Buckets:
| import { Color, Vector2, NearestFilter, Matrix4, RendererUtils, PassNode, QuadMesh, NodeMaterial } from 'three/webgpu'; | |
| import { add, float, If, Loop, int, Fn, min, max, clamp, nodeObject, texture, uniform, uv, vec2, vec4, luminance } from 'three/tsl'; | |
| const _quadMesh = /*@__PURE__*/ new QuadMesh(); | |
| const _size = /*@__PURE__*/ new Vector2(); | |
| let _rendererState; | |
| /** | |
| * A special render pass node that renders the scene with TRAA (Temporal Reprojection Anti-Aliasing). | |
| * | |
| * Note: The current implementation does not yet support MRT setups. | |
| * | |
| * References: | |
| * - {@link https://alextardif.com/TAA.html} | |
| * - {@link https://www.elopezr.com/temporal-aa-and-the-quest-for-the-holy-trail/} | |
| * | |
| * @augments PassNode | |
| * @three_import import { traaPass } from 'three/addons/tsl/display/TRAAPassNode.js'; | |
| */ | |
| class TRAAPassNode extends PassNode { | |
| static get type() { | |
| return 'TRAAPassNode'; | |
| } | |
| /** | |
| * Constructs a new TRAA pass node. | |
| * | |
| * @param {Scene} scene - The scene to render. | |
| * @param {Camera} camera - The camera to render the scene with. | |
| */ | |
| constructor( scene, camera ) { | |
| super( PassNode.COLOR, scene, camera ); | |
| /** | |
| * This flag can be used for type testing. | |
| * | |
| * @type {boolean} | |
| * @readonly | |
| * @default true | |
| */ | |
| this.isTRAAPassNode = true; | |
| /** | |
| * The clear color of the pass. | |
| * | |
| * @type {Color} | |
| * @default 0x000000 | |
| */ | |
| this.clearColor = new Color( 0x000000 ); | |
| /** | |
| * The clear alpha of the pass. | |
| * | |
| * @type {number} | |
| * @default 0 | |
| */ | |
| this.clearAlpha = 0; | |
| /** | |
| * The jitter index selects the current camera offset value. | |
| * | |
| * @private | |
| * @type {number} | |
| * @default 0 | |
| */ | |
| this._jitterIndex = 0; | |
| /** | |
| * Used to save the original/unjittered projection matrix. | |
| * | |
| * @private | |
| * @type {Matrix4} | |
| */ | |
| this._originalProjectionMatrix = new Matrix4(); | |
| /** | |
| * A uniform node holding the inverse resolution value. | |
| * | |
| * @private | |
| * @type {UniformNode<vec2>} | |
| */ | |
| this._invSize = uniform( new Vector2() ); | |
| /** | |
| * The render target that holds the current sample. | |
| * | |
| * @private | |
| * @type {?RenderTarget} | |
| * @default null | |
| */ | |
| this._sampleRenderTarget = null; | |
| /** | |
| * The render target that represents the history of frame data. | |
| * | |
| * @private | |
| * @type {?RenderTarget} | |
| * @default null | |
| */ | |
| this._historyRenderTarget = null; | |
| /** | |
| * Material used for the resolve step. | |
| * | |
| * @private | |
| * @type {NodeMaterial} | |
| */ | |
| this._resolveMaterial = new NodeMaterial(); | |
| this._resolveMaterial.name = 'TRAA.Resolve'; | |
| } | |
| /** | |
| * Sets the size of the effect. | |
| * | |
| * @param {number} width - The width of the effect. | |
| * @param {number} height - The height of the effect. | |
| * @return {boolean} Whether the TRAA needs a restart or not. That is required after a resize since buffer data with different sizes can't be resolved. | |
| */ | |
| setSize( width, height ) { | |
| super.setSize( width, height ); | |
| let needsRestart = false; | |
