Task descriptions and reference implementations for the "Introduction to WebGPU" workshop at CESCG 2023.
Three things:
- Internet browser with WebGPU support
- Local server
- (this) Workshop Git repository
- On Windows, MacOs:
- Download Google Chrome
- Check if Chrome version is already 113 (early roll-out)
- Alternatively: Download Google Chrome Canary
- Enable WebGPU via
chrome://flags(#enable-unsafe-webgpu) - Restart browser
- Enable WebGPU via
- Download Google Chrome
- On Linux:
- Download Chromium
- Option 1: https://download-chromium.appspot.com/
- Option 2: https://github.com/scheib/chromium-latest-linux then update.sh
- On Ubuntu 22.04: Don’t use snap / apt version
- Enable WebGPU (
#enable-unsafe-webgpu) and Vulkan (#enable-vulkan) backend viachrome://flagsthen restart browser - To see if it works, check
chrome://gpuand/or https://toji.github.io/webgpu-test/ - If does not work, use the built-in software renderer SwiftShader via the commandline flag
--use-webgpu-adapter=swiftshader
- Download Chromium
- Our recommendation: Visual Studio Code
- Go to Extensions > Search "Live Server" (over 33 million downloads) > Install
- Alternatively: Install from here
- Alternatively: Have Python installed; we provide a Python server
python3 server.py - Alternatively: Have Node.js installed; we provide a Node.js server
node server.js - Alternatively: Have WebStorm installed; open the HTML file in a browser
git clone https://github.com/UL-FRI-LGM/cescg2023-webgpu
During the four parts of this workshop you're going to solve tasks in order to learn how to use the WebGPU API. For each part, you'll find the task descriptions as well as the respective reference implementation in its respective folder:
You can also check out the reference implementations in the browser. Start a server in the project's root directory and navigate to reference implementations.html.