GPU acceleration of the Game of Life: CUDA Triton benchmarks https://sudonull.com/gpu-acceleration-of-the-game-of-life-cuda-triton-benchmarks
Comparison of PyTorch, CUDA, Triton for the cellular automaton "Game of Life" on A40. Results 22.5 ms, kernel code. Study optimizations for memory-bound tasks. Read the analysis.
GPU Font Rendering Slug: Dynamic Dilation and MIT License https://sudonull.com/gpu-font-rendering-slug-dynamic-dilation-and-mit-license
Learn about the Slug algorithm for GPU font rendering from Bezier curves, dynamic dilation, and opening the patent under MIT. Read the technical analysis.
Deterministic AI on Rust O(1) without GPU https://sudonull.com/deterministic-ai-on-rust-o-1-without-gpu
Development of AI core with agglutinative logic on MacBook Air M2. Determinism, 5.4 million ops/s on CPU, protection against hallucinations. Explore the architecture for middle/senior dev.
Vibe-coding SwiftUI: macOS utilities without knowing the language https://sudonull.com/vibe-coding-swiftui-macos-utilities-without-knowing-the-language
Django co-author created traffic and GPU monitors on SwiftUI using Claude in one evening. Code recombination, verification issues. For developers: prototypes without Xcode.
SIMD optimization of Mandelbrot AVX2 OpenMP CUDA https://sudonull.com/simd-optimization-of-mandelbrot-avx2-openmp-cuda
Boosting the Mandelbrot set: from 7 FPS scalar C++ to 500+ FPS on CUDA. AVX2 intrinsics, OpenMP multithreading, GPU code. Benchmarks Ryzen 5 + RTX 3050. For middle/senior developers.