Test Position for amateurs, humans, and even some stronger engines

[JVMerlino]

[d]8/3r4/5p2/2k1nQ2/8/3p4/3K4/8 b - -

This position came up in Halogen v Myrddin in a CCRL match. There is only one move that wins for Black. Myrddin failed to find it within the time control even with 5-man tablebases.

[zullil]

[d]8/3r4/5p2/2k1nQ2/8/3p4/3K4/8 b - -

This position came up in Halogen v Myrddin in a CCRL match. There is only one move that wins for Black. Myrddin failed to find it within the time control even with 5-man tablebases.

Stockfish-dev with pure NNUE eval (no hybridization) finds the correct move in about a tenth of a second. With 1 thread, 16 MB hash and no endgame tables:

Code: Select all

$ ./stockfishPureNNUE 
Stockfish 120820 by the Stockfish developers (see AUTHORS file)
position fen 8/3r4/5p2/2k1nQ2/8/3p4/3K4/8 b - -
d

 +---+---+---+---+---+---+---+---+
 |   |   |   |   |   |   |   |   | 8
 +---+---+---+---+---+---+---+---+
 |   |   |   | r |   |   |   |   | 7
 +---+---+---+---+---+---+---+---+
 |   |   |   |   |   | p |   |   | 6
 +---+---+---+---+---+---+---+---+
 |   |   | k |   | n | Q |   |   | 5
 +---+---+---+---+---+---+---+---+
 |   |   |   |   |   |   |   |   | 4
 +---+---+---+---+---+---+---+---+
 |   |   |   | p |   |   |   |   | 3
 +---+---+---+---+---+---+---+---+
 |   |   |   | K |   |   |   |   | 2
 +---+---+---+---+---+---+---+---+
 |   |   |   |   |   |   |   |   | 1
 +---+---+---+---+---+---+---+---+
   a   b   c   d   e   f   g   h

Fen: 8/3r4/5p2/2k1nQ2/8/3p4/3K4/8 b - - 0 1
Key: E7F7DF6FB0DFCBBC
Checkers: 
go infinite
info string NNUE evaluation enabled using nn-112bb1c8cdb5.nnue.
info depth 1 seldepth 1 multipv 1 score cp 48 nodes 46 nps 46000 tbhits 0 time 1 pv d7d6
info depth 2 seldepth 3 multipv 1 score cp 59 nodes 176 nps 176000 tbhits 0 time 1 pv d7d6 f5f2 c5c6
info depth 3 seldepth 4 multipv 1 score cp 63 nodes 340 nps 340000 tbhits 0 time 1 pv d7d6 f5f2 c5c6 f2f1 e5c4 d2c3
info depth 4 seldepth 6 multipv 1 score cp 23 nodes 690 nps 345000 tbhits 0 time 2 pv d7d6 f5f2 c5b4 f2e3 b4a3
info depth 5 seldepth 6 multipv 1 score cp 83 nodes 841 nps 420500 tbhits 0 time 2 pv d7d6 f5f2 c5b4 f2f4 e5c4 d2d1
info depth 6 seldepth 10 multipv 1 score cp 98 nodes 1969 nps 656333 tbhits 0 time 3 pv d7d6 f5c8 c5d4 c8c3 d4e4 c3b3
info depth 7 seldepth 10 multipv 1 score cp 167 nodes 3027 nps 756750 tbhits 0 time 4 pv c5b4 d2e1 d7d6 f5c8
info depth 8 seldepth 8 multipv 1 score cp 190 nodes 3613 nps 903250 tbhits 0 time 4 pv c5b4 d2d1 d7d6 f5c8 e5c4 c8c7 b4c3
info depth 9 seldepth 10 multipv 1 score cp 219 nodes 4214 nps 842800 tbhits 0 time 5 pv c5b4 d2d1 d7d6 f5c8 e5c4 c8a8 b4b3
info depth 10 seldepth 12 multipv 1 score cp 266 nodes 5701 nps 950166 tbhits 0 time 6 pv c5b4 d2d1 d7d6 f5c8 e5c4 c8a8 b4c3 a8a1 c3b3
info depth 11 seldepth 22 multipv 1 score cp 62 nodes 47259 nps 1243657 tbhits 0 time 38 pv d7d6 f5c8 c5d4 c8c3 d4e4 c3c5 e4f3 d2d1 d6c6 c5a7 f3g4 a7g7 g4f5 g7b7 c6c4
info depth 12 seldepth 25 multipv 1 score cp 84 nodes 84046 nps 1293015 tbhits 0 time 65 pv d7d6 f5c8 c5d4 c8c3 d4e4 c3c5 d6a6 c5e3 e4d5 e3g3 a6a3 g3g2 d5e6 g2e4 a3a1 e4b7 a1a2 d2d1 a2e2 b7a8
info depth 13 seldepth 31 multipv 1 score cp 80 nodes 197841 nps 1393246 tbhits 0 time 142 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d7 e4f4 d7g7 f4f2 g7e7 f2d4 c4b2 d1d2 e7e2 d2c1 e2e1 c1d2 e1d1 d2e3
info depth 14 seldepth 26 multipv 1 score cp 199 nodes 209473 nps 1415358 tbhits 0 time 148 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d7 e4f4
info depth 15 seldepth 31 multipv 1 score cp 156 nodes 271403 nps 1467043 tbhits 0 time 185 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d7 e4h4 c4b2 d1d2 d7c7 d2e3 c7c2 h4g5 d3d2 g5d5 b2c4 e3e2 b3a2 d5d3 a2b1 d3e4 c4b2
info depth 16 seldepth 32 multipv 1 score cp 182 nodes 294983 nps 1497375 tbhits 0 time 197 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d7 e4h4 c4b2 d1d2 d7c7 d2e3 c7c2 h4g5 d3d2 g5d5 b2c4 e3e2 b3a2 d5d3 a2b1 d3e4
info depth 17 seldepth 32 multipv 1 score cp 229 nodes 321243 nps 1522478 tbhits 0 time 211 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d7 e4h4 c4b2 d1d2 d7c7 d2e3 c7c2 h4g5 d3d2 g5d5 b2c4 e3e2 b3a2 d5c5

