	skip traces that hits transparent surfaces
r_staticShadowsCullWithTraceWorldSubDivStep	float	0.05	no	step size for triangle subdiv for culling with traceworld
r_staticShadowsCullWithUmbra	bool	1	no	Cull triangles with umbra occlusion
r_staticShadowsIncludeInstances	bool	1	no	Set to include instance geometry into static shadow cache
r_staticShadowsIncludeSun	bool	1	no	Set to include the sun into static shadow cache
r_staticShadowsMaxAlphaVertices	int	1048576	no	Maximum amount of transparent vertices in shadow cache
r_staticShadowsMaxOpaqueVertices	int	33554432	no	Maximum amount of opaque vertices in shadow cache
r_staticShadowsUseJobs	bool	1	no	cull triangles in parallel
r_staticTransparencyFadeMaxDist	float	2048	no	completely faded out at this range
r_staticTransparencyFadeStartDist	float	1024	no	start fading out at this range
r_stripFragmentBranches	bool	1	no	strip branches from fragment programs
r_swapInterval	int	-1	yes	Changes vsync settings. 0 = off, 1 = vsync 60Hz, -1 = swap-tear 60Hz
r_TAAClearAccumulationBuffer	bool	0	no	Clears the accumulation buffer
r_TAALowPassResolveFilter	int	2	no	Resolve filter type. 0: Box, 1: Triangle, 2: Gaussian, 3: Mitchel, 4: B-spline, 5: Catmul-rom, 6: Smoothstep, 7: Hanning, 8: Hamming, 9: BlackMan, 10: Blackman-Harris, 11: Lanczos
r_TAALowPassSharpening	float	1.0	no	Kernel filter width. Increasing value, increases sharpening. Decreasing increases smoothness
r_taaNegativeLodBiasVT	int	1	no	Enable negative lod bias for virtual textures, when TAA modes enabled
r_TAAResolveFilter	int	3	no	Resolve filter type. 0: Box, 1: Triangle, 2: Gaussian, 3: Mitchel, 4: B-spline, 5: Catmul-rom, 6: Smoothstep, 7: Hanning, 8: Hamming, 9: BlackMan, 10: Blackman-Harris, 11: Lanczos
r_TAASafeMode	int	0	no	0 = Disabled, 1 = enable safe mode, disable anything relying on temporal accumulation
r_TAASharpening	float	1.0	no	Kernel filter width. Increasing value, increases sharpening. Decreasing increases smoothness
r_TAATransparentsK	float	1.0	no	Transparents weight scale
r_testAutosprites	bool	0	no	display the autosprited geometry
r_testGamma	float	0	no	if > 0 draw a grid pattern to test gamma levels
r_testGammaBias	float	0	no	if > 0 draw a grid pattern to test gamma levels
r_testImageIntensity	float	1.0	no	The a multiplier to increase/decrease the brightness of the test image
r_testImageLayer	int	0	no	Which slice (layer) to display in a texture array
r_testImageMip	float	0.0	no	TestImage Mip level
r_testImageOffsetX	float	0	no	TestImage Offset in the X direction, system values range (0..1)
r_testImageOffsetY	float	0	no	TestImage Offset in the Y direction, system values range (0..1)
r_testImageScale	float	0.75	no	TestImage Scale
r_testMegaTexture	bool	0	no	draw the entire megatexture as a flat quad
r_testTextureReconstruct	bool	0	no	—
r_testVMTR	float	0	no	if > 0 displays the vmtr specified by r_currentMaterial centered on the screen
r_testVMTRLod	int	0	no	which LOD of the vmtr to display
r_threadedCommit	bool	1	no	—
r_threadedRenderGui	bool	0	no	Use a parallel job to render the GUI
r_threadedUpdateInView	int	2	no	0 = single threaded, 1 = job chain, 2 = coop job
r_timerWatchdogCrashOnTimout	bool	0	no	crash on timeout (to catch driver crashed for autobuilder
r_timerWatchdogTimout	float	0	no	render watchdog timer in seconds
r_transparencyHalfRes	int	0	no	Enables half-resolution transparency. 0: Disabled. 1: Automatic. 2: Always.
