Open dav1312 opened 2 years ago
Although this was game was completely winning, if this also occurs in games in the balance of drawn and winning then a solution might bring elo as that wasted time would help stockfish convert.
its not an issues , I Test this Position via stockfish 15 and result was correct g4f3
But I don't Know why your Best move is g4g6
!!!!!
info depth 30 seldepth 64 multipv 1 score cp 318 nodes 30879487 nps 859649 hashfull 1000 tbhits 0 time 35921 pv g4f3 h1f2 f3f2 c8g4 d1g1 d8d7 h6g7 c6b4 a2a3 e8f7 a3b4 b8g8 g7h6 d7a4 f2c2 a4a1 c2b1 a1b1 c1b1 g4f3 g1g8 f3e4 b1c1 f7g8 f4e6 c7c6 e6d8 b7b5 b2b3 e4h1 c1c2 h1e4 c2b2 e4g2 b2a3 g2e4 c4c5 d6c5 b4c5 a6a5 d8b7 g8f7 b7a5 f7e6 a3b4 e6d5
bestmove g4f3 ponder h1f2
I Test this Position via stockfish 15
@movepicker I specified the version that I was using for a reason, don't just ignore it and use something else... The results can change with the version and depth But regardless of the version, Stockfish does this in a lot of occasions and needs a proper fix, it is a real issue
Another example, black again can't do anything and Stockfish again only wants to check the king and go back to g3 for no reason instead of directly playing Bf8 This kinds of moves should be filtered or ignored or something... it just doesn't make sense to even consider them, It's a waste of time!!
Stockfish https://github.com/official-stockfish/Stockfish/commit/cc7bcd5303a645223a6cd853817d3172754243aa
Stockfish 210522 by the Stockfish developers (see AUTHORS file)
uci
id name Stockfish 210522
id author the Stockfish developers (see AUTHORS file)
option name Debug Log File type string default
option name Threads type spin default 1 min 1 max 512
option name Hash type spin default 16 min 1 max 33554432
option name Clear Hash type button
option name Ponder type check default false
option name MultiPV type spin default 1 min 1 max 500
option name Skill Level type spin default 20 min 0 max 20
option name Move Overhead type spin default 10 min 0 max 5000
option name Slow Mover type spin default 100 min 10 max 1000
option name nodestime type spin default 0 min 0 max 10000
option name UCI_Chess960 type check default false
option name UCI_AnalyseMode type check default false
option name UCI_LimitStrength type check default false
option name UCI_Elo type spin default 1350 min 1350 max 2850
option name UCI_ShowWDL type check default false
option name SyzygyPath type string default <empty>
option name SyzygyProbeDepth type spin default 1 min 1 max 100
option name Syzygy50MoveRule type check default true
option name SyzygyProbeLimit type spin default 7 min 0 max 7
option name Use NNUE type check default true
option name EvalFile type string default nn-3c0aa92af1da.nnue
uciok
setoption name Hash value 128
ucinewgame
position fen rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1 moves d2d4 g8f6 c2c4 c7c5 g1f3 c5d4 f3d4 e7e5 d4b5 d7d5 c4d5 a7a6 b5c3 f8c5 g2g3 e8g8 f1g2 c8g4 h2h3 g4f5 e2e4 f5g6 e1g1 b8d7 b1d2 c5d4 c3e2 d4a7 g3g4 h7h5 e2g3 h5g4 h3g4 a8c8 g2f3 f6e8 a2a4 e8d6 g1g2 a7d4 f3e2 d7c5 f2f3 d8b6 a1a2 b6b4 f1e1 b7b5 b2b3 b4a5 c1a3 a5d8 a4b5 a6b5 a3b4 d8b6 d1b1 c8a8 e1c1 f7f6 d2f1 f8c8 c1c2 g6e8 g3f5 d6f5 g4f5 d4g1 f1g3 g1e3 b1a1 a8a2 a1a2 b6b8 g3h1 b8b7 h1f2 e3d4 a2b1 b7b8 f2g4 c8c7 b1e1 c5e6 d5e6 c7c2 g4f6 g7f6 e1g3 g8h7
d
+---+---+---+---+---+---+---+---+
| | q | | | b | | | | 8
+---+---+---+---+---+---+---+---+
| | | | | | | | k | 7
+---+---+---+---+---+---+---+---+
| | | | | P | p | | | 6
+---+---+---+---+---+---+---+---+
| | p | | | p | P | | | 5
+---+---+---+---+---+---+---+---+
| | B | | b | P | | | | 4
+---+---+---+---+---+---+---+---+
| | P | | | | P | Q | | 3
+---+---+---+---+---+---+---+---+
| | | r | | B | | K | | 2
+---+---+---+---+---+---+---+---+
| | | | | | | | | 1
+---+---+---+---+---+---+---+---+
a b c d e f g h
