Re-exports§
pub use handles::*;
Modules§
- animation
- avatar
- bind_
pool - camera
- compute
- cooperative_
matrix - csm
- debug_
vis - dof
- dream_
animations - dream_
procedures - DreamProcedures — procedural generation tools for the Dreamwell render pipeline.
- dream_
surface - DreamSurface — unified material + texture pipeline for PBR.
- dream_
tsr - dreamlet_
catalog - dreammatter
- features
- formats
- frame_
arena - gltf
- gpu_
driven - handles
- Stable, typed handles for GPU resources. All handles are small (u32), Copy, and Hash-safe for use as map keys.
- hiz
- ibl
- lights
- Scene light management – collects light descriptors for GPU upload.
- lod
- materialize
- mesh_
shader - meshlet
- meshlet_
render - micro_
dreamlet - motion_
vectors - observer
- p3g
- particle
- passes
- Fixed render pass chain — no generalized render graph.
Each pass implements
RenderPassNodefor prepare/run lifecycle. - physics
- Dreamwell Physics GPU — compute simulation, rendering, and promotion.
- picking
- pipeline
- pipeline_
cache - post
- Post-processing chain — HDR framebuffer, bloom, tonemapping.
- prelude
- GPU crate prelude – commonly used types for rendering.
- primitives
- procedural
- quantum_
bridge - quantum_
cull - ray_
tracing - render_
scene - Render scene data – GPU-facing presentation state. This is NOT canonical world truth. It is rebuilt each frame from the client mirror via the presentation update phase.
- render_
target - ring
- rt_
denoise - rt_gi
- rt_
shadows - scene
- sdf
- shader
- skybox
- spatial_
event - sprite
- ssao
- ssgi
- ssr
- stores
- Resource stores – handle-based GPU resource management. Each store owns GPU resources and provides insert/get/remove via typed handles.
- surface_
cache - taa
- tangent
- Tangent generation utility — per-triangle accumulation + Gram-Schmidt orthogonalization. Generates tangent vectors (with bitangent sign in .w) for meshes without tangent data.
- text
- texture
- texture_
cache - translucency_
volume - tween
- upload
- volumetric
- vram
Structs§
- GpuRuntime
- GPU runtime — shared device, queue, and capability contexts. Both editor and runtime create one of these at startup.