unity-configure-rendering

Unity Configure Rendering

When to use

Configure Built-in, URP or HDRP rendering, camera/Volume behavior, quality tiers or custom renderer features. Use for an explicitly requested pipeline migration; material/shader implementation stays with the materials owner.

Activation and prerequisites

Run commands/resolve-engine.sh and commands/resolve-skill.sh unity-configure-rendering before edits or tool launches. Require effective UNITY compatibility and respect recorded task override scope. A skill request never changes engine selection.

Inspect the actual Editor and pipeline package versions, Graphics/Quality assets and per-tier overrides, selected renderer assets, graphics APIs, color space, cameras, XR and build targets. Resolve supported APIs from the installed package; preview features and packages require explicit selection.

Workflow

1. Trace the effective pipeline for each target quality level. A project-wide default alone does not establish which renderer a player uses. Record current output and budgets before changing settings.
2. Configure camera ownership, projection, clipping, stacking/composition, render targets and Volume masks/weights. Prevent duplicate presentation owners, leaked camera targets and unintended per-camera feature costs.
3. Select quality settings from measured target-device limits: render scale, antialiasing, shadows, lights, depth/opaque textures, HDR and postprocessing. Keep gameplay cues readable across tiers.
4. Implement custom features using the installed pipeline's extension points. In Render Graph, declare access/dependencies and resource lifetime explicitly; do not persist transient handles between frames. Read renderer resources before changing passes.
5. Treat Built-in-to-URP conversion as a separate reviewed migration with an asset inventory, saved originals and comparison scenes. Do not run broad converters incidentally. Use pipeline migration and cameras.

Verification

Compile and build the selected player backend; inspect actual pipeline/renderer selection, shader errors and variant inclusion. Compare camera composition, depth, transparencies, lighting and postprocessing on representative scenes and quality tiers. Exercise resize, camera enable/disable, scene changes and renderer-feature disposal. Measure GPU time and memory on target hardware.

Output

Native-authored pipeline/renderer/Volume assets, camera/resource ownership, measured quality decisions and before/after comparisons. Record environment, expected/observed results and unavailable checks. Null graphics, compilation and headless report success cannot certify rendered appearance or GPU budgets.