1
0
Fork 0
mirror of https://github.com/sockspls/badfish synced 2025-05-01 17:19:36 +00:00

Mix alpha and statScore for reduction

Idea by @xoto10, and tuning by @FauziAkram.

Passed STC:
LLR: 2.94 (-2.94,2.94) <0.00,2.00>
Total: 57832 W: 15540 L: 15199 D: 27093
Ptnml(0-2): 207, 6343, 15477, 6680, 209
https://tests.stockfishchess.org/tests/view/6338db6f35f43d649ff60fdc

passed LTC:
LLR: 2.95 (-2.94,2.94) <0.50,2.50>
Total: 50968 W: 13770 L: 13440 D: 23758
Ptnml(0-2): 25, 4905, 15306, 5211, 37
https://tests.stockfishchess.org/tests/view/6339777035f43d649ff62686

Links to the tuning sessions:
https://tests.stockfishchess.org/tests/view/63345725a004bed9a2e47b28
https://tests.stockfishchess.org/tests/view/63345728a004bed9a2e47b2a

closes https://github.com/official-stockfish/Stockfish/pull/4183

Bench: 4426602
This commit is contained in:
FauziAkram 2022-10-03 17:45:05 +03:00 committed by Stéphane Nicolet
parent f436bf77ad
commit 8bab09749d

View file

@ -787,9 +787,9 @@ namespace {
// The depth condition is important for mate finding.
if ( !ss->ttPv
&& depth < 8
&& eval - futility_margin(depth, improving) - (ss-1)->statScore / 303 >= beta
&& eval - futility_margin(depth, improving) - (ss-1)->statScore / 301 >= beta
&& eval >= beta
&& eval < 28031) // larger than VALUE_KNOWN_WIN, but smaller than TB wins
&& eval < 28692) // larger than VALUE_KNOWN_WIN, but smaller than TB wins
return eval;
// Step 9. Null move search with verification search (~22 Elo)
@ -1179,7 +1179,7 @@ moves_loop: // When in check, search starts here
- 4433;
// Decrease/increase reduction for moves with a good/bad history (~30 Elo)
r -= ss->statScore / 13628;
r -= (ss->statScore + 5 * alpha) / 15448;
// In general we want to cap the LMR depth search at newDepth, but when
// reduction is negative, we allow this move a limited search extension