r_transparencyHalfRes_ConstantCostEstimation	float	0.35	no	Estimation of the constant cost of using half-res particles (ms)
r_transparencyHalfRes_DebugRender	bool	0	no	Transparency half-resolution debug rendering
r_transparencyHalfRes_DropDuration	float	3.0	no	Switch to half-resolution rendering when the blend pass takes longer that this value (ms)
r_transparencyHalfRes_MaxDistance	float	3000	no	Max distance at which half-resolution transparency is supported
r_transparencyHalfRes_MaxSettleTime	int	2000	no	Never keep rendering in half-resolution for longer than this unless we absolutely have to (ms)
r_transparencyHalfRes_RaiseDuration	float	2.5	no	Come back to full-resolution rendering if we estimate that the blend pass will take less than this value (ms)
r_transparencyHalfRes_SettleTime	int	500	no	Time to wait before considering to go back to full res (ms)
r_transparencyHalfRes_SpeedupEstimation	float	2.5	no	Estimated speedup factor for rendering particles in half-resolution
r_transparencyProfileModels	bool	0	no	Insert a profiler block for each individual transparent surface
r_transparencyShowSurfaces	int	0	no	1: Show all transparent surfaces being rendered in the world. 2: Show prog name. 3: Show material name. 4: Show model name. 5: Show stage sort.
r_transparentSurfacesSortMode	int	0	no	0 - model center depth, 1 - depth bounds max depth
r_umbraAccurateOcclusionThreshold	float	-1	no	LOD bias. A higher value results in more accurate culling over distance but also in higher CPU cost
r_umbraAccurateOcclusionThresholdStep	float	0.9	no	The maximum LOD bias. A higher value results in more accurate culling over distance but also in higher CPU cost
r_umbraAdaptiveOcclusionThreshold	bool	0	no	Set to use the adaptive occlusion threshold to balance framerate with accuracy.
r_umbraAdaptiveOcclusionThresholdMsec	float	12.0	no	The maximum number of millseconds to let job(s) run before scaling down culling distance.
r_umbraBackfaceLimit	float	100.0f	no	Backface limit percentage
r_umbraBuildTomeMapType	int	0	no	Build tome for specified map data (.proc file). 0 = COMBO, 1 = SLOW
r_umbraCameraUncertainty	float	0.0	no	The amount of uncertainty in the camera translation. Helpful when camera is behind geo.
r_umbraDebug_showCurrentViewCell	bool	0	no	Show debug lines for current view cell
r_umbraDebug_showPortalsTraversed	bool	0	no	Show debug lines for portals traversed in query
r_umbraDebug_showStats	bool	0	no	Show statistics for queries
r_umbraDebug_showViewFrustum	bool	0	no	Show debug lines for view frustum
r_umbraDebug_showVisibilityLines	bool	0	no	Show debug lines from view location to visible objects
r_umbraDebug_showVisibleObjectBounds	bool	0	no	Show debug lines for bounds of visible objects
r_umbraDebug_showVisibleVolume	bool	0	no	Show debug lines for volume PVS
r_umbraIgnoreCameraPosition	bool	0	no	Sets QUERYFLAG_IGNORE_CAMERA_POSITION on Umbra culling jobs.
r_umbraJobKickoff	int	2	no	Submit the RenderCull request early? 0=no, 1=early, 2=super early, 3=deferred
r_umbraJobsPerWorker	int	4	no	Multiplies the number of worker threads by this value to obtain the number of jobs to run (limited by r_umbraMaxJobCount.) The higher this value, the more granular the jobs and the better they'll dynamically balance, but the more overhead per job due to job scheduling.