| if ( this.renderTarget.width !== this._sampleRenderTarget.width || this.renderTarget.height !== this._sampleRenderTarget.height ) { | |
| this._sampleRenderTarget.setSize( this.renderTarget.width, this.renderTarget.height ); | |
| this._historyRenderTarget.setSize( this.renderTarget.width, this.renderTarget.height ); | |
| this._invSize.value.set( 1 / this.renderTarget.width, 1 / this.renderTarget.height ); | |
| needsRestart = true; | |
| } | |
| return needsRestart; | |
| } | |
| /** | |
| * This method is used to render the effect once per frame. | |
| * | |
| * @param {NodeFrame} frame - The current node frame. | |
| */ | |
| updateBefore( frame ) { | |
| const { renderer } = frame; | |
| const { scene, camera } = this; | |
| _rendererState = RendererUtils.resetRendererState( renderer, _rendererState ); | |
| // | |
| this._pixelRatio = renderer.getPixelRatio(); | |
| const size = renderer.getSize( _size ); | |
| const needsRestart = this.setSize( size.width, size.height ); | |
| // save original/unjittered projection matrix for velocity pass | |
| camera.updateProjectionMatrix(); | |
| this._originalProjectionMatrix.copy( camera.projectionMatrix ); | |
| // camera configuration | |
| this._cameraNear.value = camera.near; | |
| this._cameraFar.value = camera.far; | |
| // configure jitter as view offset | |
| const viewOffset = { | |
| fullWidth: this.renderTarget.width, | |
| fullHeight: this.renderTarget.height, | |
| offsetX: 0, | |
| offsetY: 0, | |
| width: this.renderTarget.width, | |
| height: this.renderTarget.height | |
| }; | |
| const originalViewOffset = Object.assign( {}, camera.view ); | |
| if ( originalViewOffset.enabled ) Object.assign( viewOffset, originalViewOffset ); | |
| const jitterOffset = _JitterVectors[ this._jitterIndex ]; | |
| camera.setViewOffset( | |
| viewOffset.fullWidth, viewOffset.fullHeight, | |
| viewOffset.offsetX + jitterOffset[ 0 ] * 0.0625, viewOffset.offsetY + jitterOffset[ 1 ] * 0.0625, // 0.0625 = 1 / 16 | |
| viewOffset.width, viewOffset.height | |
| ); | |
| // configure velocity | |
| const mrt = this.getMRT(); | |
| const velocityOutput = mrt.get( 'velocity' ); | |
| if ( velocityOutput !== undefined ) { | |
| velocityOutput.setProjectionMatrix( this._originalProjectionMatrix ); | |
| } else { | |
| throw new Error( 'THREE:TRAAPassNode: Missing velocity output in MRT configuration.' ); | |
| } | |
| // render sample | |
| renderer.setMRT( mrt ); | |
| renderer.setClearColor( this.clearColor, this.clearAlpha ); | |
| renderer.setRenderTarget( this._sampleRenderTarget ); | |
| renderer.render( scene, camera ); | |
| renderer.setRenderTarget( null ); | |
| renderer.setMRT( null ); | |
| // every time when the dimensions change we need fresh history data. Copy the sample | |
| // into the history and final render target (no AA happens at that point). | |
| if ( needsRestart === true ) { | |
| // bind and clear render target to make sure they are initialized after the resize which triggers a dispose() | |
| renderer.setRenderTarget( this._historyRenderTarget ); | |
| renderer.clear(); | |
| renderer.setRenderTarget( this.renderTarget ); | |
| renderer.clear(); | |
| renderer.setRenderTarget( null ); | |
| renderer.copyTextureToTexture( this._sampleRenderTarget.texture, this._historyRenderTarget.texture ); | |
| renderer.copyTextureToTexture( this._sampleRenderTarget.texture, this.renderTarget.texture ); | |
| } else { | |
