Another version of a very powerful chess engine has appeared, which wants to beat Stockfish. We are very happy about this - we will do the first tests of this version of the engine for you soon.
Caissa - UCI chess engine, Author: Michaล Witanowski
Rating Chess Engines Diary CEDR=3711
Caissa 1.23 - what's new?
A smaller, focused update. At this level gains are hard to come by and my dev time is limited, but 1.23 still squeezes out steady Elo at both STC and LTC, plus a big leap in a quirky “queens-only” test.
Changes
Evaluation
New neural net.
Simplify Evaluate: leaner piece/phase accounting; removed the old castling-rights bonus.
Search
Simplify extensions.
Fix SkipQuiets so quiets are truly skipped.
Simplify TT alpha cutoffs.
Use uint8 for TT entry depth (avoids signed wrap; widens depth range) and adjust a static-eval TT write.
Tweak correction history bonus.
QSearch: adjust best score toward beta on stand-pat exceed (idea from Stockfish).
Perform Null Move Pruning only in cut nodes.
Tuning
Retuned all parameters at long time control (80+0.8).
Progression test
LTC
Elo | 15.85 +- 3.27 (95%)
Conf | 60.0+0.60s Threads=1 Hash=128MB
Games | N: 10000 W: 2532 L: 2076 D: 5392
Penta | [2, 923, 2704, 1359, 12]
STC
Elo | 7.92 +- 3.56 (95%)
Conf | 10.0+0.10s Threads=1 Hash=16MB
Games | N: 10006 W: 2466 L: 2238 D: 5302
Penta | [30, 1126, 2475, 1330, 42]
Special sauce: Queens-only test
Using more high-quality queens-only games in training continues to pay off.
Score of Caissa 1.23 BMI2 vs Caissa 1.22 BMI2: 295 - 19 - 463 [0.678] 777
... Caissa 1.23 BMI2 playing White: 159 - 11 - 218 [0.691] 388
... Caissa 1.23 BMI2 playing Black: 136 - 8 - 245 [0.665] 389
... White vs Black: 167 - 147 - 463 [0.513] 777
Elo difference: 129.0 +/- 14.7, LOS: 100.0 %, DrawRatio: 59.6 %
Thanks to everyone running games and sharing feedback!
Caissa 1.22.1 vs other engines:
| Obsidian dev-16.14 | 10/20 | +0 | 20 Games |
| PlentyChess 6.0.5 JA | 10.5/18 | +3 | 18 Games |
| Berserk 20250622 | 9.5/18 | +1 | 18 Games |
| RubiChess 20250606 JA | 9/18 | +0 | 18 Games |
| Integral 7.0.0 JA | 7/14 | +0 | 14 Games |
| Viridithas 18.0.0 JA | 6.5/13 | +0 | 13 Games |
| SF-PRO2 02.05.2025 | 6.5/13 | +0 | 13 Games |
| Motor 0.9.0 | 7.5/12 | +3 | 12 Games |
| Stockfish 17.1 | 6/12 | +0 | 12 Games |
| Horsie 1.1.0 | 5.5/11 | +0 | 11 Games |
| Alexandria 8.0.0 JA | 5.5/11 | +0 | 11 Games |
| Stormphrax 7.0.0 JA | 5/10 | +0 | 10 Games |
| Dragon 3.3 | 5/10 | +0 | 10 Games |
| Texel 1.13a4 JA | 7.5/9 | +6 | 9 Games |
| Starzix 6.1 JA | 5.5/9 | +2 | 9 Games |
| Yuliana 7.0 | 4.5/9 | +0 | 9 Games |
| Obsidian dev-16.13 | 4.5/9 | +0 | 9 Games |
| Uralochka 3.42 dev14 | 5/8 | +2 | 8 Games |
| Titan 1.1.0 | 5/7 | +3 | 7 Games |
| Clarity 8.0.0 JA | 4.5/7 | +2 | 7 Games |
Comments
Post a Comment