r_umbraMaxAccurateOcclusionThreshold	float	100	no	The maximum LOD bias. A higher value results in more accurate culling over distance but also in higher CPU cost
r_umbraMaxJobCount	int	16	no	The maximum number of jobs to split queries into. This is limited by Umbra to 16, so setting it higher will cause queries to fail.
r_umbraMaxShadowQueriesInvisible	int	1	no	Max non-visible lights shadow occlusion jobs per frame
r_umbraMaxShadowQueriesVisible	int	6	no	Max visible lights shadow occlusion jobs per frame
r_umbraMinAccurateOcclusionThreshold	float	11	no	The maximum LOD bias. A higher value results in more accurate culling over distance but also in higher CPU cost
r_umbraOverrideSmallestOccluder	float	0.0f	no	Used to override the normal smallest occluder cvar
r_umbraPrimeHTile	int	0	no	0 - Normal depth clear, 1 - Prime depth HTile with Umbra occlusion buffer, 2 - Prime Depth HTile with Umbra Compute Version
r_umbraShowOcclusionBuffer	int	0	no	1 = histogram, 2 = depth, 3 = leaks to infinity, 4 = potential leaks (mostly false positives)
r_umbraShowShadowJobs	bool	0	no	Show shadow occlusion jobs
r_umbraSkipParallelShadowCuller	bool	0	no	Enable/disable Umbra shadow culler for parallel lights
r_umbraSmallestHole	float	0.1f	no	Smallest hole computation parameter
r_umbraSmallestHoleSnap	float	0.25f	no	Smallest hole computation parameter for Snap
r_umbraSmallestOccluder	float	2.0f	no	Smallest occluder computation parameter
r_umbraTempPath	string	(empty)	no	Directory for storing Umbra temp and cache files, if empty the files will be placed inside generated/umbra
r_umbraUseLocalLightQuery	bool	0	no	Use local light queries but disables LOD
r_umbraYieldToCollisionJobs	bool	1	no	HACK: Ensure the collision jobs have completed before starting umbra jobs so the main thread doesn't stall for the length of umbra when there are only a couple of collision jobs left.
r_useAnimBlendJobs	int	1	no	0 = off, 1 = threaded
r_useCompressedLightProbes	bool	1	no	0 == use the uncompressed F16 lightprobes, 1 == use the compressed BC6H lightprobes
r_useCubeMapSeamless	bool	1	no	Enable GL_TEXTURE_CUBE_MAP_SEAMLESS
r_useDistanceFont	int	1	no	1 - the game uses distance fonts instead of coverage fonts - Note: This cvar should only be set from command line and toggled at run time
r_useDynamicEnvironment	int	1	no	generate a new environment map every frame
r_useDynamicLightingJobs	int	1	no	0 = off, 1 = threaded
r_useFoliageGenJobs	int	1	no	0 = off, 1 = threaded
r_useHardwareTextures	bool	0	no	Use hardware textures instead of virtual textures for implicit gui images
r_useInfluenceSphereCullJob	int	1	no	0 = off, 1 = threaded
r_useInfluenceSpheres	bool	1	no	enable influence spheres
r_useIterativeKDTree	bool	1	no	Use iterative rather than recursive version of kdtree functions
r_useKdPartitioning	bool	1	no	Use k-d tree partitioning for world areas and BSP -- set BEFORE loading map!