Fen: 1q2b3/7k/4Pp2/1p2pP2/1B1bP3/1P3PQ1/2r1B1K1/8 w - - 2 46
Key: CF2FE70B73A7DB27
Checkers:
go depth 28
info string NNUE evaluation using nn-3c0aa92af1da.nnue enabled
info depth 1 seldepth 1 multipv 1 score cp 367 nodes 77 nps 38500 tbhits 0 time 2 pv g3h2 e8h5
info depth 2 seldepth 2 multipv 1 score cp 367 nodes 113 nps 56500 tbhits 0 time 2 pv g3h2 e8h5 h2h5 h7g7
info depth 3 seldepth 3 multipv 1 score cp 200 nodes 263 nps 131500 tbhits 0 time 2 pv g3g4 c2e2 g2f1
info depth 4 seldepth 4 multipv 1 score cp 321 nodes 323 nps 161500 tbhits 0 time 2 pv g3h3 h7g7
info depth 5 seldepth 6 multipv 1 score cp 178 nodes 1536 nps 384000 tbhits 0 time 4 pv g2h3 b8a7 e2b5 a7a1
info depth 6 seldepth 6 multipv 1 score cp 0 nodes 2472 nps 494400 tbhits 0 time 5 pv b4d2 c2d2 g3h3 h7g7
info depth 7 seldepth 8 multipv 1 score cp 0 nodes 3350 nps 558333 tbhits 0 time 6 pv g3h4 h7g7 h4g3 g7h7
info depth 8 seldepth 7 multipv 1 score cp 49 nodes 3922 nps 560285 tbhits 0 time 7 pv g3h4 h7g8 h4f6 c2e2 g2f1
info depth 9 seldepth 7 multipv 1 score cp 147 nodes 4943 nps 617875 tbhits 0 time 8 pv g3h4 h7g8 b4e7 c2e2 g2h3
info depth 10 seldepth 15 multipv 1 score cp 112 nodes 14371 nps 756368 tbhits 0 time 19 pv g3h4 h7g8 h4f6 b8a7 f6f8 g8h7 f5f6 c2e2 g2h3 e2f2 e6e7 f2f3 h3g4 f3f4 g4g5
info depth 11 seldepth 18 multipv 1 score cp 147 nodes 23913 nps 797100 tbhits 0 time 30 pv g3h4 h7g8 h4f6 c2e2 g2h3 b8a7 f6f8 g8h7 f5f6 e2f2 b4e7 f2f3 h3g4
info depth 12 seldepth 15 multipv 1 score cp 147 nodes 30831 nps 811342 tbhits 0 time 38 pv g3h4 h7g8 h4f6 c2e2 g2h3 b8a7 f6f8 g8h7 f5f6 e2f2 b4e7 f2f3 h3g4 f3f4 g4g5
info depth 13 seldepth 13 multipv 1 score cp 242 nodes 39012 nps 848086 tbhits 0 time 46 pv g3h4 h7g8 h4g4 g8h7 b4f8 b8a7 e6e7 c2e2 g2h3
info depth 14 seldepth 19 multipv 1 score cp 528 nodes 78112 nps 1041493 tbhits 0 time 75 pv g3h3 h7g7 h3g4 g7h7 b4f8 c2c7 e6e7 c7e7 g4h4 h7g8 f8e7 b8a7 e7f6 a7h7 h4g5 g8f7 f6e7
info depth 15 seldepth 23 multipv 1 score cp 588 nodes 97377 nps 1106556 tbhits 0 time 88 pv g3h4 h7g8 h4g4 g8h7 b4f8 c2c7 e6e7 c7e7 g4h4 h7g8 f8e7 b8a7 e7f6 a7h7 h4h7 g8h7
info depth 16 seldepth 25 multipv 1 score cp 632 nodes 205642 nps 1117619 tbhits 0 time 184 pv g3h3 h7g7 h3g4 g7h7 b4f8 c2c7 e6e7 c7e7 g4h4 h7g8 f8e7 b8a7 e7f6 a7h7 h4g5 g8f7 f3f4 e8c6 e2f3
info depth 17 seldepth 31 multipv 1 score cp 786 nodes 279888 nps 1147081 tbhits 0 time 244 pv g3h3 h7g7 h3g3 g7h7 g3h4 h7g8 h4g4 g8h7 b4f8 c2c7 e6e7 c7e7 g4h4 h7g8 f8e7 b8a7 e7f6 a7h7 h4g5 g8f7 f3f4 e8c6 e2f3 f7e8 f6e5 d4e5 f4e5
info depth 18 seldepth 35 multipv 1 score cp 774 nodes 325554 nps 1154446 tbhits 0 time 282 pv g3h4 h7g8 h4g4 g8h7 b4f8 c2c7 e6e7 c7e7 g4h4 h7g8 f8e7 b8a7 e7f6 a7h7 h4g3 g8f8 f6e5 d4e5 g3e5 h7h6 f3f4 h6h7 e5c3 e8c6 e2f3
info depth 19 seldepth 32 multipv 1 score cp 755 nodes 397131 nps 1134660 tbhits 0 time 350 pv g3h4 h7g8 h4g4 g8h7 b4f8 c2c7 e6e7 c7e7 g4h4 h7g8 f8e7 b8a7 e7f6 a7h7 h4g5 g8f8 f3f4 e8c6 e2f3 f8e8 f6e5 e8d7 g2g3 d7c8 e5d4 c8b7
info depth 20 seldepth 34 multipv 1 score cp 802 nodes 501381 nps 1129236 tbhits 0 time 444 pv g3h4 h7g8 h4g4 g8h7 b4f8 c2c7 e6e7 c7e7 g4h4 h7g8 f8e7 b8c7 e7f6 c7h7 h4g5 g8f8 f3f4 d4c5 f4e5 e8c6 e2f3 b5b4 g5h4 h7h4 f6h4 c5e7 f5f6 e7f6 h4f6
info depth 21 seldepth 35 multipv 1 score cp 768 nodes 578623 nps 1127920 tbhits 0 time 513 pv g3h4 h7g8 h4g4 g8h7 b4f8 c2c7 e6e7 c7e7 g4h4 h7g8 f8e7 b8c7 e7f6 c7h7 h4g5 g8f8 f3f4 d4c5 f4e5 e8c6 e2f3 b5b4 g2g3 c6d7 e5e6 c5d6 g3g4 d7c6 g5h4
info depth 22 seldepth 43 multipv 1 score cp 776 nodes 919837 nps 1073322 tbhits 0 time 857 pv g3h4 h7g8 h4g4 g8h7 b4f8 c2c7 e6e7 c7e7 g4h4 h7g8 f8e7 b8c7 e7f6 c7h7 h4g5 g8f8 f3f4 d4e3 f6e5 e3c5 e2f3 h7g8 f3g4 g8h7 g5d8 h7h6 f5f6 h6g6 g2f3 f8g8 e5d6 c5d6 d8d6