...

info depth 51 seldepth 90 multipv 1 score cp 819 nodes 632673357 nps 2044727 hashfull 1000 tbhits 0 time 309417 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d7 e4f3 b3c3 d1e1 d7d4 e1f2 d4d6 f2e1 c3b4 f3b7 d6b6 b7d7 b4c3 d7h3 c3c2 h3h7 b6e6 e1f2 e6e2 f2g3 c2c3 h7h1 d3d2 g3f4 e2e3 h1h8 c3c2 h8d4 d2d1q d4c4 e3c3 c4a2 c2c1 a2a1 c1d2 a1d1 d2d1 f4e5 c3c5 e5d6 c5a5 d6e6 d1d2 e6d6 d2d3 d6e6 d3d4 e6f6 d4e3 f6e7 a5e5 e7d6 e3d4 d6c6 e5a5 c6c7 d4d3

[JVMerlino]

[d]8/3r4/5p2/2k1nQ2/8/3p4/3K4/8 b - -

This position came up in Halogen v Myrddin in a CCRL match. There is only one move that wins for Black. Myrddin failed to find it within the time control even with 5-man tablebases.

Stockfish-dev with pure NNUE eval (no hybridization) finds the correct move in about a tenth of a second.

This is no surprise.

[zullil]

[d]8/3r4/5p2/2k1nQ2/8/3p4/3K4/8 b - -

This position came up in Halogen v Myrddin in a CCRL match. There is only one move that wins for Black. Myrddin failed to find it within the time control even with 5-man tablebases.

Stockfish-dev with pure NNUE eval (no hybridization) finds the correct move in about a tenth of a second.

This is no surprise.

Surprised me.

[Vinvin]

10 best moves after the only winning move 1...Kb4 :