r_useLightPVS	bool	0	no	Use PVS to reduce the number of areas a shadow casting light affects
r_useModelUpdateFlags	bool	1	no	Call UpdateInView/UpdateInViewJoin only for models that implement it
r_useParticleGenJobs	int	1	no	0 = off, 1 = job chain, 2 = coop job
r_usePBOTextureFeedback	bool	1	no	use async pixel buffer read for texture feedback query
r_userOptionSkipGPUCulling	int	-1	yes	Can be modified from Video Options. If it's positive, r_skipGPUTriangleCulling is set to this value when the game starts
r_useSMP	int	2	no	1 = run the renderer in a separate thread, 2 = run the renderer in jobs
r_useStateCaching	bool	1	no	avoid redundant state changes in GL_*() calls
r_useStaticShadows	bool	1	no	load static shadows for map (change requires re-load)
r_useStaticTransparencyGenJobs	int	1	no	0 = off, 1 = threaded
r_useTransparencyEmissivePass	int	1	no	render quads marked emissive to the emissive pass
r_useTransparencySort	bool	1	no	0 = do not sort the transparency quads
r_useTransparencySortJobs	int	1	no	0 = off, 1 = threaded
r_useUmbraCulling	bool	1	no	enable Umbra occlusion culling and computation
r_useUmbraPartitioning	bool	0	no	Use Umbra partitioning for world areas and BSP
r_useUmbraSubspaceOcclusion	int	2	no	0 = off, 1 = enable subspace bounds sorting and occlusion (no nodes), 2 = no sorting
r_useVSM	int	0	no	variance shadow maps
r_useWaterSimulationJobs	int	1	no	0 = off, 1 = threaded
r_useZiplineGenJobs	int	1	no	0 = off, 1 = threaded
r_validateRenderProgs	int	1	no	Validates shaders and points out potential shader patch issues
r_verboseShaderCompilation	bool	0	no	Print cmdline for each shader compilation task
r_viewBoundsRange	float	0.5	no	Distance at which to stop occlusion culling and decide that models are scene models
r_vignette	bool	1	no	enable vignetting
r_vmtrPhysicalPagesAspectRatio	float	2.0	yes	Set to change the aspect ratio of physical pages for the vmtr.
r_vulkanPreferPhysicalDevice	string	(empty)	no	Prefer this vulkan device name if encountered
r_waitForQuery	bool	1	no	Stall until occlusion query returns
r_waitForVirtualMaterialFinishFeedback	bool	0	no	forces the finish feedback function to wait rather than timeout
r_water	bool	1	no	0 = disabled, 1 = enabled (default)
r_waterCaustics	bool	1	yes	0 = disabled, 1 = enabled (default)
r_waterCausticsGridVertsScale	int	0	no	1 = 256 x 256 photons, 2 = 512 x 256 photons, 3 = 1024 x 256 photons, etc
r_waterCausticsResolution	int	1024	no	1024 x 1024 default
r_waterDisplacementCapture	bool	1	no	Enable vmem to cpu transfer of water positions
r_waterDisplacementDebug	bool	0	no	Debug water positions nearby camera
r_waterGbufferPrimeDepthHTile	bool	0	no	Enable priming g-buffer depth h-tile
r_waterGridDebug	int	0	no	0 = disabled, 1 = enabled (default)
r_waterGridDepthResolution	int	256	no	256x256 default. Low res depth buffer for water cage mesh.
r_waterGridOffsetFov	float	7.5	no	0 = no offset. Adds Fov offset to water grid view projection
r_waterGridQuadPrimitive	bool	0	no	0 = triangles, 1 = quadrangles
r_waterGridResolution	int	256	yes	256x256 default ( pc ), 224 x 224 for xb1 / low
r_waterGridTAA	bool	1	no	0 = disabled, 1 = enabled (default)
r_waterHalfRes	bool	1	no	Enable half-resolution usage for water for faster performance
r_waterHitsSim	bool	1	yes	0 = disabled, 1 = enabled (default)
r_waterHitsSimResolution	int	512	no	512x512 default
r_waterInterleaveUpdates	int	1	yes	0 = disabled, 1 = enabled ( interleaves updates for techniques with similar costs )
r_waterLodViewDistance	float	16.0f	yes	Disable/enable water related phenomena (for now caustics / hits simulation ) after threshold distance
r_watermarkDelay	int	2000	no	The number of milliseconds between watermark position changes
r_waterNormalsAnisoFilter	float	4.0	no	0 = no anisotropic filter. 2.0, 4.0, 8.0, 16.0
r_waterPostProcess	bool	1	yes	Enable inside/outside water post process transitions
r_waterQualityFFT	int	0	no	0 = 256x256, 1 = 512x512
r_waterReflections	int	1	yes	0 = disabled, 1 = enabled (default)
r_waterReflectionsTAA	bool	1	no	0 = disabled, 1 = enabled (default)
r_waterUpdateFFT	int	1	no	0 = disabled, 1 = enabled (default)
r_waterUpdateGridVerts	bool	1	no	0 = disabled, 1 = enabled (default)
r_windowHeight	int	0	yes	window height when running in a window
r_windowPosX	int	0	yes	window X position when running in a window
r_windowPosY	int	0	yes	window Y position when running in a window
r_windowWidth	int	0	yes	window width when running in a window
r_writePackedMaterials	bool	0	no	debug option to write all the packed materials out to viewable tga files as they are retrieved
r_znear	float	0.06	no	near Z clip plane distance
r_znearOverride	float	0.0	no	Override near Z if greater than zero

vt_

114 cvars.

