# Traffic DEFAULT shader preview · r454

A host Chrome/WebGL2 comparison tested the extracted GLOBALB `DEFAULT` effect
inputs on the ambient traffic `paint_body` material. This is an opt-in viewer
diagnostic: `?debug&trafficModel=PARCELVAN&trafficShader=default-state`.

Both screenshots use the same Route 736 scene, traffic spawn pose, camera target,
viewport (960×540), and packaged city assets. The baseline uses the viewer's
normal fallback; the diagnostic feeds `DiffuseMin/Range`, `SpecularMin/Range`,
`EnvmapMin/Range`, `SpecularPower`, and `EnvmapPower` from the installed
`DEFAULT` payload to the decompiled shader branch. It also projects the nearest
authored type-2 lights into the provisional harmonic packet.

The comparison shows no useful recovery of traffic body appearance. The van
remains an almost featureless white box; the experimental state only changes
its shading slightly. The standalone class table says the random class-0x3f
selector cannot choose from the installed class-0x30 paint catalog, and this
experiment does not recover the separate explicit descriptor's missing state,
material flags, paint mask, or draw-time bindings. The normal renderer is
unchanged. The diagnostic remains opt-in and is not a visual fix.

- Baseline: [609-r454-traffic-default-state-baseline.png](../webgl-city/review/images/609-r454-traffic-default-state-baseline.png)
- Experiment: [610-r454-traffic-default-state-experiment.png](../webgl-city/review/images/610-r454-traffic-default-state-experiment.png)
- Selector evidence: [ambient-traffic-paint-selection-r453.md](ambient-traffic-paint-selection-r453.md)