Code: Select all

 58/33	00:12	 161 659 508	12 616 835	-M16	2.Qf1 Nc4+ 3.Ke1 d2+ 4.Ke2 Re7+ 5.Kf2 Nb2 6.Qa6 d1Q 7.Qb6+ Kc4 8.Qa6+ Kb3 9.Qb6+ Ka2 10.Qa6+ Kb1 11.Qf1 Nd3+ 12.Kg1 Re1 13.Qxe1 Qxe1+ 14.Kg2 Qf2+ 15.Kh3 Qf3+ 16.Kh2 Ne1 17.Kg1 Qg2+
 59/33	00:12	 161 659 508	12 616 835	-M16	2.Qe4+ Nc4+ 3.Ke1 f5 4.Qf4 d2+ 5.Kd1 Rd3 6.Qb8+ Kc3 7.Qh8+ Rd4 8.Qh3+ Kb4 9.Qh1 Ne3+ 10.Ke2 d1Q+ 11.Qxd1 Nxd1 12.Kf3 Rg4 13.Ke2 Kc3 14.Kf3 Kc2 15.Ke2 Rf4 16.Ke1 Kd3 17.Kxd1 Rf1+
 59/35	00:12	 161 659 508	12 616 835	-M17	2.Qf4+ Nc4+ 3.Ke1 Rd5 4.Qg4 d2+ 5.Kd1 f5 6.Qf4 Rd3 7.Qb8+ Kc3 8.Qh8+ Rd4 9.Qh3+ Kb4 10.Qh1 Ne3+ 11.Ke2 d1Q+ 12.Qxd1 Nxd1 13.Kf3 Rg4 14.Ke2 Kc3 15.Kf3 Kc2 16.Ke2 Rf4 17.Ke1 Kd3 18.Kxd1 Rf1+
 59/37	00:12	 161 659 508	12 616 835	-M18	2.Kc1 Rd6 3.Kd1 Nc4 4.Qe4 f5 5.Qe7 f4 6.Qe1+ Kb3 7.Qe8 d2 8.Qf7 Kc3 9.Qg7+ Rd4 10.Qc7 Rd5 11.Qxf4 Nb2+ 12.Ke2 d1Q+ 13.Ke3 Qd3+ 14.Kf2 Rf5 15.Ke1 Qd1+ 16.Kf2 Rxf4+ 17.Kg2 Qf3+ 18.Kg1 Qf2+ 19.Kh1 Rh4+
 59/39	00:12	 161 659 508	12 616 835	-M19	2.Qxd7 Nxd7 3.Kxd3 Kc5 4.Ke4 Kc4 5.Ke3 Kd5 6.Kf4 Kd4 7.Kf5 Ke3 8.Kg6 Ke4 9.Kh5 f5 10.Kh4 f4 11.Kh3 Ke3 12.Kg4 f3 13.Kf5 f2 14.Ke6 Kd4 15.Kxd7 Kd5 16.Kc7 f1Q 17.Kb6 Kd6 18.Kb7 Qb5+ 19.Kc8 Qb2 20.Kd8 Qb8+
 59/39	00:12	 161 659 508	12 616 835	-M19	2.Qe6 Nc4+ 3.Kd1 Rh7 4.Qe1+ Kb3 5.Qe4 Rh3 6.Kc1 Re3 7.Qb7+ Kc3 8.Kb1 Re1+ 9.Ka2 Re2+ 10.Ka1 Re5 11.Qc6 d2 12.Qa4 d1Q+ 13.Qxd1 Ra5+ 14.Kb1 Nd2+ 15.Qxd2+ Kxd2 16.Kb2 Rb5+ 17.Ka3 Kc3 18.Ka2 Ra5+ 19.Kb1 Ra4 20.Kc1 Ra1+
 59/43	00:12	 161 659 508	12 616 835	-M21	2.Kd1 Rd6 3.Qe4+ Nc4 4.Ke1 d2+ 5.Kd1 f5 6.Qb7+ Ka3 7.Qa7+ Kb2 8.Qg1 f4 9.Ke2 d1Q+ 10.Qxd1 f3+ 11.Kxf3 Rxd1 12.Ke4 Kc3 13.Kf4 Rd4+ 14.Kg5 Rd6 15.Kf5 Kd4 16.Kf4 Ne3 17.Kg3 Rg6+ 18.Kf3 Rf6+ 19.Kg3 Ke4 20.Kh3 Kf3 21.Kh2 Kf2 22.Kh1 Rh6+
 59/43	00:12	 161 659 508	12 616 835	-M21	2.Ke3 d2 3.Qb1+ Kc5 4.Qd1 Nc4+ 5.Kf3 Rd3+ 6.Ke2 Nb2 7.Qc2+ Kd5 8.Qd1 Nxd1 9.Kxd3 Ne3 10.Kxd2 Ke4 11.Kc3 f5 12.Kd2 f4 13.Ke2 Nf5 14.Kd1 Kd3 15.Ke1 Ke3 16.Kf1 f3 17.Ke1 f2+ 18.Kf1 Ng3+ 19.Kg2 f1Q+ 20.Kxg3 Qg1+ 21.Kh3 Kf4 22.Kh4 Qh2+
 59/53	00:12	 161 659 508	12 616 835	-M26	2.Ke1 Rd6 3.Qe4+ Nc4 4.Qb7+ Kc3 5.Qf3 Re6+ 6.Kf2 Re2+ 7.Kg1 Re1+ 8.Kg2 Ne3+ 9.Kg3 d2 10.Qxf6+ Kc2 11.Qc6+ Kb1 12.Qb5+ Kc1 13.Qd3 Nc2 14.Kf4 d1Q 15.Qxd1+ Rxd1 16.Kf5 Ne3+ 17.Kf4 Nc4 18.Kg4 Rd5 19.Kf4 Re5 20.Kg4 Kd2 21.Kf4 Ke2 22.Kg4 Ke3 23.Kg3 Rg5+ 24.Kh4 Kf4 25.Kh3 Nd2 26.Kh4 Nf3+ 27.Kh3 Rg3+
 59/63	00:12	 161 659 508	12 616 835	-M31	2.Qxf6 Nc4+ 3.Kd1 Rd6 4.Qe7 Kb3 5.Qe4 Rd8 6.Qf3 Kb4 7.Qc6 Kc3 8.Qf6+ Rd4 9.Qf3 Rg4 10.Qf6+ Kb3 11.Qf2 Re4 12.Kc1 Re2 13.Qg3 Re3 14.Qf2 d2+ 15.Kd1 Rc3 16.Ke2 Nb2 17.Kxd2 Rc2+ 18.Ke3 Nd1+ 19.Ke4 Nxf2+ 20.Kd4 Rc6 21.Kd5 Ra6 22.Kd4 Nd1 23.Ke5 Kc4 24.Ke4 Ra5 25.Kf3 Kd3 26.Kf4 Ne3 27.Kg3 Rf5 28.Kh3 Ke2 29.Kg3 Ke1 30.Kh3 Kf2 31.Kh4 Kf3 32.Kh3 Rh5+