Name	Type	Default	Exposed	Description
vt_autoReloadBinaryVirtualMaterials	int	3	no	Automatically reload bvmaterials when changed
vt_autoReloadBinaryVirtualMaterialsDelay	int	16	no	Automatically reload bvmaterials when changed
vt_autoReloadSuspendReloadOnMaterialChange	bool	0	no	Stop reloading of vmtr when material checksum is changed
vt_binaryVirtualMaterialsAsync	bool	1	no	Set to load binary virtual materials in the background
vt_binaryVmtrStats	int	0	no	1 = print stats for binary vmtr loading, 2 = add evictions
vt_containers	bool	0	no	Set to enable loading from containers
vt_dumpHDPData	int	0	no	set to enable writing of all compressed data to the hard drive for analysis
vt_eliminateSeamsInVmtr	bool	0	no	Worse packing to eliminate black borders in vmtrs.
vt_errorOnNoPageFile	bool	1	no	Fatal error if page file is missing or can't be created
vt_feedbackHighWater	float	90	no	More than this percentage of pages needed this frame and we consider ourselves oversubscribed.
vt_feedbackLodBias	float	0	no	Additional floating point bias for feedback LOD
vt_feedbackLODIncrement	float	0.125	no	Base increment value for adjusting dynamic LOD bias.
vt_feedbackLowWater	float	80	no	Less than this percentage of pages needed this frame and we consider ourselves undersubscribed.
vt_filePath	string	(empty)	yes	Path for the virtual texture
vt_filePathVmtrOverride	string	generated/pagefiles	yes	Optionally get vmtr files from a different source than vt_filePath
vt_filter	int	2	yes	filter physical images, 0 = off, 1 = bilinear, 2 = trilinear
vt_genMissingPages	int	0	no	Gen a page in memory for pages that don't exist on disk.
vt_glSyncOnPageUpload	bool	1	no	sync with fences around the page upload (to make sure we don't write to the pbo before the last upload
vt_loadMegaTexture2	bool	1	no	Allow loading of the new megatexture file format.
vt_lockExplicitPages	int	1	no	Set to true to lock pages
vt_lodBias	float	0	no	Additional floating point bias for virtual texture page tables
vt_maxAniso	float	4	yes	Floating point max ansio for physical textures.
vt_maxLockedPagesPercent	int	20	yes	Amount of physical page pool to lock in percent; -1 = use default, [0..50] = use that percentage
vt_maxLod	float	12	no	Set the coarsest mip that can be accessed.
vt_maxPageFileLod	int	6	no	Set the coarsest mip that can be accessed by the pagefile.
vt_maxPPF	int	32	yes	maximum number of transcoded pages per frame
vt_minLockedVmtrLOD	int	6	yes	Lock from coarsest LOD down to this LOD level in the vmtr physical page pool; [0..10] = lock from coarsest(11) down to this mip LOD
vt_minLod	int	0	no	Set the finest mip that can be accessed.