info depth 23 seldepth 51 multipv 1 score cp 806 nodes 1852452 nps 1037788 hashfull 89 tbhits 0 time 1785 pv g3h4 h7g8 h4g4 g8h7 b4f8 c2c7 e6e7 c7e7 g4h4 h7g8 f8e7 b8c7 e7f6 c7h7 h4g5 g8f8 f3f4 d4e3 f6e5 e3c5 e2f3 c5e7 f5f6 e7c5 e5c3 h7g6 g5g6 e8g6 b3b4 c5e3 g2g3 g6f7 f3g4 f7e8 e4e5 e3c1 e5e6
info depth 24 seldepth 49 multipv 1 score cp 806 nodes 1959884 nps 1043046 hashfull 94 tbhits 0 time 1879 pv g3h4 h7g8 h4g4 g8h7 b4f8 c2c7 e6e7 c7e7 g4h4 h7g8 f8e7 b8c7 e7f6 c7h7 h4g5 g8f8 f3f4 d4e3 f6e5 e3c5 g5f6 f8g8 f6d8 g8f8 e5d6 c5d6 d8d6 f8g8 d6d8 g8f8 d8f6 f8g8 f6g5 g8f8 e2f3 b5b4 g5f6 e8f7 f6d8 f8g7 d8g5 g7f8 f3d1 f7e8 g5f6 f8g8 f6d8 h7g7 d8g5
info depth 25 seldepth 46 multipv 1 score cp 816 nodes 2353899 nps 1048974 hashfull 111 tbhits 0 time 2244 pv g3h4 h7g8 h4g4 g8h7 b4f8 c2c7 e6e7 c7e7 g4h4 h7g8 f8e7 b8c7 e7f6 c7h7 h4g5 g8f8 f3f4 d4e3 f6e5 e3c5 g5f6 f8g8 f6d8 h7e7 d8e7 c5e7 e5c3 e8c6 e2f3 b5b4 c3a1 e7h4 e4e5 h4e1 f3c6 e1c3 c6d5 g8g7
info depth 26 currmove g3h3 currmovenumber 2
info depth 26 currmove g2h3 currmovenumber 3
info depth 26 currmove g3g7 currmovenumber 4
info depth 26 currmove g3g6 currmovenumber 5
info depth 26 currmove e6e7 currmovenumber 6
info depth 26 currmove b4e7 currmovenumber 7
info depth 26 currmove g3g8 currmovenumber 8
info depth 26 currmove g3g4 currmovenumber 9
info depth 26 currmove g2h2 currmovenumber 10
info depth 26 currmove g3f2 currmovenumber 11
info depth 26 currmove g3h2 currmovenumber 12
info depth 26 currmove g2h1 currmovenumber 13
info depth 26 currmove b4d6 currmovenumber 14
info depth 26 currmove g3g5 currmovenumber 15
info depth 26 currmove b4d2 currmovenumber 16
info depth 26 currmove b4c3 currmovenumber 17
info depth 26 currmove b4f8 currmovenumber 18
info depth 26 currmove b4c5 currmovenumber 19
info depth 26 currmove b4a5 currmovenumber 20
info depth 26 currmove b4a3 currmovenumber 21
info depth 26 currmove b4e1 currmovenumber 22
info depth 26 currmove g2f1 currmovenumber 23
info depth 26 currmove g3f4 currmovenumber 24
info depth 26 currmove f3f4 currmovenumber 25
info depth 26 currmove g3e1 currmovenumber 26
info depth 26 currmove g3e5 currmovenumber 27
info depth 26 seldepth 50 multipv 1 score cp 839 nodes 3410579 nps 1047153 hashfull 161 tbhits 0 time 3257 pv g3h4 h7g8 h4g3 g8h7 b4f8 c2c7 e6e7 c7e7 g3h4 h7g8 f8e7 b8c7 e7f6 c7h7 h4g5 g8f8 f3f4 d4e3 f6e5 e3c5 g5f6 f8g8 f6d8 h7e7 d8e7 c5e7 e5c3 e8c6 e2f3 b5b4 c3d4 g8f8 e4e5 c6d7 f5f6 e7f6 e5f6 d7e6 f3d1 e6f7 g2f2
info depth 27 currmove g3h4 currmovenumber 1
info depth 27 currmove g3h3 currmovenumber 2
info depth 27 currmove g3g7 currmovenumber 3
info depth 27 currmove g3g6 currmovenumber 4
info depth 27 currmove e6e7 currmovenumber 5
info depth 27 currmove g2h3 currmovenumber 6
info depth 27 currmove b4e7 currmovenumber 7
info depth 27 currmove g3g8 currmovenumber 8
info depth 27 currmove g3h2 currmovenumber 9
info depth 27 currmove g3g4 currmovenumber 10
info depth 27 currmove g2h2 currmovenumber 11
info depth 27 currmove g3g5 currmovenumber 12
info depth 27 currmove g3f4 currmovenumber 13
info depth 27 currmove g3e1 currmovenumber 14
info depth 27 currmove b4f8 currmovenumber 15
info depth 27 currmove b4a5 currmovenumber 16
info depth 27 currmove b4c5 currmovenumber 17
info depth 27 currmove b4d6 currmovenumber 18
info depth 27 currmove f3f4 currmovenumber 19
info depth 27 currmove g3f2 currmovenumber 20
info depth 27 currmove b4d2 currmovenumber 21
info depth 27 currmove b4a3 currmovenumber 22
info depth 27 currmove b4e1 currmovenumber 23
info depth 27 currmove b4c3 currmovenumber 24
info depth 27 currmove g2f1 currmovenumber 25
info depth 27 currmove g2h1 currmovenumber 26
info depth 27 currmove g3e5 currmovenumber 27
info depth 27 seldepth 44 multipv 1 score cp 856 nodes 4021549 nps 1045916 hashfull 192 tbhits 0 time 3845 pv g3h4 h7g8 h4g3 g8h7 b4f8 c2c7 e6e7 c7e7 g3h4 h7g8 f8e7 b8c7 e7f6 c7h7 h4g5 g8f8 f3f4 d4e3 f6e5 e3c5 g5f6 f8g8 f6d8 h7e7 d8e7 c5e7 e5c3 e8c6 e2f3 e7c5 b3b4 c5e3 e4e5 c6d7 f3d5 g8f8 e5e6 d7c8 c3e5 e3d2 f5f6 d2b4
