Nodes: Access Remaining Compute Builtins #29469
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
📦 Bundle sizeFull ESM build, minified and gzipped.
🌳 Bundle size after tree-shakingMinimal build including a renderer, camera, empty scene, and dependencies.
|
cmhhelgeson
force-pushed
the
webgpu_access_compute_builtin
branch
from
7374f40
to
b3c0fbf
Compare
LD2Studio
pushed a commit
to LD2Studio/LD2Studio-Editor
that referenced
this pull request
Sep 27, 2024
BatchedMesh: add `deleteInstance()` (mrdoob#29449) * BatchedMesh: add `deleteInstance()` * BatchedMesh: prioritize reusing freed instance ids when adding instance Update Three.js r169 Lensflare: Add WebGPU version. (mrdoob#29451) * lensflare example * fix depth * remove forceWebGL * rename * fix typo * follow original code * Update LensflareMesh.js --------- Co-authored-by: aardgoose <angus.sawyer@email.com> Co-authored-by: Michael Herzog <michael.herzog@human-interactive.org> Examples: Update lensflare colors. (mrdoob#29458) WebGLRenderer: add reverse-z via EXT_clip_control (mrdoob#29445) * WebGLRenderer: add reverse-z via EXT_clip_control * WebGLRenderer: move conversion methods to utils Updated builds. Docs: Improve `WebGLRenderer` page. (mrdoob#29459) WebGLProgram: add USE_REVERSEDEPTHBUF define (mrdoob#29461) CylinderGeometry: Don't add degenerate triangles (mrdoob#29460) * CylinderGeometry: Don't add degenerate triangles * Change comparison RenderObject: Introduce `getGeometryCacheKey()` (mrdoob#29465) * RenderObject: Added `getGeometryCacheKey()` * NodeMaterialObserver: Improve skinnedmesh and morph supports * cleanup Update actions/setup-node digest to 0a44ba7 (mrdoob#29466) Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com> Update github/codeql-action digest to 294a9d9 (mrdoob#29467) Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com> Update devDependencies (non-major) (mrdoob#29468) Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com> Updated builds. CurveModifierGPU: WebGPURenderer port. (mrdoob#29453) * curve mod * lint * update screenshot * use correct constant * use filtered texture --------- Co-authored-by: aardgoose <angus.sawyer@email.com> Examples: Clean up. (mrdoob#29473) * Examples: Clean up. * E2E: Update screenshot. SpriteNodeMaterial: Add sizeAttenuation (mrdoob#29394) * SpriteNodeMaterial: Add sizeAttenuation * Regenerate screenshot * revert mat4 mul with vec4(,1)? * revert * using let and toVar fixes the issue? * merge upstream/dev * revert screenshot * does reverting SpriteNode even fix? * revert fix * revert everything fix tsl galaxy? * checkout dev SpriteNodeMaterial.js... * smaller change test * convert scale to var * convert scale to var * simply adding toVar breaks tsl_galaxy? * fix multiple spritematerial with different center position * feedbacks * cleanup * feedbacks * cleanup * cleanup --------- Co-authored-by: sunag <sunagbrasil@gmail.com> WebGPURenderer: Prevent out of bounds `textureLoad` access in WGSL (mrdoob#29470) Co-authored-by: aardgoose <angus.sawyer@email.com> SkyMesh,WaterMesh: Fix NodeMaterial imports (mrdoob#29477) WebGPURenderer: Reuse LightNode when available (mrdoob#29480) * WebGPURenderer: Reuse LightNode when available * cleanup DecalGeometry: Transform normal with normal matrix (mrdoob#29476) * Transform normal with normal matrix * Transform normal with normal matrix Examples: add reflection mask in `retargeting_readyplayer` (mrdoob#29485) BatchedMesh: Add `getGeometryRangeAt` (mrdoob#29409) * BatchedMesh: Add getGeometryRangeAt Co-authored-by: Luigi Denora <luigidenora@users.noreply.github.com> * Fix eslint error Co-authored-by: Luigi Denora <luigidenora@users.noreply.github.com> * Doc changed * Add optional target --------- Co-authored-by: Luigi Denora <luigidenora@users.noreply.github.com> WebGPURenderer: Introduce hash-based cache key (mrdoob#29479) * introduce numeric cache key * cleanup * cleanup * simplification * revision * rev * rev * cleanup Nodes: Access Remaining Compute Builtins (mrdoob#29469) * init * remove duplicated function Add decal as child of mesh (mrdoob#29486) Update webgl_postprocessing_ssaa.html Update webgpu_postprocessing_ssaa.html ToonOutlinePassNode: Add FX pass for toon outlines. (mrdoob#29483) * ToonOutlineNode: Add FX pass for toon outlines. * ToonOutlinePassNode: Refactor code. * E2E: Update screenshot. * ToonOutlinePassNode: Clean up. * ToonOutlinePassNode: More clean up. * ToonOutlinePassNode: More clean up. Update Addons.js (mrdoob#29493) leftovers after removing `PackedPhongMaterial ` mrdoob#29382 ReferenceNode: Fix null reference using `getNodeType()` (mrdoob#29498) * fix null reference using `getNodeType()` * add `sprite.center` check NodeBuilder: Introduce `addFlowCodeHierarchy()` / `NodeBlock` (mrdoob#29495) * VolumeNodeMaterial: simplify a little * cleanup * NodeBuilder: Introduce `addFlowCodeHierarchy()` Examples: Improve shadow map size in `webgpu_tsl_angular_slicing` (mrdoob#29499) WebGPURenderer: respect the `renderer.shadowMap.enabled` property (mrdoob#29492) * enable/disable shadow * enable shadowmaps * enable more examples * and another one * Update Nodes.js --------- Co-authored-by: aardgoose <angus.sawyer@email.com> Updated builds. GLTFLoader: Remove deprecated code. (mrdoob#29502) TiltLoader: Remove loader. (mrdoob#29471) Update Chinese translation of InstancedMesh. (mrdoob#29506) r169
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
Adds the ability to access compute built-ins that are already defined by default directly within TSL. There are a few changes I made that I think might deserve further consideration.
Enabling subgroups by default if the feature is accessible: I largely did this to obviate the need for further logic within ComputeBuiltin that enables subgroups and activates the built-in should it be requested, or creating a separate node specifically for subgroup_size. However, we might not want to enable subgroups by default in every shader.
Creating ComputeBuiltinNode: A lot of the logic is similar to AttributeNode, however, I figured it would be better to separate accessing these builtins from accessing vertex attributes, as they serve different purposes, even if they are both considered 'attributes' within the context of the NodeBuilder.