vt_minMipLevelInstalled	int	0	no	minimum mip level installed on the hard disk (2 = all mips except finest two installed, 16 = no mips installed)
vt_nopreload	bool	0	no	—
vt_noSeekCacheLines	int	4	no	number of cache lines we can skip without seeking
vt_numAnalyzeJobs	int	4	yes	Number of analyze jobs to run if vt_useAnalyzeJobs = 2
vt_numBinaryVirtualMips	int	12	no	Set to the number of binary virtual mips to build
vt_numBinaryVmtrThreads	int	1	no	Number of threads to create for jobs.
vt_packPageFiles	bool	0	yes	Use the virtual texture file name without any suffixes
vt_pageBorderAdjust	float	0	no	Debug adjustment of a page's mapping about it's center
vt_pageFadeInSteps	int	3	no	—
vt_pageImageSizeUnique	int	8192	yes	Dimensions of the physical page images. vt_restart after changing this.
vt_pageImageSizeVmtr	int	8192	yes	Dimensions of the physical page images. vt_restart after changing this.
vt_pageImageSizeVmtrUnique	int	128	yes	Dimensions of the physical page images.
vt_printInfo	bool	0	no	Print virtual texture information during initialization, likely only interesting to rendering and engine coders
vt_productionFilePath	string	virtualtextures	yes	path appended for virtual textures in production builds
vt_qualityBC6HLightmap	int	2	no	0 = highest, 1 = high, 2 = medium, 3 = low, 4 = lowest
vt_qualityBC7ColorMask	int	2	no	0 = highest, 1 = high, 2 = medium, 3 = low, 4 = lowest
vt_qualityBC7ColorMaskTranscode	int	4	no	BC7 quality when transcoding. 0 = highest, 1 = high, 2 = medium, 3 = low, 4 = lowest
vt_qualityHDPDiffuse	int	30	no	amount of quantization to apply in the range [0-255] where lower is better quality
vt_qualityHDPLightmap	int	30	no	amount of quantization to apply in the range [0-255] where lower is better quality
vt_qualityHDPLossless	bool	0	no	force lossless HD-Photo compression
vt_qualityHDPNormal	int	30	no	amount of quantization to apply in the range [0-255] where lower is better quality
vt_qualityHDPPower	int	30	no	amount of quantization to apply in the range [0-255] where lower is better quality
vt_qualityLightmapRoundDenominator	float	0	no	Lightmap round factor 1.0/val. <= 0 to turn it off
vt_reconstructMissingMips	bool	1	no	Reconstructs the missing mips if set
vt_recordStats	string	(empty)	no	record virtual texture system statistics to the given file name
vt_regionHashBucketSize	int	1024	no	Number of region hash buckets on a side
vt_scaleNormalsBySpecularCutoff	float	64.0	no	cutoff for the specular below which the normals start getting blended to blue
vt_shapePPF	bool	1	no	Shape maximum number of transcoded pages per frame
vt_showBinaryVirtualMaterialsTimings	int	0	no	Set to show the timinigs of the virtual material system
vt_showExplicitPages	bool	0	no	Report the number of explicit pages each frame.
