THREE.JS / FOUNDATIONS 125

Six views, one target.

A cube render target stores six camera views in one cube texture. This lab updates the six faces with CubeCamera, uses the result for a live reflection, and reads individual faces with faceIndex.

Initializing -

Creating a six-face cube render target.

backend: -
target: -
target class: -
texture: -
active face: -

Face topologysame target, six views

renderer target selected face preview mode

animate environment

Cube target policywaiting

size: -
faces: 6
color space: -
mipmaps: -

CubeCamera writes one scene from six coordinate-system-aware directions.

Backend capability snapshot pending.

0 / +Xunread

center pixel: -

readbacks: 0
bytes: -

1 / -Xunread

center pixel: -

readbacks: 0
bytes: -

2 / +Yunread

center pixel: -

readbacks: 0
bytes: -

3 / -Yunread

center pixel: -

readbacks: 0
bytes: -

4 / +Zunread

center pixel: -

readbacks: 0
bytes: -

5 / -Zunread

center pixel: -

readbacks: 0
bytes: -

Update and readbackwaiting

CPU frames: -
CPU avg: -
CPU p95: -
face writes: -
readbacks: 0
last read: -

Each CubeCamera update writes six faces; face readback is a separate explicit address.

scene -> CubeCamera -> cube render target[face 0..5] -> envMap / faceIndex readback

The face order is 0/+X, 1/-X, 2/+Y, 3/-Y, 4/+Z, 5/-Z. CubeCamera updates its coordinate-system-aware cameras before writing those six target faces. Hide the reflective object during the capture to avoid feedback.

Compare array-layer readback next →

LAB 125 / CUBE FACE TARGET TOPOLOGY