info depth 28 currmove g3h4 currmovenumber 1
info depth 28 currmove g3h3 currmovenumber 2
info depth 28 currmove g3g7 currmovenumber 3
info depth 28 currmove g2h3 currmovenumber 4
info depth 28 currmove g3g6 currmovenumber 5
info depth 28 currmove e6e7 currmovenumber 6
info depth 28 currmove b4e7 currmovenumber 7
info depth 28 currmove b4f8 currmovenumber 8
info depth 28 currmove g3g8 currmovenumber 9
info depth 28 currmove g3h2 currmovenumber 10
info depth 28 currmove g3g5 currmovenumber 11
info depth 28 currmove g3g4 currmovenumber 12
info depth 28 currmove b4c5 currmovenumber 13
info depth 28 currmove b4d6 currmovenumber 14
info depth 28 currmove b4d2 currmovenumber 15
info depth 28 currmove f3f4 currmovenumber 16
info depth 28 currmove g2f1 currmovenumber 17
info depth 28 currmove b4c3 currmovenumber 18
info depth 28 currmove b4a5 currmovenumber 19
info depth 28 currmove b4e1 currmovenumber 20
info depth 28 currmove b4a3 currmovenumber 21
info depth 28 currmove g3e1 currmovenumber 22
info depth 28 currmove g3f4 currmovenumber 23
info depth 28 currmove g3f2 currmovenumber 24
info depth 28 currmove g2h1 currmovenumber 25
info depth 28 currmove g2h2 currmovenumber 26
info depth 28 currmove g3e5 currmovenumber 27
info depth 28 seldepth 47 multipv 1 score cp 845 nodes 4467400 nps 1035798 hashfull 213 tbhits 0 time 4313 pv g3h4 h7g8 h4g3 g8h7 b4f8 c2c7 e6e7 c7e7 g3h4 h7g8 f8e7 b8c7 e7f6 c7h7 h4g5 g8f8 f3f4 d4e3 f6e5 e3c5 g5f6 f8g8 f6d8 h7e7 d8e7 c5e7 e5c3 b5b4 e2c4 g8f8 c3e5 e8c6 c4d5 c6b5 f5f6 e7c5 g2g3 b5d3 f4f5
bestmove g3h4 ponder h7g8
I Test this Position also and result was b4f8
But I don't Know why your Best move is g3h4
!
info depth 28 seldepth 45 multipv 1 score cp 796 nodes 12582745 nps 1087908 hashfull 999 tbhits 0 time 11566 pv b4f8 c2c7 e6e7 c7e7 g3h3 h7g8 f8e7 b8c7 e7f6 c7h7 f3f4 h7h3 g2h3 d4c5 f6e5 g8h7 e5f6 e8f7 e2b5 f7b3 f6c3 b3c2 b5c6 h7h6 f5f6 h6h5 h3g3 h5g6 c6e8 g6h6 e4e5 c2f5 e8c6 h6g6 g3f3 f5e6 c6e4 g6h5 f3g3 e6f7 e4f5 f7b3 f5d7 b3c2 e5e6
bestmove b4f8 ponder c2c7
my stockfish version is stockfish_15_x64_avx2
, download this version & do test to see your problem solved or not ...
https://stockfishchess.org/files/stockfish_15_win_x64_avx2.zip
my stockfish version is
stockfish_15_x64_avx2
, download this version & do test to see your problem solved or not ...
Again... stop using an older/different version and expect the same result at the same depth...
Stockfish 15 by the Stockfish developers (see AUTHORS file)
position fen rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1 moves d2d4 g8f6 c2c4 g7g6 b1c3 f8g7 e2e4 d7d6 f2f3 e8g8 c1g5 b8c6 g1e2 a7a6 d1d2 f8e8 h2h4 h7h5 e1c1 a8b8 g2g4 h5g4 h4h5 f6h5 c3d5 f7f6 g5h6 g4f3 e2g3 h5g3 h6g7 g6g5 d2h2 g3h1 h2h8 g8f7 h8h7 e8g8 g7h6 f7e6 h7f5 e6f7 f5h7 f7e6 d5f4 e6d7 f1h3 g5g4 h7f5 d7e8 f5h5 e8d7 h3g4 g8g4 h5g4 d7e8
d
+---+---+---+---+---+---+---+---+
| | r | b | q | k | | | | 8
+---+---+---+---+---+---+---+---+
| | p | p | | p | | | | 7
+---+---+---+---+---+---+---+---+
| p | | n | p | | p | | B | 6
+---+---+---+---+---+---+---+---+
| | | | | | | | | 5
+---+---+---+---+---+---+---+---+
| | | P | P | P | N | Q | | 4
+---+---+---+---+---+---+---+---+
| | | | | | p | | | 3
+---+---+---+---+---+---+---+---+
| P | P | | | | | | | 2
+---+---+---+---+---+---+---+---+
| | | K | R | | | | n | 1
+---+---+---+---+---+---+---+---+
a b c d e f g h
Fen: 1rbqk3/1pp1p3/p1np1p1B/8/2PPPNQ1/5p2/PP6/2KR3n w - - 1 29
Key: D1C84C7B7836A670
Checkers:
go depth 21
info string NNUE evaluation using nn-6877cd24400e.nnue enabled
info depth 1 seldepth 1 multipv 1 score cp 206 nodes 59 nps 29500 tbhits 0 time 2 pv g4g8 e8d7
info depth 2 seldepth 2 multipv 1 score cp 320 nodes 116 nps 58000 tbhits 0 time 2 pv g4f3 e8d7
info depth 3 seldepth 3 multipv 1 score cp 290 nodes 234 nps 78000 tbhits 0 time 3 pv g4g8 e8d7 g8g4 f6f5 g4f5 d7e8
