OffscreenCanvas API in JS for UI Performance
Master the OffscreenCanvas API to offload rendering from the main thread. Covers worker-based 2D and WebGL rendering, animation loops inside workers, bitmap transfer, double buffering, chart rendering pipelines, image processing, and performance measurement strategies.
OffscreenCanvas decouples rendering from the main thread. It provides a canvas that can be used inside Web Workers, preventing heavy draw calls from blocking the UI. Two modes exist: transferControlToOffscreen (worker owns the canvas element) and standalone OffscreenCanvas (worker-only, returns ImageBitmap).
For the fundamentals of Web Workers, see Advanced Web Workers for High Performance JS.
Transfer Control Mode
// === main.js ===
// Transfer an on-screen canvas element to a worker for rendering
const canvas = document.getElementById("renderCanvas");
const offscreen = canvas.transferControlToOffscreen();
const worker = new Worker("render-worker.js");
// Transfer the OffscreenCanvas to the worker (one-time, ownership moves)
worker.postMessage({ canvas: offscreen }, [offscreen]);
// Main thread is completely free for UI interaction
// The worker now owns the canvas and can render at 60fps independently
// Send configuration updates without blocking rendering
worker.postMessage({ type: "config", color: "#ff6600", speed: 2 });
// Resize handling
window.addEventListener("resize", () => {
worker.postMessage({
type: "resize",
width: canvas.clientWidth * devicePixelRatio,
height: canvas.clientHeight * devicePixelRatio
});
});
// === render-worker.js ===
let ctx = null;
let animationId = null;
let config = { color: "#3366ff", speed: 1 };
let width = 0;
let height = 0;
self.onmessage = (event) => {
const data = event.data;
if (data.canvas) {
// Received the transferred OffscreenCanvas
const canvas = data.canvas;
ctx = canvas.getContext("2d");
width = canvas.width;
height = canvas.height;
startRenderLoop();
}
if (data.type === "config") {
config = { ...config, ...data };
}
if (data.type === "resize") {
width = data.width;
height = data.height;
}
};
function startRenderLoop() {
let frame = 0;
function render() {
frame++;
// Clear
ctx.clearRect(0, 0, width, height);
// Draw animated content (runs on worker thread, never blocks UI)
const centerX = width / 2;
const centerY = height / 2;
const radius = Math.min(width, height) * 0.3;
const particleCount = 200;
for (let i = 0; i < particleCount; i++) {
const angle = (i / particleCount) * Math.PI * 2 + frame * 0.01 * config.speed;
const r = radius * (0.5 + 0.5 * Math.sin(i * 0.1 + frame * 0.02));
const x = centerX + Math.cos(angle) * r;
const y = centerY + Math.sin(angle) * r;
const size = 2 + Math.sin(i + frame * 0.05) * 2;
ctx.fillStyle = config.color;
ctx.globalAlpha = 0.5 + 0.5 * Math.sin(i * 0.3 + frame * 0.03);
ctx.beginPath();
ctx.arc(x, y, size, 0, Math.PI * 2);
ctx.fill();
}
ctx.globalAlpha = 1;
requestAnimationFrame(render); // requestAnimationFrame works in workers
}
render();
}Standalone OffscreenCanvas (Bitmap Mode)
// === main.js ===
// Create OffscreenCanvas independently (not tied to a DOM canvas)
// Worker renders to it, then transfers ImageBitmap to main thread for display
const displayCanvas = document.getElementById("display");
const displayCtx = displayCanvas.getContext("bitmaprenderer");
const worker = new Worker("bitmap-worker.js");
worker.onmessage = (event) => {
if (event.data.bitmap) {
// Transfer the rendered bitmap to the visible canvas
displayCtx.transferFromImageBitmap(event.data.bitmap);
// The bitmap is consumed; no need to close it
}
};
worker.postMessage({
type: "start",
width: displayCanvas.width,
height: displayCanvas.height
});
// === bitmap-worker.js ===
let offscreen = null;
let ctx = null;
self.onmessage = (event) => {
if (event.data.type === "start") {
const { width, height } = event.data;
// Create standalone OffscreenCanvas (no DOM element)
offscreen = new OffscreenCanvas(width, height);
ctx = offscreen.getContext("2d");
renderFrame();
}
};
function renderFrame() {
const w = offscreen.width;
const h = offscreen.height;
// Draw content on the worker thread
ctx.fillStyle = "#1a1a2e";
ctx.fillRect(0, 0, w, h);
// Draw a grid of cells
const cellSize = 10;
const cols = Math.floor(w / cellSize);
const rows = Math.floor(h / cellSize);
for (let r = 0; r < rows; r++) {
for (let c = 0; c < cols; c++) {
const alive = Math.random() > 0.7;
if (alive) {
ctx.fillStyle = `hsl(${(r + c) * 3}, 70%, 50%)`;
ctx.fillRect(c * cellSize, r * cellSize, cellSize - 1, cellSize - 1);
}
}
}
// Convert to ImageBitmap and transfer to main thread
const bitmap = offscreen.transferToImageBitmap();
self.postMessage({ bitmap }, [bitmap]); // Transfer (zero-copy)
// Schedule next frame
setTimeout(renderFrame, 1000 / 30); // 30fps
}
// ALTERNATIVE: use createImageBitmap for async conversion
// async function renderFrameAsync() {
// // ... draw to ctx ...
