Blog "Chess Engines Diary" - needs a new laptop for chess engine tests

♔ We use for testing only 2 old laptops:
1. Acer - Intel Core i3-3217U ( 5 years old)
2. Lenovo - Intel Core i7-4710MQ is over 3 years old (He works 24 hours a day)
♕♞ A modern laptop that will be designed only for chess engine tests costs (with software) about $ 1000-1200 (4200-4500 PLN). We will no longer invest private money, and chess engine testing costs are big. That is why we are starting the collection for the purchase of a laptop for testing.

♕ Our blog "Chess Engines Diary" is visited by around 2,000-3.000 users daily. If only everyone paid 1 dollar - we will pay the required amount in one day!


Sunday, April 9, 2017

Stockfish 17040802_joergoster - new version!

Stockfish, chess engine UCI
Leader rating list JCER = 3405
🔬 Author: joergoster 


Timestamp: 1491610500 

Fix zugzwang pruning issues 

By adding pos.non_pawn_material(pos.side_to_move()) as a precondition in step 13, which is already in use in Futility Pruning (child node) and Null Move Pruning for similar reasons. 

Pawn endgames, especially those with only 1 or 2 pawns, are simply heavily influenced by zugzwang situations. 

Since we are using a bitbase for KPK endgames, I see no reason to accept buggy evals as shown in #760 

Patch looks neutral at STC 
LLR: 2.32 (-2.94,2.94) [-3.00,1.00] 
Total: 79580 W: 10789 L: 10780 D: 58011 

and LTC 
LLR: 2.95 (-2.94,2.94) [-3.00,1.00] 
Total: 27071 W: 3502 L: 3390 D: 20179 

Bench: 6259071 

Closes #1051 
Closes #760  

⭐⭐⭐   Stockfish 17040802_joergoster - download

No comments:

Post a Comment