Chess Engines Diary

The 3rd season of the CEDR chess engine league has begun. The highest amateur group, E-01, was won by the Spaghet MariNNara 1.0 engine, with Tucano 12.00 JA finishing second and Avalanche 2.1.0 t hird. These three engines have advanced to the elite TOP-64 League. Chess Engines Diary Team All CEDR  625.428 games download  (01.04.2025)  💾  70 games from the tournament download 👍 @chessenginesdiary Country - Poland, City -  Malbork   🕓 Time 3'+3"  💻Asus Zenbook, AMD Ryzen 9 6900HX, RAM 32,0 GB 🖬 GUI-Aquarium Pl Engine 1 2 3 4 5 6 7 Score WP Pr 1 Spaghet MariNNara 1.0 = 11 b + 8 w + 19 b = 3 w + 4 b = 2 w = 5 b 5.0 22.5 21.0 2 Tucano 12.00 JA + 15 w = 14 w + 7 b = 4 w = 3 b = 1 b + 13 w 5.0 21.0 20.5 3 Avalanche 2.1.0 + 6 w = 4 w + 14 b = 1 b = 2 w = 5 w = 8 b 4.5 23.5 20.0 4 StockDory Starfish 0.1 + 10 w = 3 b + 9 w = 2 b - 1 w + 13 b = 6 w 4.5 22.5 19.5 5 Patricia ...

The Spaghet MariNNara 1.0 engine, tested for the first time in the league, promises to be very interesting. He won the A-08 group with a very good result. In 2nd place is the Bagatur 5.1a engine and in 3rd place is the Thrawn 2.1 engine. The fourth engine was also promoted to the A-07 group: Javelin 3.0.0 . 👍Promotion:  Spaghet MariNNara 1.0, Bagatur 5.1a, Thrawn 2.1, Javelin 3.0.0 .  👎Relegated: Mcu-max 1.0.5, Phalp 22032400 JA, Alibaba 1.40 JA, Catto 0.5.2 .  All CEDR games  381.115 download    💾  190 games from the tournament download @chessenginesdiary Country - Poland, City -  Malbork   🕓 Time 3'+3"  💻Asus Zenbook, AMD Ryzen 9 6900HX, RAM 32,0 GB 🖬 GUI-Banksia P Engine CEDR Score Chg +/-/= 1: Spaghet MariNNara 1.0 2400 18.5 / 19 +87 (+18 -0 =1) 2: Bagatur 5.1a 3086 18.0 / 19 +11 (...

Spaghet MariNNara 1.0 - c hess engine based on BBC This release marks the inclusion of NNUE, with the architecture being (768 -> 2048)x2 -> 1 and trained on Leela data This release also has improved readability due to my search_stack rewrite Notable search features include: Singular Extensions Multicut Continuation History Capture History Better TT replacement Estimated CCRL Elo: ~3450 (tested against PlentyChess 0.2.1) Features to look forward to in the next major release: Multithreading Output Buckets SIMD NNUE inference The old version of Spaghet fails to make a single draw against this update, so there will be no progtest Spaghet MariNNara 1.0