WebGL For More Than Graphics: Is It Possible?
summary
Live demos show WebGL used as a general‑purpose GPU: an edge detection shader running at 60fps, a particle system pushing 300,000 particles, and a 1024×1024 matrix multiply where GPU gives thousands‑fold speedups versus CPU (large CPU run takes minutes). Implementation details include writing GLSL fragment/vertex shaders, moving data via textures and reading back with getImageData, plus limits: old GLSL, no WebCL/compute shaders in browsers, serialization and shader timeouts.
AI-written summary · about
chapters (14)
- 0:00Intro
- 6:24Small Math Operations
- 6:53WebGL
- 8:19Pixels
- 9:33Matrix Multiplication
- 16:53Big O notation
- 18:10Proof of concept
- 18:24Hash collision
- 18:55Signal analysis
- 19:32GLSL
- 20:16Bit Operations
- 20:50Shader Timeouts
- 21:25Shaders
- 21:49Web Worker
related talks
25mArbitrary Computation on the GPU Using WebGL
24mArbitrary Computation on the GPU Using WebGL
23mArbitrary Computation on the GPU Using WebGL
34mNext-Generation 3D Graphics on the Web
25mHow to Program Your GPU (and Not Do It Alone)
38mWebGPU: Next-generation 3D Graphics on the Web
34mWebGL Practical Application
05mA Brief Intro to GPU Shader Programming in WebGL