// const bitmap = await createImageBitmap(offscreen);
// self.postMessage({ bitmap }, [bitmap]);
// }WebGL in a Worker
// === webgl-worker.js ===
let gl = null;
let program = null;
let positionBuffer = null;
let frame = 0;
self.onmessage = (event) => {
if (event.data.canvas) {
const canvas = event.data.canvas;
gl = canvas.getContext("webgl2");
if (!gl) {
self.postMessage({ error: "WebGL2 not supported in worker" });
return;
}
initWebGL();
renderLoop();
}
};
function initWebGL() {
// Vertex shader
const vsSource = `#version 300 es
in vec2 aPosition;
uniform float uTime;
out vec2 vPos;
void main() {
float wave = sin(aPosition.x * 3.0 + uTime) * 0.1;
gl_Position = vec4(aPosition.x, aPosition.y + wave, 0.0, 1.0);
vPos = aPosition;
}
`;
// Fragment shader
const fsSource = `#version 300 es
precision mediump float;
in vec2 vPos;
uniform float uTime;
out vec4 fragColor;
void main() {
float r = 0.5 + 0.5 * sin(vPos.x * 5.0 + uTime);
float g = 0.5 + 0.5 * cos(vPos.y * 5.0 + uTime * 0.7);
float b = 0.5 + 0.5 * sin(uTime * 0.5);
fragColor = vec4(r, g, b, 1.0);
}
`;
// Compile shaders
const vs = compileShader(gl.VERTEX_SHADER, vsSource);
const fs = compileShader(gl.FRAGMENT_SHADER, fsSource);
program = gl.createProgram();
gl.attachShader(program, vs);
gl.attachShader(program, fs);
gl.linkProgram(program);
if (!gl.getProgramParameter(program, gl.LINK_STATUS)) {
self.postMessage({ error: gl.getProgramInfoLog(program) });
return;
}
// Create geometry (triangle strip quad)
const positions = new Float32Array([
-1, -1, 1, -1, -1, 1, 1, 1
]);
positionBuffer = gl.createBuffer();
gl.bindBuffer(gl.ARRAY_BUFFER, positionBuffer);
gl.bufferData(gl.ARRAY_BUFFER, positions, gl.STATIC_DRAW);
}
function compileShader(type, source) {
const shader = gl.createShader(type);
gl.shaderSource(shader, source);
gl.compileShader(shader);
if (!gl.getShaderParameter(shader, gl.COMPILE_STATUS)) {
const info = gl.getShaderInfoLog(shader);
gl.deleteShader(shader);
throw new Error(`Shader compile error: ${info}`);
}
return shader;
}
function renderLoop() {
frame++;
gl.viewport(0, 0, gl.canvas.width, gl.canvas.height);
gl.clearColor(0.05, 0.05, 0.15, 1.0);
gl.clear(gl.COLOR_BUFFER_BIT);
gl.useProgram(program);
// Set time uniform
const timeLoc = gl.getUniformLocation(program, "uTime");
gl.uniform1f(timeLoc, frame * 0.02);
// Bind position attribute
const posLoc = gl.getAttribLocation(program, "aPosition");
gl.bindBuffer(gl.ARRAY_BUFFER, positionBuffer);
gl.enableVertexAttribArray(posLoc);
gl.vertexAttribPointer(posLoc, 2, gl.FLOAT, false, 0, 0);
// Draw
gl.drawArrays(gl.TRIANGLE_STRIP, 0, 4);
requestAnimationFrame(renderLoop);
}Double Buffering Pattern
// Use two OffscreenCanvases: one renders while the other displays
class DoubleBufferedRenderer {
#frontBuffer;
#backBuffer;
#frontCtx;
#backCtx;
#width;
#height;
#rendering = false;
constructor(width, height) {
this.#width = width;
this.#height = height;
// Two off-screen buffers
this.#frontBuffer = new OffscreenCanvas(width, height);
this.#backBuffer = new OffscreenCanvas(width, height);
this.#frontCtx = this.#frontBuffer.getContext("2d");
this.#backCtx = this.#backBuffer.getContext("2d");
}
// Draw to back buffer (not yet visible)
renderToBackBuffer(drawFn) {
this.#backCtx.clearRect(0, 0, this.#width, this.#height);
drawFn(this.#backCtx, this.#width, this.#height);
}
// Swap front and back buffers
swap() {
[this.#frontBuffer, this.#backBuffer] = [this.#backBuffer, this.#frontBuffer];
[this.#frontCtx, this.#backCtx] = [this.#backCtx, this.#frontCtx];
}
// Get the front buffer as ImageBitmap for display
getDisplayBitmap() {
return this.#frontBuffer.transferToImageBitmap();
}
// Render frame with automatic swap
renderFrame(drawFn) {
this.renderToBackBuffer(drawFn);
this.swap();
return this.getDisplayBitmap();
}
}
// USAGE IN WORKER
// const renderer = new DoubleBufferedRenderer(800, 600);
//
// function animate() {