vt_showFeedbackAnalysis	int	0	no	Report all the unique pages visible
vt_showFeedbackBuffer	int	0	no	visualize the feedback buffer
vt_showFeedbackStreaming	int	0	no	Display the feedback pages as they are streamed in
vt_showMissingBvimages	bool	0	no	Set to show missing bvimages
vt_showPageBorders	bool	0	yes	Shows the page borders
vt_showPageColors	bool	0	yes	use a random color for each page
vt_showPageInfo	bool	0	yes	Shows page info: source #, layout #, compression, mono/color specular
vt_showPageLoads	int	0	yes	Print each page that is loaded
vt_showPageMips	int	0	yes	Colorize pages by virtual mip level, if > 1, only color pages beyond that level-1
vt_showPageNumbers	bool	0	yes	Draw page numbers
vt_showPageSizes	int	0	yes	Draw page sizes (1 = show size in bytes, 2 = color green to red from small to large)
vt_showPageSources	bool	0	yes	Shades vmtr and unique pages differently
vt_showPageUsage	int	0	yes	color each page from red to green based on the percentage of the page texels that are actually visible
vt_showStats	bool	0	yes	show virtual texture system statistics
vt_showVmtrAccesses	bool	0	no	shows all vmtr accesses in feedback buffer
vt_singleStep	int	-1	no	single step through the transcode process
vt_skipBVImages	bool	0	no	set to skip upload of bvimages
vt_skipBVStaleCheck	bool	0	no	Disable the bvimage stalechecking
vt_skipExplicitPageFeedback	bool	0	no	don't append explicit pages from gui surfaces.
vt_skipFindingVmtrFromAbsolutePoint	bool	0	no	Skips the lookup and subsequent reading of binary virtual materials during transcode
vt_skipLZDecompress	bool	0	no	Skips the LZW decode
vt_skipPhysicalTextureUploads	bool	0	no	—
vt_skipStreaming	bool	0	no	Skip streaming but still transcode and update the page tables
vt_skipTranscode	bool	0	no	Skip transcoding (but don't skip reading)
vt_skipUniquePageFiles	bool	0	no	If true then set empty baked pagefiles rather than trying to open them (useful when running outside office networks)
vt_skipUpdatePageImages	bool	0	no	Skip updating page table and physical images on the PC (but still read/transcode)
vt_skipVirtualMaterials	bool	0	no	1 = Don't load virtual materials - ONLY FOR TESTING!
vt_stripPageBorders	bool	0	no	Strip the page border by upscaling the center
vt_supressCpuVmtrs	bool	1	no	Set to use supress cpuvmtr building and loading entirely
vt_textureFilter	int	2	no	0 = Box, 1 = Cubic, 2 = Mitchell, 3 = Lanczos
vt_timeBC6HCompression	bool	0	no	—
vt_toksvigAA	bool	1	no	Power will be affected by the estimated bump variance to reduce shader aliasing
vt_truncateConstantPayloads	bool	1	no	Truncate constant payloads in a binary virtual image
vt_uncompressedPhysicalImages	bool	0	yes	Use uncompressed physical page images. vt_restart after changing this.
vt_useAnalyzeJobs	int	2	yes	0 = off, 1 = threaded, 2 = threaded, multiple sequential jobs
vt_useBinaryVirtualMaterials	bool	1	no	Set to use binary virtual materials in the feedback system
vt_useCachedOpenFilesForBinaryVmtrs	bool	0	no	Force the use of a limited number of opened BVMTR files, should only be applicable to PS4 in top of tree ( development ) runs
vt_useDLCVmtrPageFiles	bool	1	yes	Loads the dlc vmtr list for storing pages in the dlc quadrant.
vt_useIntroVmtrPageFiles	bool	0	no	Loads the intro vmtr list for storing pages in the first quadrant.
vt_useLZ4Compression	bool	1	no	Use LZ4 instead of LZW
vt_useMinLodTexture	bool	0	no	Use a min-LOD texture to gradually adjust the LOD when new texture pages become resident
vt_useNetworkPagesWithPreload	bool	0	no	Use the network page files and vmtr even when using linear preload files
vt_useNonBufferedIO	bool	0	no	use non-buffered IO
vt_usePageFileCache	bool	1	no	Use the pageFileCache for all reads
vt_usePageMinLodJobs	int	1	no	0 = off, 1 = threaded
vt_usePageSortJobs	int	1	yes	0 = off, 1 = threaded
vt_usePageTablePBOs	bool	1	no	use the PBOs for page table updates on the PC
vt_usePersistentPBO	bool	1	no	Use persistent PBOs for virtual texturing uploads
vt_usePreCompressedVmtrs	bool	1	no	Set to 1 to load pre-compressed vmtrs for faster vmtr generation times.
