VIDEOMONTAZHERDAVINCI RESOLVE · VIDEO · AI

FUSION · NODE REFERENCE

Renderer3D

Render a 3D scene into a 2D image

Renderer3D is the scene3d → image boundary: it accepts a 3D scene plus an optional 2D EffectMask, chooses the camera/render path, and produces the beauty image with only the auxiliary render data you actually need.

Fusion 3DIntermediate

Renderer3D

Overview

What Renderer3D does

Everything before Renderer3D can remain scene3d: geometry, cameras, lights, materials, and nested Merge3D hierarchies are not yet a pixel frame. Renderer3D evaluates the scene and creates the Image output that starts the normal 2D compositing path.

The node has two different input domains. The required orange SceneInput accepts the 3D scene stream. The blue EffectMask accepts a normal 2D mask that limits where the Renderer3D result applies; it does not become geometry inside the scene.

Controls define viewpoint and render path: Camera, Eye, Layer Name, Output Type, and Renderer Type. Software Renderer, Hardware Renderer, and Hardware UV Renderer are not interchangeable labels: the UV mode renders surfaces in unwrapped texture space for baking/texture workflows rather than a normal camera beauty frame.

Output Channels can attach utility data to the beauty result: Z, Coverage, Background Color, Normal, Texture Coordinates, Object ID, Material ID, World Position, Vector, and Back Vector. Enable these deliberately because every extra channel has a memory/render cost and should have a downstream purpose.

In current Resolve 21, object/material Cryptomatte layers and metadata are generated through Renderer3D's hardware renderer. Enable Cryptomatte therefore belongs to a hardware-renderer workflow rather than being treated as a renderer-independent switch.

The Image tab controls technical output properties such as Process Mode, resolution, pixel aspect, bit depth, DataWindow/overscan, and common color/gamma handling. Floating-point depth is especially relevant when downstream work depends on Z, World Position, or other technical channels.

When to use it

  • Convert a Fusion 3D scene into a normal 2D image

  • Render Text3D, Shape3D, ImagePlane3D, or imported geometry

  • Evaluate scene lighting and shadows

  • Render selected Z, Normal, World Position, Object ID, Material ID, or motion-vector data

  • Generate depth data for downstream depth-aware effects

  • Generate object/material Cryptomatte through the hardware renderer

  • Render texture-space output with Hardware UV Renderer for baking workflows

  • Match resolution, pixel aspect, bit depth, and overscan to the target pipeline