info depth 4 seldepth 4 multipv 1 score cp 290 nodes 307 nps 76750 tbhits 0 time 4 pv g4g8 e8d7 g8g4 f6f5
info depth 5 seldepth 5 multipv 1 score cp 302 nodes 543 nps 108600 tbhits 0 time 5 pv g4f3 e8d7 f3g4 d7e8 g4g7
info depth 6 seldepth 6 multipv 1 score cp 450 nodes 697 nps 139400 tbhits 0 time 5 pv g4f3 e8d7 f3g4 d7e8
info depth 7 seldepth 9 multipv 1 score cp 357 nodes 2802 nps 350250 tbhits 0 time 8 pv g4g6 e8d7 g6g4 d7e8 g4f3 e8d7 f3h3 d7e8
info depth 8 seldepth 12 multipv 1 score cp 330 nodes 4162 nps 416200 tbhits 0 time 10 pv g4g6 e8d7 g6g4 d7e8 g4f3 e8d7 f3h3 d7e8 h3h1 c8g4 d1g1
info depth 9 seldepth 11 multipv 1 score cp 345 nodes 6023 nps 463307 tbhits 0 time 13 pv g4g6 e8d7 g6g4 d7e8 g4f3 e8d7 f3h3 d7e8 h3h1 c8g4
info depth 10 seldepth 21 multipv 1 score cp 242 nodes 75022 nps 833577 tbhits 0 time 90 pv g4g6 e8d7 g6g4 d7e8 g4f3 h1f2 d1g1 e8d7 f3f2 d8h8
info depth 11 seldepth 20 multipv 1 score cp 209 nodes 148304 nps 915456 tbhits 0 time 162 pv g4f3 h1f2 d1g1 e8d7 f3f2 d8h8 h6g7 h8h7 f4d5 h7e4 g7f6 e4d3
info depth 12 seldepth 19 multipv 1 score cp 193 nodes 176985 nps 926623 tbhits 0 time 191 pv g4g8 e8d7 g8g4 d7e8 g4f3 h1f2 d1g1 e8d7 f3f2 d8h8 h6g7 h8h7 f4d5 d7e8 g7f6 c8e6 d5c7 e8d7 c7e6
info depth 13 seldepth 23 multipv 1 score cp 238 nodes 184968 nps 934181 tbhits 0 time 198 pv g4h5 e8d7 h5g4 d7e8 g4f3 h1f2 d1g1 e8d7 f3f2 d8h8
info depth 14 seldepth 28 multipv 1 score cp 250 nodes 444404 nps 949581 tbhits 0 time 468 pv g4h5 e8d7 h5g4 d7e8 g4f3 h1f2 f3f2 e8d7 d1g1 b7b5 g1g7 b5c4 f2h4 c6d4 h4f6 c7c5 f4d5 d7c6 g7e7
info depth 15 seldepth 25 multipv 1 score cp 183 nodes 639399 nps 952904 tbhits 0 time 671 pv g4g6 e8d7 g6f5 d7e8 f5h5 e8d7 h5h3 d7e8 h3f3 h1f2 f3f2 c8g4 d1g1 d8d7 h6g7 e8d8 f2h4 c6d4 g7f6 e7f6 h4f6 d8c8 f6d4 b7b5 d4d5
info depth 16 seldepth 33 multipv 1 score cp 217 nodes 902565 nps 951069 tbhits 0 time 949 pv g4f3 h1f2 f3f2 c8g4 d1g1 d8d7 h6g7 e7e5 f4d5 d7g7 g1g4 g7h6 c1d1 e8f7 g4h4 h6g6 d5f6 e5d4 f6d7 f7e6 d7b8 c6b8 f2d4 e6d7
info depth 17 seldepth 31 multipv 1 score cp 196 nodes 1039022 nps 938592 hashfull 352 tbhits 0 time 1107 pv g4g6 e8d7 g6g4 d7e8 g4f3 h1f2 f3f2 c8g4 d1g1 d8d7 h6g7 e7e5 f4d5 d7g7 g1g4 g7h6 c1d1 e8f7 g4h4 h6g6 d5f6 e5d4 f6d7 f7e6 d7b8 c6b8 f2d4 b8d7 d4e3
info depth 18 seldepth 32 multipv 1 score cp 195 nodes 1184858 nps 928572 hashfull 402 tbhits 0 time 1276 pv g4g6 e8d7 g6g4 d7e8 g4f3 h1f2 f3f2 c8g4 d1g1 d8d7 h6g7 e7e5 f4d5 d7g7 g1g4 g7h6 c1d1 e8f7 g4h4 h6g6 d5f6 e5d4 f6d7 f7e7 d7b8 c6b8 f2d4 b8d7 d4e3 b7b6 e3h3 g6g1 d1e2
info depth 19 seldepth 38 multipv 1 score cp 233 nodes 1332252 nps 931644 hashfull 442 tbhits 0 time 1430 pv g4g6 e8d7 g6g4 d7e8 g4f3 h1f2 f3f2 c8g4 d1g1 d8d7 h6g7 e7e5 f4d5 d7g7 g1g4 g7h6 c1d1 e8f7 g4h4 h6g6 d5f6 e5d4 f6d7 f7e7 d7b8 c6b8 f2d4 b8d7 d4f2 d7c5 f2f5 g6g1 d1e2
info depth 20 seldepth 34 multipv 1 score cp 227 nodes 1470783 nps 933830 hashfull 482 tbhits 0 time 1575 pv g4g6 e8d7 g6g4 d7e8 g4h5 e8d7 h5h3 d7e8 h3f3 h1f2 f3f2 c8g4 d1g1 d8d7 h6g7 e7e5 f4d5 d7g7 g1g4 g7h6 c1d1 e8f7 g4h4 h6g6 d5f6 e5d4 f6d7 f7e7 d7b8 c6b8 f2d4 b8d7 d4f2 d7e5
info depth 21 seldepth 34 multipv 1 score cp 252 nodes 1812422 nps 939078 hashfull 573 tbhits 0 time 1930 pv g4g6 e8d7 g6g4 d7e8 g4f3 h1f2 f3f2 c8g4 d1g1 d8d7 h6g7 e8d8 f2h4 c6d4 g7f6 e7f6 h4f6 d8c8 f6d4 b7b5 f4d5 c7c5 d4f6 c8b7 e4e5 b5c4 d5e3
bestmove g4g6 ponder e8d7
Happy now? sigh
yes yes , that is Like Zugzwang Issues ,maybe exist a bug in Nnue Evaluation , Thanks Dav for Report This Issues
maybe exist a bug in Nnue Evaluation
No, this happens in Stockfish Classical too
Stockfish 291021 64 POPCNT by T. Romstad, M. Costalba, J. Kiiski, G. Linscott
position fen rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1 moves d2d4 g8f6 c2c4 g7g6 b1c3 f8g7 e2e4 d7d6 f2f3 e8g8 c1g5 b8c6 g1e2 a7a6 d1d2 f8e8 h2h4 h7h5 e1c1 a8b8 g2g4 h5g4 h4h5 f6h5 c3d5 f7f6 g5h6 g4f3 e2g3 h5g3 h6g7 g6g5 d2h2 g3h1 h2h8 g8f7 h8h7 e8g8 g7h6 f7e6 h7f5 e6f7 f5h7 f7e6 d5f4 e6d7 f1h3 g5g4 h7f5 d7e8 f5h5 e8d7 h3g4 g8g4 h5g4 d7e8
