0 views collected around this technical thread.
The article explains GPU fundamentals—from the end of Dennard scaling and why GPUs excel in parallel throughput, through CUDA programming basics like the SAXPY kernel and SIMT versus SIMD execution, to the evolution of the SIMT stack, modern scheduling, and a three‑step core architecture design.
The article delivers an in‑depth technical dissection of AMD’s FidelityFX Super Resolution 1.0, detailing the EASU spatial upscaling pipeline—its Lanczos2‑based polynomial fitting, 12‑point sampling, gradient calculations, and edge handling—and the RCAS contrast‑adaptive sharpening stage, while also outlining mobile‑friendly optimizations such as half‑precision arithmetic and reduced texture fetches.
This article introduces WebGL, covering its definition, history, basic concepts, working principles, and practical examples of drawing shapes using both native WebGL API and the Three.js framework.
Edge deep learning inference on mobile devices faces hardware and software fragmentation, diverse CPUs, GPUs, DSPs, and NPUs, and limited programmability; optimization techniques such as model selection, quantization, and architecture‑specific tuning enable real‑time performance, with most inference on CPUs, GPUs offering 5–10× speedups, and co‑processor support varying across Android and iOS.