Skip to content

Latest commit

 

History

History

Rust

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
 
 
 
 
 
 
 
 

GitHub crates.io docs.rs

FastNoise Lite

FastNoise Lite is an extremely portable open source noise generation library with a large selection of noise algorithms. This library focuses on high performance while avoiding platform/language specific features, allowing for easy ports to as many possible languages.

Features

  • 2D & 3D sampling
  • OpenSimplex2 noise
  • OpenSimplex2S noise
  • Cellular (Voronoi) noise
  • Perlin noise
  • Value noise
  • Value Cubic noise
  • OpenSimplex2-based domain warp
  • Basic Grid Gradient domain warp
  • Multiple fractal options for all of the above
  • Supports f32 or f64 precision floats for X/Y sampling positions (see "Feature Flags" below)
  • no_std (see "Feature Flags" below)

Feature Flags

Optionally enable these in your Cargo.toml file with your fastnoise-lite dependency.

  • "f64": Uses f64, instead of the default f32, X/Y coordinate sampling inputs.
  • "std": Uses Rust's standard library for floating point operations (sqrt(), trunc(), and abs()). Either this or "libm" must be enabled. This is enabled by default if no feature flags are specified (but note that specifying "f64" will mean this is no longer default, and must be specified too).
  • "libm": Enables no_std support. Either this or "std" must be enabled. Uses the libm optional dependency (through the num-traits optional dependency) to allow floating point operations (sqrt(), trunc(), and abs()) without relying on the Rust standard library's implementations. FNL is dependency-free except when using this feature flag, which pulls in these optional dependencies.

Getting Started

Additional documentation is available in the project's Getting Started and Documentation pages from its GitHub wiki.

Below is an example for creating a 128x128 array of OpenSimplex2 noise.

use fastnoise_lite::*;

// Create and configure the FastNoise object
let mut noise = FastNoiseLite::new();
noise.set_noise_type(Some(NoiseType::OpenSimplex2));

const WIDTH: usize = 128;
const HEIGHT: usize = 128;
let mut noise_data = [[0.; HEIGHT]; WIDTH];

// Sample noise pixels
for x in 0..WIDTH {
    for y in 0..HEIGHT {
        // Domain warp can optionally be employed to transform the coordinates before sampling:
        // let (x, y) = noise.domain_warp_2d(x as f32, y as f32);
        
        let negative_1_to_1 = noise.get_noise_2d(x as f32, y as f32);
        // You may want to remap the -1..1 range data to the 0..1 range:
        noise_data[x][y] = (neg_1_to_1 + 1.) / 2.;
        
        // (Uses of `as f32` above should become `as f64` if you're using FNL with the "f64" feature flag)
    }
}

// Do something with this data...