| // resolve | |
| renderer.setRenderTarget( this.renderTarget ); | |
| _quadMesh.material = this._resolveMaterial; | |
| _quadMesh.render( renderer ); | |
| renderer.setRenderTarget( null ); | |
| // update history | |
| renderer.copyTextureToTexture( this.renderTarget.texture, this._historyRenderTarget.texture ); | |
| } | |
| // copy depth | |
| renderer.copyTextureToTexture( this._sampleRenderTarget.depthTexture, this.renderTarget.depthTexture ); | |
| // update jitter index | |
| this._jitterIndex ++; | |
| this._jitterIndex = this._jitterIndex % ( _JitterVectors.length - 1 ); | |
| // restore | |
| if ( originalViewOffset.enabled ) { | |
| camera.setViewOffset( | |
| originalViewOffset.fullWidth, originalViewOffset.fullHeight, | |
| originalViewOffset.offsetX, originalViewOffset.offsetY, | |
| originalViewOffset.width, originalViewOffset.height | |
| ); | |
| } else { | |
| camera.clearViewOffset(); | |
| } | |
| velocityOutput.setProjectionMatrix( null ); | |
| RendererUtils.restoreRendererState( renderer, _rendererState ); | |
| } | |
| /** | |
| * This method is used to setup the effect's render targets and TSL code. | |
| * | |
| * @param {NodeBuilder} builder - The current node builder. | |
| * @return {PassTextureNode} | |
| */ | |
| setup( builder ) { | |
| if ( this._sampleRenderTarget === null ) { | |
| this._sampleRenderTarget = this.renderTarget.clone(); | |
| this._historyRenderTarget = this.renderTarget.clone(); | |
| this._sampleRenderTarget.texture.minFiler = NearestFilter; | |
| this._sampleRenderTarget.texture.magFilter = NearestFilter; | |
| const velocityTarget = this._sampleRenderTarget.texture.clone(); | |
| velocityTarget.isRenderTargetTexture = true; | |
| velocityTarget.name = 'velocity'; | |
| this._sampleRenderTarget.textures.push( velocityTarget ); // for MRT | |
| } | |
| // textures | |
| const historyTexture = texture( this._historyRenderTarget.texture ); | |
| const sampleTexture = texture( this._sampleRenderTarget.textures[ 0 ] ); | |
| const velocityTexture = texture( this._sampleRenderTarget.textures[ 1 ] ); | |
| const depthTexture = texture( this._sampleRenderTarget.depthTexture ); | |
| const resolve = Fn( () => { | |
| const uvNode = uv(); | |
| const minColor = vec4( 10000 ).toVar(); | |
| const maxColor = vec4( - 10000 ).toVar(); | |
| const closestDepth = float( 1 ).toVar(); | |
| const closestDepthPixelPosition = vec2( 0 ).toVar(); | |
| // sample a 3x3 neighborhood to create a box in color space | |
| // clamping the history color with the resulting min/max colors mitigates ghosting | |
| Loop( { start: int( - 1 ), end: int( 1 ), type: 'int', condition: '<=', name: 'x' }, ( { x } ) => { | |
| Loop( { start: int( - 1 ), end: int( 1 ), type: 'int', condition: '<=', name: 'y' }, ( { y } ) => { | |
| const uvNeighbor = uvNode.add( vec2( float( x ), float( y ) ).mul( this._invSize ) ).toVar(); | |
| const colorNeighbor = max( vec4( 0 ), sampleTexture.sample( uvNeighbor ) ).toVar(); // use max() to avoid propagate garbage values | |
| minColor.assign( min( minColor, colorNeighbor ) ); | |
| maxColor.assign( max( maxColor, colorNeighbor ) ); | |
| const currentDepth = depthTexture.sample( uvNeighbor ).r.toVar(); | |
| // find the sample position of the closest depth in the neighborhood (used for velocity) | |
| If( currentDepth.lessThan( closestDepth ), () => { | |
| closestDepth.assign( currentDepth ); | |
