HN
Today

Stockfish 19

Stockfish 19, the latest iteration of the world's leading open-source chess engine, has been officially released, boasting significant strength improvements and a host of technical enhancements. This highly anticipated update introduces universal binaries, a refined neural network architecture, and expanded platform support, appealing to the technical and competitive spirit of the Hacker News community. Its continued dominance in engine championships solidifies its reputation as a marvel of computational chess.

12
Score
1
Comments
#7
Highest Rank
1h
on Front Page
First Seen
Sep 10, 9:00 AM
Last Seen
Sep 10, 9:00 AM

The Lowdown

The Stockfish team proudly announced the release of Stockfish 19, marking another significant milestone in the development of the world's strongest open-source chess engine. This new version brings substantial improvements in playing strength, alongside a suite of technical upgrades designed to enhance performance, usability, and platform compatibility.

  • Strength Enhancement: Stockfish 19 demonstrates an Elo gain of up to 44 points over its predecessor, Stockfish 18, and wins over three times as many game pairs, reinforcing its position as the top chess engine.
  • Simplified Deployment: The introduction of universal binaries streamlines the download process, automatically optimizing performance for diverse CPU architectures without manual selection.
  • Advanced NNUE Architecture: The SFNNv16 network features a more efficient design, reducing binary size while boosting strength through new pawn-pair features and the retirement of a less effective secondary neural network. Training methodologies have also been refined with techniques like Quantization-Aware Training and extensive re-scoring using a strong Leela net.
  • Broadened Platform Support: Stockfish 19 extends its reach with native support for RISC-V (RVV) and LoongArch (LSX/LASX), 1GB Linux huge pages, and WebAssembly targets, alongside an overhauled shared-memory implementation across various operating systems.
  • Robust Error Handling: Stricter validation for board positions, FEN strings, and UCI commands ensures greater stability, with critical errors now leading to immediate termination and detailed feedback.

This release underscores the continuous innovation and community-driven development that keeps Stockfish at the forefront of computational chess, providing both stronger play and a more refined user experience for enthusiasts and professionals alike.