FastNoise2 for Unity (UPM Package)

FastNoise2 by Auburn packaged for Unity Package Manager (UPM). This package provides efficient, high-performance noise generation capabilities directly integrated into Unity, leveraging native textures, Burst, Jobs, and IL2CPP compatibility.

---

Features

• Native Texture Support:

  • NativeTexture2D<T> and NativeTexture3D<T>: Efficient native texture wrappers for 2D and 3D textures
  • Zero-copy operations with Unity textures for maximum performance
  • Support for multiple texture formats (byte2/3/4, ushort2/3/4, float)
  • Value bounds and normalization utilities

• Comprehensive Sampling Extensions:

  • Bilinear sampling for smooth interpolation
  • Read pixel functionality for precise texture access
  • Normalized sampling with configurable value ranges

• Job System Integration:

  • Full Burst/Jobs/IL2CPP compatibility
  • Specialized noise generation jobs:
    • Uniform grid generation (2D/3D)
    • Tileable noise generation
    • Texture normalization jobs

• Authoring Tools:

  • BakedNoiseTextureAsset for creating and configuring baked noise textures
  • FastNoiseGraph for storing and serializing noise node configurations
  • Integration with FastNoise Tool for visual noise design

• Editor Extensions:

  • Property drawers for noise graph assets
  • Menu items for launching the FastNoise Tool
  • Custom editors for noise assets

• Cross-Platform Native Support:

  • Optimized native libraries for Windows, macOS, and Linux
  • Universal architecture support (x64, arm64)

---

Platform Support

• ✅ Linux x64, Android x64
• ✅ macOS Universal, macOS Intel
• ✅ Windows x64

---

Getting Started

Installation via Unity Package Manager (UPM)

1. Open Unity and navigate to Window → Package Manager.
2. Click the + button and select Add package from git URL....
3. Paste the following URL and click Add:

https://github.com/api-haus/com.auburn.fastnoise2.git

Unity will automatically download and install the package.

---

Usage Examples

Using Jobs for Noise Generation

public class NoiseJobsExample : MonoBehaviour
{
    void Start()
    {
			using FastNoise nodeTree = FastNoise.FromEncodedNodeTree("DQAFAAAAAAAAQAgAAAAAAD8AAAAAAA==");

			Texture2D texture = new Texture2D(512, 512, TextureFormat.RFloat, false);
			NativeTexture2D<float> noiseTexture2D = new NativeTexture2D<float>(512, Allocator.TempJob);

			// Create a bounds reference for tracking min/max values
			using NativeReference<ValueBounds> boundsRef = new NativeReference<ValueBounds>(Allocator.Temp);

			// Generate noise directly into the native texture with built-in bounds tracking
			nodeTree.GenUniformGrid2D(
				noiseTexture2D,
				boundsRef,
				0, 0,
				noiseTexture2D.Width, noiseTexture2D.Height,
				0.02f, 1337);

			// Log the bounds for verification
			Debug.Log($"Noise bounds: Min={boundsRef.Value.Min}, Max={boundsRef.Value.Max}");

			noiseTexture2D.ApplyTo(texture);

			File.WriteAllBytes("texNative.png", texture.EncodeToPNG());
			Object.DestroyImmediate(texture);
    }
}

---

Important Notes for macOS and Linux Users

Handling Unsigned Libraries and Binaries

On macOS and Linux, you may encounter issues with unsigned native libraries. To resolve these issues, you can use the following commands:

macOS:

chmod +x path/to/library
xattr -dr com.apple.quarantine path/to/library

Linux:

chmod +x path/to/library

Replace path/to/library with the actual path to the native library file.

⚠️ Disclaimer

Perform these actions at your own risk. Modifying file permissions and removing security attributes can expose your system to potential security risks. Ensure you trust the source of the libraries before proceeding. There is always a risk of repository being overtaken by malicious actor.

---

Roadmap / TODO

• Update to NewFastSIMD branch
• Complete Noise Editor integration (serialization to/from Noise Tool)
• Support for all texture dimensions (4D) and job types
• Comprehensive documentation and additional examples
• macOS CodeSigning
• Continuous Integration (CI), Semantic Versioning (SemVer), and OpenUPM support

---

Contributing

Contributions are welcome! Please submit pull requests or open issues on the GitHub repository.

---

License

This project is licensed under the MIT License. See the LICENSE file for details.