| closestDepthPixelPosition.assign( uvNeighbor ); | |
| } ); | |
| } ); | |
| } ); | |
| // sampling/reprojection | |
| const offset = velocityTexture.sample( closestDepthPixelPosition ).xy.mul( vec2( 0.5, - 0.5 ) ); // NDC to uv offset | |
| const currentColor = sampleTexture.sample( uvNode ); | |
| const historyColor = historyTexture.sample( uvNode.sub( offset ) ); | |
| // clamping | |
| const clampedHistoryColor = clamp( historyColor, minColor, maxColor ); | |
| // flicker reduction based on luminance weighing | |
| const currentWeight = float( 0.05 ).toVar(); | |
| const historyWeight = currentWeight.oneMinus().toVar(); | |
| const compressedCurrent = currentColor.mul( float( 1 ).div( ( max( max( currentColor.r, currentColor.g ), currentColor.b ).add( 1.0 ) ) ) ); | |
| const compressedHistory = clampedHistoryColor.mul( float( 1 ).div( ( max( max( clampedHistoryColor.r, clampedHistoryColor.g ), clampedHistoryColor.b ).add( 1.0 ) ) ) ); | |
| const luminanceCurrent = luminance( compressedCurrent.rgb ); | |
| const luminanceHistory = luminance( compressedHistory.rgb ); | |
| currentWeight.mulAssign( float( 1.0 ).div( luminanceCurrent.add( 1 ) ) ); | |
| historyWeight.mulAssign( float( 1.0 ).div( luminanceHistory.add( 1 ) ) ); | |
| return add( currentColor.mul( currentWeight ), clampedHistoryColor.mul( historyWeight ) ).div( max( currentWeight.add( historyWeight ), 0.00001 ) ); | |
| } ); | |
| // materials | |
| this._resolveMaterial.fragmentNode = resolve(); | |
| return super.setup( builder ); | |
| } | |
| /** | |
| * Frees internal resources. This method should be called | |
| * when the effect is no longer required. | |
| */ | |
| dispose() { | |
| super.dispose(); | |
| if ( this._sampleRenderTarget !== null ) { | |
| this._sampleRenderTarget.dispose(); | |
| this._historyRenderTarget.dispose(); | |
| } | |
| this._resolveMaterial.dispose(); | |
| } | |
| } | |
| export default TRAAPassNode; | |
| // These jitter vectors are specified in integers because it is easier. | |
| // I am assuming a [-8,8) integer grid, but it needs to be mapped onto [-0.5,0.5) | |
| // before being used, thus these integers need to be scaled by 1/16. | |
| // | |
| // Sample patterns reference: https://msdn.microsoft.com/en-us/library/windows/desktop/ff476218%28v=vs.85%29.aspx?f=255&MSPPError=-2147217396 | |
| const _JitterVectors = [ | |
| [ - 4, - 7 ], [ - 7, - 5 ], [ - 3, - 5 ], [ - 5, - 4 ], | |
| [ - 1, - 4 ], [ - 2, - 2 ], [ - 6, - 1 ], [ - 4, 0 ], | |
| [ - 7, 1 ], [ - 1, 2 ], [ - 6, 3 ], [ - 3, 3 ], | |
| [ - 7, 6 ], [ - 3, 6 ], [ - 5, 7 ], [ - 1, 7 ], | |
| [ 5, - 7 ], [ 1, - 6 ], [ 6, - 5 ], [ 4, - 4 ], | |
| [ 2, - 3 ], [ 7, - 2 ], [ 1, - 1 ], [ 4, - 1 ], | |
| [ 2, 1 ], [ 6, 2 ], [ 0, 4 ], [ 4, 4 ], | |
| [ 2, 5 ], [ 7, 5 ], [ 5, 6 ], [ 3, 7 ] | |
| ]; | |
| /** | |
| * TSL function for creating a TRAA pass node for Temporal Reprojection Anti-Aliasing. | |
| * | |
| * @tsl | |
| * @function | |
| * @param {Scene} scene - The scene to render. | |
| * @param {Camera} camera - The camera to render the scene with. | |
| * @returns {TRAAPassNode} | |
| */ | |
| export const traaPass = ( scene, camera ) => nodeObject( new TRAAPassNode( scene, camera ) ); | |
Xet Storage Details
- Size:
- 12.4 kB
- Xet hash:
- 253a42301e7c0688a6d0c9d611341b0fe13be391ba8a5a443f40d3399ea63ee5
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.