// const bitmap = renderer.renderFrame((ctx, w, h) => {
// // Complex drawing that takes variable time
// ctx.fillStyle = "#222";
// ctx.fillRect(0, 0, w, h);
//
// for (let i = 0; i < 1000; i++) {
// ctx.fillStyle = `hsl(${i * 0.36}, 80%, 60%)`;
// ctx.fillRect(
// Math.random() * w,
// Math.random() * h,
// 5, 5
// );
// }
// });
//
// self.postMessage({ bitmap }, [bitmap]);
// requestAnimationFrame(animate);
// }
// animate();Image Processing Pipeline
// Process images entirely off the main thread
// === image-processor-worker.js ===
const filters = {
grayscale(pixels) {
const data = pixels.data;
for (let i = 0; i < data.length; i += 4) {
const avg = data[i] * 0.299 + data[i + 1] * 0.587 + data[i + 2] * 0.114;
data[i] = data[i + 1] = data[i + 2] = avg;
}
return pixels;
},
sepia(pixels) {
const data = pixels.data;
for (let i = 0; i < data.length; i += 4) {
const r = data[i], g = data[i + 1], b = data[i + 2];
data[i] = Math.min(255, r * 0.393 + g * 0.769 + b * 0.189);
data[i + 1] = Math.min(255, r * 0.349 + g * 0.686 + b * 0.168);
data[i + 2] = Math.min(255, r * 0.272 + g * 0.534 + b * 0.131);
}
return pixels;
},
blur(pixels, width, height, radius = 3) {
const data = pixels.data;
const copy = new Uint8ClampedArray(data);
const size = radius * 2 + 1;
const area = size * size;
for (let y = radius; y < height - radius; y++) {
for (let x = radius; x < width - radius; x++) {
let r = 0, g = 0, b = 0;
for (let dy = -radius; dy <= radius; dy++) {
for (let dx = -radius; dx <= radius; dx++) {
const idx = ((y + dy) * width + (x + dx)) * 4;
r += copy[idx];
g += copy[idx + 1];
b += copy[idx + 2];
}
}
const idx = (y * width + x) * 4;
data[idx] = r / area;
data[idx + 1] = g / area;
data[idx + 2] = b / area;
}
}
return pixels;
},
sharpen(pixels, width, height) {
const data = pixels.data;
const copy = new Uint8ClampedArray(data);
// Sharpening kernel: [0,-1,0,-1,5,-1,0,-1,0]
const kernel = [0, -1, 0, -1, 5, -1, 0, -1, 0];
for (let y = 1; y < height - 1; y++) {
for (let x = 1; x < width - 1; x++) {
for (let c = 0; c < 3; c++) {
let val = 0;
for (let ky = -1; ky <= 1; ky++) {
for (let kx = -1; kx <= 1; kx++) {
const idx = ((y + ky) * width + (x + kx)) * 4 + c;
val += copy[idx] * kernel[(ky + 1) * 3 + (kx + 1)];
}
}
data[(y * width + x) * 4 + c] = Math.max(0, Math.min(255, val));
}
}
}
return pixels;
}
};
self.onmessage = async (event) => {
const { type, bitmap, pipeline, width, height } = event.data;
if (type === "process") {
// Create OffscreenCanvas from ImageBitmap
const canvas = new OffscreenCanvas(width, height);
const ctx = canvas.getContext("2d");
ctx.drawImage(bitmap, 0, 0);
bitmap.close(); // Release the source bitmap
let imageData = ctx.getImageData(0, 0, width, height);
// Apply filter pipeline sequentially
for (const filterName of pipeline) {
if (filters[filterName]) {
imageData = filters[filterName](imageData, width, height);
}
}
ctx.putImageData(imageData, 0, 0);
// Return processed result as ImageBitmap
const result = canvas.transferToImageBitmap();
self.postMessage({ result }, [result]);
}
};
// === main.js ===
// async function processImage(imgElement, filterPipeline) {
// const bitmap = await createImageBitmap(imgElement);
//
// return new Promise((resolve) => {
// const worker = new Worker("image-processor-worker.js");
//
// worker.onmessage = (event) => {
// const resultCanvas = document.getElementById("result");
// const ctx = resultCanvas.getContext("bitmaprenderer");
// ctx.transferFromImageBitmap(event.data.result);
// worker.terminate();
// resolve();
// };
//
// worker.postMessage({
// type: "process",
// bitmap,
// pipeline: filterPipeline,
// width: imgElement.naturalWidth,
// height: imgElement.naturalHeight
// }, [bitmap]);
// });
// }
//
// // Apply grayscale + sharpen pipeline
// const img = document.querySelector("img");
// await processImage(img, ["grayscale", "sharpen"]);| Feature | Main Thread Canvas | OffscreenCanvas (Transfer) | OffscreenCanvas (Bitmap) |
