Limitations

Current limitations:

• Forge material functions are interpreted, not VM/JIT compiled
• subsurface is not yet rendered as true subsurface transport
• the ray renderer still contains backend-specific recursion logic
• the path tracer is the more correct target for custom BSDF work
• the language is still evolving and should not be treated as stable yet

Near-term priorities:

• stabilize the Forge material runtime contract
• improve renderer correctness around custom BSDFs
• document the language and material system as they evolve
• move hot Forge shading code toward a VM and then JIT