1
0
Fork 0
mirror of https://github.com/sockspls/badfish synced 2025-04-29 08:13:08 +00:00

Remove evaluation grain

Passed non-regression STC:
https://tests.stockfishchess.org/tests/view/66fa345a86d5ee47d953b86e
LLR: 2.96 (-2.94,2.94) <-1.75,0.25>
Total: 39776 W: 10528 L: 10306 D: 18942
Ptnml(0-2): 134, 4674, 10063, 4870, 147

Passed non-regression LTC:
https://tests.stockfishchess.org/tests/view/66facfb886d5ee47d953b8a8
LLR: 2.94 (-2.94,2.94) <-1.75,0.25>
Total: 64230 W: 16484 L: 16305 D: 31441
Ptnml(0-2): 38, 7195, 17483, 7348, 51

closes https://github.com/official-stockfish/Stockfish/pull/5613

bench 1013135
This commit is contained in:
Linmiao Xu 2024-09-29 04:22:37 -04:00 committed by Joost VandeVondele
parent 2b9154882a
commit 0186904f53

View file

@ -83,9 +83,6 @@ Value Eval::evaluate(const Eval::NNUE::Networks& networks,
int material = (smallNet ? 553 : 532) * pos.count<PAWN>() + pos.non_pawn_material();
v = (nnue * (76898 + material) + optimism * (8112 + material)) / (smallNet ? 74411 : 76256);
// Evaluation grain (to get more alpha-beta cuts) with randomization (for robustness)
v = (v / 16) * 16 - 1 + (pos.key() & 0x2);
// Damp down the evaluation linearly when shuffling
v -= v * pos.rule50_count() / 212;