|---|---|---|---|
| Where it renders | Main thread | Worker thread | Worker thread |
| DOM access | Yes | No | No |
| Blocks UI | Yes | No | No |
| requestAnimationFrame | Yes | Yes | No (use setTimeout) |
| 2D Context | Yes | Yes | Yes |
| WebGL/WebGL2 | Yes | Yes | Yes |
| Browser support | Universal | Chrome 69+, Firefox 105+ | Chrome 69+, Firefox 105+ |
| Use case | Simple UI | Continuous animation | On-demand rendering |
| Optimization | Impact | When to Use |
|---|---|---|
| transferControlToOffscreen | Frees main thread completely | Continuous animations, games |
| transferToImageBitmap | Zero-copy frame delivery | On-demand chart/image rendering |
| Double buffering | Smooth frame presentation | Complex scenes with variable render time |
| Worker pools + OffscreenCanvas | Parallel rendering | Multiple independent canvases |
| SharedArrayBuffer for vertex data | Avoid cloning geometry | Scenes with shared geometry across workers |
Rune AI
Key Insights
- OffscreenCanvas provides two modes: transfer mode moves an on-screen canvas to a worker, while standalone mode creates a worker-only canvas that returns ImageBitmap: Transfer mode is ideal for continuous animations; standalone mode suits on-demand rendering
- WebGL and 2D contexts both work inside Web Workers when used with OffscreenCanvas: Complex shaders, particle systems, and data visualizations can run at full frame rate without blocking the UI thread
- Double buffering with two OffscreenCanvases prevents visual tearing by rendering to one buffer while displaying the other: Swap the buffers after each frame to present the completed render
- Image processing pipelines benefit greatly from worker-based OffscreenCanvas by applying filters (blur, sharpen, grayscale) without freezing the interface: Process the image as ImageData, apply convolution kernels, and return the result as an ImageBitmap
- requestAnimationFrame works in workers with transferred canvases, providing vsync-aligned rendering independent of the main thread's workload: For standalone canvases without a DOM element, use setTimeout-based frame scheduling instead
Frequently Asked Questions
What is the difference between transferControlToOffscreen and standalone OffscreenCanvas?
Does requestAnimationFrame work inside Web Workers?
How do I handle canvas resizing with OffscreenCanvas in a worker?
Can I use OffscreenCanvas with React or other UI frameworks?
Conclusion
OffscreenCanvas moves rendering off the main thread entirely, ensuring smooth UI interactions even during heavy draw operations. Combined with Web Workers and transferable objects, it enables high-performance visualization pipelines. For the Web Worker fundamentals behind these patterns, revisit Advanced Web Workers for High Performance JS. For reactive state management in UI frameworks, see Creating Advanced UI Frameworks in JavaScript.
More in this topic
Advanced Web Workers for High Performance JS
Master Web Workers for truly parallel JavaScript execution. Covers dedicated and shared workers, structured cloning, transferable objects, SharedArrayBuffer with Atomics, worker pools, task scheduling, Comlink RPC patterns, module workers, and performance profiling strategies.
JavaScript Macros and Abstract Code Generation
Master JavaScript code generation techniques for compile-time and runtime metaprogramming. Covers AST manipulation, Babel plugin authorship, tagged template literals as macros, code generation pipelines, source-to-source transformation, compile-time evaluation, and safe eval alternatives.
Creating Advanced UI Frameworks in JavaScript
Build a modern UI framework from scratch in JavaScript. Covers virtual DOM implementation, diff algorithms, reactive state management, component lifecycle, template compilation, event delegation, batched rendering, hooks system, server-side rendering, and hydration.