vt_useTranscodeJobs	int	1	yes	0 = off, 1 = threaded
vt_useTransferQueue	bool	1	no	Use transfer queue for page updates if available
vt_useUnsynchronizedPageTablePBOs	bool	0	no	use unsynchronized PBOs for page table updates on the PC
vt_useVmtrPageFiles	bool	0	no	Enable or disable reading and writing to the vmtr page files.
vt_useVmtrRegionHash	bool	0	no	If set, then uses the region hash to look up vmtrs during development
vt_virtualSizeLog2	int	11	no	Log2 of the virtualTexture dimensions for general texture allocation. Must be set on the command line.
vt_writeDebugImage	bool	1	no	1 = write out vmtr debug image

Smaller prefixes

30 cvars.

Name	Type	Default	Exposed	Description
atlas_printFilter	int	-1	no	fikter usage print by atlas id
atlas_printUsage	bool	0	no	set to 1 print out how much of the atlas has been used during allocation
bink_audioScaleCinematic	float	0.7	no	cinematic audio volume scale
bink_audioSyncTolerance	int	2	no	number of frames of of drift between the bink and the sound engine audio to tolerate
bink_audioSyncToSoundEngine	bool	1	no	trigger bink audio using audio engine playback instead of playing back the embedded audio
bink_debug	bool	0	no	dump information
bink_GotoFrameInsteadOfNextFrame	bool	0	no	Use BinkGoto Instead of BinkNextFrame, because of com_fixedTic
bink_ioMegs	int	4	no	larger buffers for less seeks during load
bink_lerpInAudioFrames	int	200	no	number of frames to ramp up the audio over
bink_lerpOutAudioFrames	int	30	no	number of frames to ramp down the audio over
bink_useIOThreadForBinkIO	int	2	no	1: use a background thread to read data for bink in large chunks, 2: use stream file cache
bvmaterial_forceUseCooked	bool	0	no	generate a new cooked version if stale
bvmaterial_parallelGeneration	bool	1	no	generate separate bvimages in parallel
bvmaterial_precompressed_parallelGeneration	bool	1	no	generate separate bvimages in parallel
decal_ForceFadeOutDuration	time_ms	10000	no	How long to take when fading a killed decal
decal_forcePersistent	bool	0	no	Force all decals to be added as persistent
decal_MaxDecalGroupingRadius	float	3.0f	no	Max Radius we use to group decals
gl_threadWaitSelection	int	2	no	How long to wait for serialization jobs to finish. 0 = shortest. wait immediately, 1 = middle. wait at the end of ReadSnapshot, 2 = longest. wait until right before PostSerializeRead
md6_allowFailedLodModelToSucceed	bool	1	no	This is only here to allow nightly to no break 2017.05.29, remove once flashback fergus mesh is fixed
md6_allowGenericRemap	bool	0	no	allow remapping materials to generic ones for more aggressive merging
md6_attemptRewrite	bool	1	no	attempt to rewrite source if out of date (disabled for cooker)
md6_generateTexcoordRebase	bool	1	no	Re-base texture coordinates if trying to generic map MD6 surfaces.
multiView_60Hz	bool	1	no	0 = alternate frame rendering, 1 = render both each frame
stereoRender_guiOffset	float	0	no	shift guis so they don't appear at infinity in HMDs
stereoRender_screenSeparation	float	0.25	no	screen units from center to eyes
stereoRender_separation	float	1.5	no	world units from center to eyes
stereoRender_swapEyes	float	0	no	swap target buffers for left and right eyes
vr_enable	bool	0	no	Enable to run in VR mode.
vr_metersToGameUnits	float	1.0	no	How many game units one meter represents.
vr_radialMaskingRatio	float	0.30	no	Radius of non-masked area