Use classical eval for Bishop vs Pawns

NNUE evaluation is incapable of recognizing trivially drawn bishop endgames
(the wrong-colored rook pawn), which are in fact ubiquitous and stock standard
in chess analysis. Switching off NNUE evaluation in KBPs vs KPs endgames is
a measure that stops Stockfish from trading down to a drawn version of these
endings when we presumably have advantage. The patch is able to edge over master
in endgame positions.

Patch tested for Elo gain with the "endgame.epd" book, and verified for
non-regression with our usual book (see the pull request for details).

STC:
LLR: 2.93 (-2.94,2.94) {-0.20,1.10}
Total: 33232 W: 6655 L: 6497 D: 20080
Ptnml(0-2): 4, 2342, 11769, 2494, 7
https://tests.stockfishchess.org/tests/view/6074a52981417533789605b8

LTC:
LLR: 2.93 (-2.94,2.94) {0.20,0.90}
Total: 159056 W: 29799 L: 29378 D: 99879
Ptnml(0-2): 7, 9004, 61085, 9425, 7
https://tests.stockfishchess.org/tests/view/6074c39a81417533789605ca

Closes https://github.com/official-stockfish/Stockfish/pull/3427

Bench: 4503918

blah

diff --git a/src/evaluate.cpp b/src/evaluate.cpp
index 789e28594cd6d8f2a7c555750b165903088f1afd..ba3de70bb2524387b553f48e01c1f1557f713eae 100644 (file)
--- a/src/evaluate.cpp
+++ b/src/evaluate.cpp
@@ -1104,23 +1104,26 @@ Value Eval::evaluate(const Position& pos) {
          return nnue;
       };
 
-      // If there is PSQ imbalance use classical eval, with small probability if it is small
+      // If there is PSQ imbalance we use the classical eval. We also introduce
+      // a small probability of using the classical eval when PSQ imbalance is small.
       Value psq = Value(abs(eg_value(pos.psq_score())));
       int   r50 = 16 + pos.rule50_count();
       bool  largePsq = psq * 16 > (NNUEThreshold1 + pos.non_pawn_material() / 64) * r50;
       bool  classical = largePsq || (psq > PawnValueMg / 4 && !(pos.this_thread()->nodes & 0xB));
 
       // Use classical evaluation for really low piece endgames.
-      // The most critical case is a bishop + A/H file pawn vs naked king draw.
-      bool strongClassical = pos.non_pawn_material() < 2 * RookValueMg && pos.count<PAWN>() < 2;
+      // One critical case is the draw for bishop + A/H file pawn vs naked king.
+      bool lowPieceEndgame =   pos.non_pawn_material() == BishopValueMg
+                            || (pos.non_pawn_material() < 2 * RookValueMg && pos.count<PAWN>() < 2);
 
-      v = classical || strongClassical ? Evaluation<NO_TRACE>(pos).value() : adjusted_NNUE();
+      v = classical || lowPieceEndgame ? Evaluation<NO_TRACE>(pos).value() 
+                                       : adjusted_NNUE();
 
       // If the classical eval is small and imbalance large, use NNUE nevertheless.
-      // For the case of opposite colored bishops, switch to NNUE eval with
-      // small probability if the classical eval is less than the threshold.
-      if (   largePsq
-          && !strongClassical
+      // For the case of opposite colored bishops, switch to NNUE eval with small
+      // probability if the classical eval is less than the threshold.
+      if (    largePsq
+          && !lowPieceEndgame
           && (   abs(v) * 16 < NNUEThreshold2 * r50
               || (   pos.opposite_bishops()
                   && abs(v) * 16 < (NNUEThreshold1 + pos.non_pawn_material() / 64) * r50