# Culling Pipeline

## Current Behavior

Flora currently uses the **Batch Renderer Group (BRG)** culling path.

> The old per-project/per-scene culling pipeline selection fields are obsolete and have no runtime effect.

## What This Affects

* CPU-side draw submission scaling
* Culling behavior per camera/view/light
* Memory tradeoffs for per-instance runtime data

***

## Batch Renderer Group

BRG-compatible shaders usually work out of the box, but you must avoid stripping required BRG variants in builds.

**Keep BRG shader variants**

* **Edit → Project Settings → Graphics**
* Set **BatchRendererGroup variants** to **Keep all**

> See Unity’s [BRG Getting Started Guide](https://docs.unity3d.com/Manual/batch-renderer-group-getting-started.html) for details.

***

### Performance

* BRG reduces CPU overhead in large scenes.
* Benefits are strongest when instance counts/types are high.

***

### Memory

BRG stores additional per-instance matrix-related data to support current Flora runtime behavior. This generally trades more instance-buffer memory for better scalability and feature support.

***

### Known Limitation: Child LOD Renderer Transforms

Child renderers under an `LODGroup` with non-identity transforms do not render using child local transforms in BRG mode.

Mitigation:

* Bake child transforms into meshes (for example via [FBX Exporter](https://docs.unity3d.com/Packages/com.unity.formats.fbx@5.1/manual/index.html)).

***

## Migration Note (Older Flora Docs/Projects)

Older Flora versions documented a selectable RenderMesh pipeline path. In current Flora runtime, this is historical context only and not a selectable runtime path.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://flora.magneticarcade.com/rendering/culling-pipeline.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.