d
+---+---+---+---+---+---+---+---+
| | r | b | q | k | | | | 8
+---+---+---+---+---+---+---+---+
| | p | p | | p | | | | 7
+---+---+---+---+---+---+---+---+
| p | | n | p | | p | | B | 6
+---+---+---+---+---+---+---+---+
| | | | | | | | | 5
+---+---+---+---+---+---+---+---+
| | | P | P | P | N | Q | | 4
+---+---+---+---+---+---+---+---+
| | | | | | p | | | 3
+---+---+---+---+---+---+---+---+
| P | P | | | | | | | 2
+---+---+---+---+---+---+---+---+
| | | K | R | | | | n | 1
+---+---+---+---+---+---+---+---+
a b c d e f g h
Fen: 1rbqk3/1pp1p3/p1np1p1B/8/2PPPNQ1/5p2/PP6/2KR3n w - - 1 29
Key: D1C84C7B7836A670
Checkers:
go depth 21
info depth 1 seldepth 1 multipv 1 score cp 306 nodes 71 nps 71000 tbhits 0 time 1 pv f4e6
info depth 2 seldepth 2 multipv 1 score cp 172 nodes 188 nps 94000 tbhits 0 time 2 pv g4f3 e8d7 f3h1
info depth 3 seldepth 3 multipv 1 score cp 29 nodes 334 nps 167000 tbhits 0 time 2 pv g4f3 b7b5 f3h1 b5c4
info depth 4 seldepth 4 multipv 1 score cp 96 nodes 396 nps 132000 tbhits 0 time 3 pv g4f3 b7b5 f3h1 b5c4
info depth 5 seldepth 6 multipv 1 score cp 124 nodes 517 nps 172333 tbhits 0 time 3 pv g4f3 b7b5 f3h1 b5c4
info depth 6 seldepth 8 multipv 1 score cp 48 nodes 1207 nps 301750 tbhits 0 time 4 pv g4f3 d8d7 f3h5 e8d8 h5h1 d7g4 a2a3
info depth 7 seldepth 11 multipv 1 score cp 273 nodes 1593 nps 318600 tbhits 0 time 5 pv g4f3 d8d7 d1g1
info depth 8 seldepth 10 multipv 1 score cp 266 nodes 4967 nps 827833 tbhits 0 time 6 pv g4f3 e8d7 d4d5 c6b4 d1h1 b4a2 c1b1 a2b4 f3g4 d7e8
info depth 9 seldepth 16 multipv 1 score cp 118 nodes 21827 nps 1364187 tbhits 0 time 16 pv g4f3 e8d7 f4d5 d8g8 f3h1 d7d8 d1g1 g8h8 g1g7 c8e6 c1b1
info depth 10 seldepth 18 multipv 1 score cp 242 nodes 27276 nps 1363800 tbhits 0 time 20 pv g4f3 e8d7 f3g4 d7e8 g4g2 e8d7 f4d5
info depth 11 seldepth 21 multipv 1 score cp 130 nodes 67255 nps 1681375 tbhits 0 time 40 pv g4f3 h1f2 f3f2 e8d7 f4d5 d8h8 h6e3 h8h3 d1g1 b7b5 c4b5 b8b5
info depth 12 seldepth 19 multipv 1 score cp 290 nodes 80098 nps 1704212 tbhits 0 time 47 pv g4f3 h1f2 f3f2 e8d7 d4d5 c6e5
info depth 13 seldepth 20 multipv 1 score cp 90 nodes 158010 nps 1795568 tbhits 0 time 88 pv g4f3 c6d4 d1d4 e7e5 d4d1 e5f4 d1h1 c8e6 h6f4 e8e7 f4d2 d8g8 d2c3 e7d7 f3f6 d7c6
info depth 14 seldepth 23 multipv 1 score cp 131 nodes 177658 nps 1794525 tbhits 0 time 99 pv g4f3 c6d4 d1d4 e7e5 d4d1 e5f4 d1h1 e8e7 h1g1 c8e6 f3f4 e6f7 h6g7
info depth 15 seldepth 18 multipv 1 score cp 136 nodes 194771 nps 1786889 tbhits 0 time 109 pv g4f3 c6d4 d1d4 e7e5 d4d1 e5f4 d1h1 e8e7 h1g1 c8e6 f3f4 e6f7 h6g7 d8e8 f4f5 e8a4 f5f6 e7e8
info depth 16 seldepth 28 multipv 1 score cp 265 nodes 260423 nps 1846971 tbhits 0 time 141 pv g4f3 c6d4 d1d4 e7e5 f4d5 e5d4 d5f6 e8f7 f3h5 f7e7 h6g5 d8f8 f6d5 e7d7 h5h1
info depth 17 seldepth 31 multipv 1 score cp 200 nodes 495455 nps 1841840 tbhits 0 time 269 pv g4f3 h1f2 f3f2 c8g4 d1g1 d8d7 h6g7 c6b4 a2a3 e7e5 a3b4 e5f4 g7f6 d7e6 d4d5 e6f6 g1g4 e8d7 g4f4 f6g5 c1c2 b8g8 f4f7 d7c8 f2f5 c8b8 f7f8 b8a7 f8g8 g5g8
info depth 18 seldepth 35 multipv 1 score cp 250 nodes 576452 nps 1847602 tbhits 0 time 312 pv g4f3 h1f2 f3f2 c8g4 d1g1 d8d7 h6g7 c6d4 f2d4 e7e5 d4a7 b8d8 g7f6 e5f4 f6d8 e8d8 a7b7 d8e7 b7a6
info depth 19 seldepth 32 multipv 1 score cp 126 nodes 1286937 nps 1889775 tbhits 0 time 681 pv g4f3 c6d4 d1d4 e7e5 d4d1 e5f4 h6f4 h1f2 f3f2 c8e6 d1h1 e6g8 b2b3 b7b5 f2g2 e8d7 c4c5 d7c6 c5d6 c7d6 h1d1 c6b7 d1d6
info depth 20 seldepth 40 multipv 1 score cp 149 nodes 2136010 nps 1922601 hashfull 796 tbhits 0 time 1111 pv g4f3 c6d4 d1d4 e7e5 d4d1 e5f4 h6f4 d8d7 d1h1 d7g7 f3h5 g7f7 h5h8 f7f8 h8h2 b7b6 b2b3 c8g4 h2g2 f8g8 h1h6 e8e7 g2h1 b6b5 h6h7 e7e6 c1b2 b5c4
