Text3D

Use Text3D when lettering must exist inside a 3D scene with real depth, bevels, perspective, lighting response and X/Y/Z placement. This is a different pipeline from Text+: Text+ immediately produces a 2D image with alpha, while Text3D remains scene data until Renderer3D.

Text3D has its own SceneInput. It can accept an existing 3D scene stream and add its generated text to that stream. Complex scenes often use Merge3D for explicit branching, but SceneInput is useful for a simple sequential 3D chain without inventing a 2D composite too early.

Extrusion and Bevel define the text geometry. Extrusion Depth creates side walls; Bevel Depth/Width shape the transition around the glyph edges and become meaningful in combination with actual depth. Extreme values can quickly destroy thin strokes, so establish readable geometry before tuning light and material.

Text3D uses a built-in shading system. It does not expose a normal general-purpose Material input like many other 3D objects. Material Type = Image exposes the conditional 2D ColorImage input, and a separate image-based bevel material can expose BevelTexture. For richer 3D materials, use downstream ReplaceMaterial3D rather than inventing a Text3D material socket.

Keep three transform scopes separate. Layout places and orients the text layout; the internal Transform controls Characters, Words or Lines; the final 3D Transform positions the complete Text3D object in the scene. Repeated labels such as Rotation, Pivot and Size belong to different scopes and are not interchangeable.

A final 2D frame exists only after Renderer3D. Once the scene is rendered, normal 2D tools such as Glow, Color Corrector, Blur, Merge and MediaOut can process the pixels. This keeps the boundary clear: text geometry and lighting before Renderer3D, pixel finishing after it.