Nodes · 3D
Render3D
3Dv1core
Overview
Renders a 3D scene into the comp: point scene/camera at your 3D branch and a Camera3D, and the CG composites over the bg input as a premultiplied element. aux adds depth/N/P/id channels.
How to
How-to
- Name the
scenebranch and thecamera; set the format; wire abgto comp over. mode = uvrenders texture space;auxadds data layers;lightgroupssplits the beauty per light.
Inputs
| Input | What it takes |
|---|---|
| bg | a clean background to reveal (optional — without one the node estimates it and says so) |
Outputs
| Output | Type |
|---|---|
| out | Image |
Parameters
| Parameter | Type | Default | What it does |
|---|---|---|---|
| slot | Int | 0 | The internal channel this render publishes on. Auto-assigned — leave it. |
| format_width | Int | 1920 | Render width in pixels. |
| format_height | Int | 1080 | Render height in pixels. |
| scene | Text | The NAME of the 3D node to render — usually your Camera3D or the top Merge3D. Everything upstream of it is in the scene. | |
| camera | Text | The NAME of the Camera3D to render through. Empty = auto-frame the scene. | |
| mode | Choice | camera | camera = render the scene through a camera. uv = UV-stabilize: the tracked model's texture-space view of its plate — the surface frozen, so downstream paint/clone lands on a non-moving target. In uv mode set tracker; scene/camera/bg are ignored. camerauv |
| tracker | Text | uv mode: the NAME of the BS_GeoTracker to stabilize. Its baked track is the pose, its geo input the mesh, its src input the plate — run ▶ Track on it first. | |
| atlas_res | Int | 1024 | uv mode: the square texture-space output resolution. Confidence rides in alpha (facing × visibility × in-frame), exactly like the built texture. |
| aux | Choice | none | Extra channels beside rgba: depth (camera-space distance), or full = depth + N + P + object id. Data-grade — never filtered or colour-managed. nonedepthfull |
| lightgroups | Bool | false | Publish a lightgroup.<light> layer per scene light — each light's isolated contribution, gradeable downstream (regrade the key without touching the fill). The layers sum to the beauty. Costs one extra render per light; capped at 8. |
| samples | Choice | 1 | How many sub-steps the effect integrates over. More is smoother and slower. 14816 |
| disabled | Bool | false | Bypass: the node passes its first input through untouched (D in the graph). The knobs keep their values for when it comes back. |
| label | Text | A note shown under the node in the graph — what this node is FOR. Never affects the render or the cache. | |
| hide_input | Bool | false | Draw the node without its input wire — a tidy graph for a node whose source is obvious (a Dot, a viewer tap). |
See also
- The 3D system — How to put geometry, cameras and volumes in the comp — without becoming a 3D application.
- Unreal — How to get Unreal's pixels into the comp, live or rendered.