[Vinvin]

Stockfish-dev with pure NNUE eval (no hybridization) finds the correct move in about a tenth of a second. With 1 thread, 16 MB hash and no endgame tables:
...

Stockfish before-NNUE finds the good move after 0.004 sec, depth = 1, 1 thread.

Code: Select all

FEN: 8/3r4/5p2/2k1nQ2/8/3p4/3K4/8 b - - 0 1

Stockfish_20073110_x64_modern:
Found 511 tablebases
 1/1	00:00	 42	10 500	-0,01	1. ... Kb4
 2/2	00:00	 66	16 500	-0,13	1. ... Kb4 2.Qxf6
 3/3	00:00	 110	27 500	-0,13	1. ... Kb4 2.Qxf6
 4/3	00:00	 161	40 250	-0,87	1. ... Kb4 2.Qe4+ Kb3
 5/7	00:00	 262	65 500	-1,21	1. ... Kb4 2.Qe4+ Nc4+ 3.Kc1 Kb3
 6/9	00:00	 1 556	222 285	-0,60	1. ... Kb4 2.Kc1 Rf7 3.Kd1 Kb3
 7/9	00:00	 2 570	321 250	-0,85	1. ... Kb4 2.Kc1 Rc7+ 3.Kd2 Rc3 4.Kd1
 8/14	00:00	 5 665	515 000	-0,79	1. ... Kb4 2.Kc1 Rd6 3.Qe4+ Kc3 4.Qa4 Nc4
 9/15	00:00	 6 992	537 846	-1,47	1. ... Kb4 2.Kc1 Rd6 3.Qe4+ Nc4 4.Qe7

[zullil]

Stockfish-dev with pure NNUE eval (no hybridization) finds the correct move in about a tenth of a second. With 1 thread, 16 MB hash and no endgame tables:
...

Stockfish before-NNUE finds the good move after 0.004 sec, depth = 1, 1 thread.

