NIXIEFX RUNTIME LAB · 0.1.20

Smaller area, explicit emission policy

Compare a fixed rate with an area-scaled rate. The host applies the multiplier; the runtime does not detect the viewport.

Loading runtime…

Reproduce

Choose 800 or 400 pixels, restart, and compare active counts at 120 fixed simulation steps. Width and height halve together. The authored rate is 120/s; lifetime is 0.75 s; seed is 14.

Common mistakes

Halving width does not halve area. Apply the multiplier to the original rate, never to a previously scaled rate. Keep burst feedback separate from ambient coverage.

Counts are real adapter measurements. Update time excludes GPU completion. Camera, browser scheduling, and device affect appearance and speed. This is a Three.js demo; PixiJS visual parity is not claimed. Preview bloom is not exported host postprocessing; PixiJS lit shading and mesh-surface emission are unsupported.

For a web game VFX editor, inspect the runtime and support-report reference alongside the preview.