nova-easing

A collection of generic easing functions, supporting portable SIMD.

This crate provides a variety of easing functions that can be applied to scalar types (f32, f64) and, with the nightly feature, to SIMD vectors from std::simd.

Compare easings.net

Usage

Add this to your Cargo.toml:

[dependencies]
nova-easing = "0.1.0"

Scalar Easing

The easing functions can be used on f32 and f64 values.

use nova_easing::EasingArgument;

fn main() {
    let x = 0.5f32;
    let eased = x.ease_in_out_sine();
    println!("Eased value: {}", eased); // Eased value: 0.5
}

SIMD Easing (Nightly Only)

To use easing functions with std::simd types, you need to enable the nightly feature in your Cargo.toml and use a nightly Rust toolchain.

[dependencies]
nova-easing = { version = "0.1.0", features = ["nightly"] }

Then you can apply easing functions to SIMD vectors:

#![feature(portable_simd)]

use nova_easing::EasingArgument;
use std::simd::f32x4;

fn main() {
    let values = f32x4::from_array([0.1, 0.3, 0.5, 0.8]);
    let eased_values = values.ease_in_quad();
    println!("Eased values: {:?}", eased_values.as_array());
}

API Overview

The crate provides easing functions for f32, f64, and SIMD types (f32x4, f64x4, etc. with nightly feature).

All easing functions follow the pattern ease_{in|out|in_out}_{type}, where type is one of: quad, cubic, quart, quint, sine, circ, back, bounce, expo, elastic.

Available Functions

• ease_in_quad, ease_out_quad, ease_in_out_quad
• ease_in_cubic, ease_out_cubic, ease_in_out_cubic
• ease_in_quart, ease_out_quart, ease_in_out_quart
• ease_in_quint, ease_out_quint, ease_in_out_quint
• ease_in_sine, ease_out_sine, ease_in_out_sine
• ease_in_circ, ease_out_circ, ease_in_out_circ
• ease_in_back, ease_out_back, ease_in_out_back
• ease_in_bounce, ease_out_bounce, ease_in_out_bounce
• ease_in_expo, ease_out_expo, ease_in_out_expo
• ease_in_elastic, ease_out_elastic, ease_in_out_elastic

For visual plots of each function, see easings.net.

Performance Notes

• SIMD versions leverage hardware vectorization for significant performance gains when processing multiple values simultaneously.
• Scalar versions are optimized for single-value operations.
• All implementations are branchless where possible for better performance.

Building the Demo

The included demo binary generates visual plots for all easing functions. To build and run it, you will need a nightly Rust toolchain and enable the nightly and demo features:

rustup default nightly
cargo run --bin demo --features nightly,demo

This will create PNG plots in the demo_plots/ directory, with subdirectories for f32 and f32x4 variants.

Note: The demo feature is optional and includes the plotters dependency for generating plots. It is not required for using the easing functions.

Running Benchmarks

To run performance benchmarks for all easing functions:

rustup default nightly
cargo bench --features nightly

This will run benchmarks for all 180 functions (30 easing functions × 6 types: f32, f64, f32x4, f32x8, f64x2, f64x4) and generate HTML reports in the target/criterion/ directory, providing detailed performance comparisons.

License

This project is licensed under the MIT License.