info depth 21 seldepth 39 multipv 1 score cp 172 nodes 2432768 nps 1935376 hashfull 848 tbhits 0 time 1257 pv g4g6 e8d7 g6g4 d7e8 g4f3 c6d4 d1d4 e7e5 f4d5 e5d4 d5f6 e8e7 f3h1 d8h8 h1h4 h8f6 h6g5 f6g5 h4g5 e7d7 e4e5 b7b5 g5g4 d7c6 g4e4 c6d7 e4d4 b5c4 e5d6 c7d6
bestmove g4g6 ponder e8d7
I Test this Position via komodo 12 and result was g4f3
, I did not see still this issues in other engine like komodo , fritz , ...
info depth 21 time 6773 nodes 12114232 score cp 138 lowerbound nps 1788451 hashfull 330 tbhits 0 pv g4f3 h1f2 f3f2
info time 6774 nodes 12114232 nps 1788182 hashfull 330
bestmove g4f3 ponder h1f2
Thanks Dav for found This Issues in Stockfish engine
I think that this is the commit that causes the issue. https://github.com/official-stockfish/Stockfish/commit/95d7369e54f20715345cf5408040f3c7d1ec8415
I think that this is the commit that causes the issue. 95d7369
This doesn't seem to be a new issue created by a recent commit as it can be reproduced with very old versions of Stockfish like (as I already mentioned) Stockfish Classical but I tried with even older versions like Stockfish 9 and same things happen
A Discord user claims that weird shuffling can be found even in simple cases of KRvK, and that Stockfish 14 does not have a shuffling problem, while SFs 12, 13 and 15 all have a shuffling problem (even while 14 doesn't)
The shuffling in KRvK is probably because the NNUE nets don't score positions in KRvK in a way that makes the path to mate obvious, it is just a winning position. The classical eval has terms that pushes the king to the edge.
More generally, SF doesn't repeat moves 'on purpose' so there are no terms that obviously favor repetition. Nevertheless, this behavior could of course emerge from whatever combination of search heuristics.
there are no terms that obviously favor repetition.
By "on purpose" I don't really mean "favor" as in "preferring" I mean more like "knowing and not caring". If Stockfish sees in his PV that a sequence of moves leads to the exact same position that it has now it should just filter those moves and go directly to the move that doesn't repeat because there really isn't a point on playing the other ones.
Let w1-b1-w2-b2-w3-b3-w4-... be the pv with SF playing move w1 and lets assume a repetition of the current position after move b3. You suggest that we should cut the first 6 plies of the pv and directly play w4, right? Lets assume that SF got this PV with a search of depth 30. Then w4 was searched with a depth of 24 or maybe even lower. Why should we trust a move that was searched with this low depth? Lets assume that w1 and b1 are played and now SF searches on this positions, gets a deeper search and finds a new move w2' that is better than w2 from the old pv. This new best move could not be played if we had cut the pv and already played the depth 24 move w4. Red hearts don't gain Elo.
Lets assume that w1 and b1 are played and now SF searches on this positions, gets a deeper search and finds a new move w2' that is better than w2 from the old pv. This new best move could not be played if we had cut the pv and already played the depth 24 move w4.
The repetitions to which I refer:
I think that limiting a filter to apply only in these situations could avoid trusting low depth moves Or maybe not filtering them but giving more importance and depth to the progress move to see if Stockfish would stick to it? Not sure if Stockfish can understand what progress in a position is tho
Red hearts don't gain Elo.
Please ignore the reactions, they are from a spammer that is now banned
Lets assume that w1 and b1 are played and now SF searches on this positions, gets a deeper search and finds a new move w2' that is better than w2 from the old pv. This new best move could not be played if we had cut the pv and already played the depth 24 move w4.
The repetitions to which I refer:
* Are usually just 2 moves so: w1, b1, w2, b2, w3
My argument "we may find a better w2' after w1-b1 are played" is still valid in your case.
I think that limiting a filter to apply only in these situations could avoid trusting low depth moves Or maybe not filtering them but giving more importance and depth to the progress move to see if Stockfish would stick to it? Not sure if Stockfish can understand what progress in a position is tho
Evaluating 2-fold repetition as a draw even in root (if expected score is > 0) should do what you want. No idea if this was already tested at some time.
Evaluating 2-fold repetition as a draw even in root (if expected score is > 0) should do what you want. No idea if this was already tested at some time.
No idea if it was ever tested to be honest I like the idea, maybe consider it a draw only if it happens in the next 2 moves and don't consider it a draw if it happened before? So if w1 has already been played, don't consider w2 a draw but if you didn't play w1 yet then consider w2 a draw? 🤔 That would make Stockfish avoid thinking about w2 and look for other potential moves, look for w2' and if it doesn't find it then it would look at w3
there was old work on 2-fold vs 3-fold, multiple PRs IIRC see https://github.com/official-stockfish/Stockfish/pull/1126 as a starter
Hmm yeah I didn't consider MultiPV... but I guess this behaviour could've been disabled only when MultiPV was more than 1? Why not? That way you "hide" the trick when MultiPV is 1 and it would still work for real games
Or show the repeating move with the same eval (or maybe 1cp less?) than the top move? But score it internally as a draw so resources are not wasted looking at it
In this case I'm talking about 5r1k/6n1/B1b4N/8/8/8/P4P2/1K5R w - - 0 1
Nf7+
Repeating the same position throughout the game is not an issue, but wanting to repeat moves when you have a clear advantage and when it's just making you waste time makes no sense.
I don't see how purposely repeating moves is considered an issue. Stockfish has determined a winning route, but has also determined that it can reach that same winning route via transposition by repeating moves, and if the opponent doesn't play the sharpest line which likely also repeats moves, the opponent would be even worse off than if Stockfish simply went directly for the winning route. Keep in mind the best PV will always try to optimize its evaluation, and it has determined that it has an opportunity to reach an even stronger evaluation. Isn't that the purpose of Stockfish?
The only argument I can think of is that if you have a clearly winning position, you should prioritize moves that reset the 50move rule over moves that might raise your evaluation?
Another factor I haven't seen discussed (but touched on by @RogerThiede) is that it isn't guaranteed that a repetition in the PV is a repetition in the game. SF may not be decided between whether to repeat or choose another line. If A -> B -> A -> D
and A -> B -> C
are both good sequences but D looks slightly better than C at eval time, should we skip a chance to find something when at B?[^1]
This gets messier when factoring in contempt (see Roger again), the time control, and ponder settings: probing for more eval time is a good thing in critical moments during competitive play. NNUE might even be optimizing for this in training.
In cases where it does matter (mate-in-N, 48-50 DTZ) SF should correct itself already, so I agree that this is a non-issue.
[^1]: On my tests of the original position with 82bb21d at depth 30, SF does exactly this, opting for g4g8
and switching between g8g4
and g8e6
at 1rbq2Q1/1ppkp3/p1np1p1B/8/2PPPN2/5p2/PP6/2KR3n w - - 3 30
.
bug, i agree for nothing but black king as it's checked on d7
how come it takes at least 4 sec to the inevitable forced move ?
I came up with a dirty trick that might solve the problem. It would be easy to test for regression, but it is a functional change nonetheless.
There is a general issue with tree search that winning a pawn in 5 moves has the same evaluation as winning a pawn in 1 move (all else being equal). So the engine has no reason to select one over the other and this may lead to shuffling. Some people like to dampen evaluations by depth (as is done for mates) to mitigate this problem. However this may potentially lead to hash table issues.
I forgot to put the link: https://github.com/dsmsgms/Stockfish/commit/23494555411e120573a4f2b62e519aa9b1648204
I might test it in a latter occasion.
Here is a potentially relevant page: https://www.ics.uci.edu/~eppstein/180a/990202a.html
Here is also a very bootleg solution I came up with: https://github.com/silversolver1/Stockfish/commit/56177af44a0a7787dea3ac01918d20a0920e5e22
This is obviously a very naive attempt at a solution but this is the reasoning:
this is clearly already done for values in the range of (greater than) VALUE_TB_WIN_IN_MAX_PLY and (less than) VALUE_TB_LOSS_IN_MAX_PLY
so why not then increase the range of values that this is done for, by decreasing the minimum values for which this ply adjustment occurs at?
For example, by decreasing VALUE_TB_WIN_IN_MAX_PLY to, say, VALUE_KNOWN_WIN * 2, we could have the effect of weighting clearly won positions by ply, without waiting so long for the positions to actually enter tb territory
I was looking at Navs stream (https://www.twitch.tv/videos/1488780810?t=18h26m47s) and I saw Stockfish doing some very very weird things. It did 7 useless moves to just take a pawn that it could've taken in 1, wasting a lot of time in the process. So I decided to check with homefish and see what it would do and the results were even more weird. Stockfish wanted to purposely repeat moves for absolutely no reason. Repeating the same position throughout the game is not an issue, but wanting to repeat moves when you have a clear advantage and when it's just making you waste time makes no sense.
Stockfish https://github.com/official-stockfish/Stockfish/commit/22b7909809c731aea691184dd7c1a2b02c5946af
In this position black is basically forced to go back and forth with the king from d7 to e8 waiting for white to do something and this is what Stockfish wants to do:
Stockfish basically wants to go to g6 to check the king expecting it to go back to d7 and then go back to g4?!?! WHY? It literally makes no sense, just take the pawn (g4f3) now!
This "bug" makes Stockfish look dumb during analysis and during games