Code: Select all

FEN: 8/3r4/5p2/2k1nQ2/8/3p4/3K4/8 b - - 0 1

Stockfish_20073110_x64_modern:
Found 511 tablebases
 1/1	00:00	 42	10 500	-0,01	1. ... Kb4
 2/2	00:00	 66	16 500	-0,13	1. ... Kb4 2.Qxf6
 3/3	00:00	 110	27 500	-0,13	1. ... Kb4 2.Qxf6
 4/3	00:00	 161	40 250	-0,87	1. ... Kb4 2.Qe4+ Kb3
 5/7	00:00	 262	65 500	-1,21	1. ... Kb4 2.Qe4+ Nc4+ 3.Kc1 Kb3
 6/9	00:00	 1 556	222 285	-0,60	1. ... Kb4 2.Kc1 Rf7 3.Kd1 Kb3
 7/9	00:00	 2 570	321 250	-0,85	1. ... Kb4 2.Kc1 Rc7+ 3.Kd2 Rc3 4.Kd1
 8/14	00:00	 5 665	515 000	-0,79	1. ... Kb4 2.Kc1 Rd6 3.Qe4+ Kc3 4.Qa4 Nc4
 9/15	00:00	 6 992	537 846	-1,47	1. ... Kb4 2.Kc1 Rd6 3.Qe4+ Nc4 4.Qe7

How about without 511 endgame tables?

[Vinvin]

Stockfish-dev with pure NNUE eval (no hybridization) finds the correct move in about a tenth of a second. With 1 thread, 16 MB hash and no endgame tables:
...

Stockfish before-NNUE finds the good move after 0.004 sec, depth = 1, 1 thread.

Code: Select all

FEN: 8/3r4/5p2/2k1nQ2/8/3p4/3K4/8 b - - 0 1

Stockfish_20073110_x64_modern:
Found 511 tablebases
 1/1	00:00	 42	10 500	-0,01	1. ... Kb4
 2/2	00:00	 66	16 500	-0,13	1. ... Kb4 2.Qxf6
 3/3	00:00	 110	27 500	-0,13	1. ... Kb4 2.Qxf6
 4/3	00:00	 161	40 250	-0,87	1. ... Kb4 2.Qe4+ Kb3
 5/7	00:00	 262	65 500	-1,21	1. ... Kb4 2.Qe4+ Nc4+ 3.Kc1 Kb3
 6/9	00:00	 1 556	222 285	-0,60	1. ... Kb4 2.Kc1 Rf7 3.Kd1 Kb3
 7/9	00:00	 2 570	321 250	-0,85	1. ... Kb4 2.Kc1 Rc7+ 3.Kd2 Rc3 4.Kd1
 8/14	00:00	 5 665	515 000	-0,79	1. ... Kb4 2.Kc1 Rd6 3.Qe4+ Kc3 4.Qa4 Nc4
 9/15	00:00	 6 992	537 846	-1,47	1. ... Kb4 2.Kc1 Rd6 3.Qe4+ Nc4 4.Qe7

How about without 511 endgame tables?

4 times faster without EGTB : 0.001 sec
But who doesn't use EGTB in 2020 ???

Code: Select all

2020-08-13 02:23:09,603<--1:readyok
2020-08-13 02:23:09,603<--1:info depth 1 seldepth 1 multipv 1 score cp 1 nodes 42 nps 42000 tbhits 0 time 1 pv c5b4
2020-08-13 02:23:09,612<--1:info depth 2 seldepth 2 multipv 1 score cp 13 nodes 66 nps 66000 tbhits 0 time 1 pv c5b4 f5f6
2020-08-13 02:23:09,614<--1:info depth 3 seldepth 4 multipv 1 score cp 28 nodes 203 nps 203000 tbhits 0 time 1 pv c5b4 f5f6 e5c4 d2c1 d3d2 c1d1
2020-08-13 02:23:09,617<--1:info depth 4 seldepth 6 multipv 1 score cp 166 nodes 269 nps 269000 tbhits 0 time 1 pv c5b4 f5f6
2020-08-13 02:23:09,621<--1:info depth 5 seldepth 9 multipv 1 score cp 92 nodes 936 nps 468000 tbhits 0 time 2 pv c5b4 d2d1 d7d6 f5f4 b4b3 f4d2
2020-08-13 02:23:09,625<--1:info depth 6 seldepth 8 multipv 1 score cp 146 nodes 1369 nps 684500 tbhits 0 time 2 pv c5b4 f5f4 e5c4 d2c1 d7e7 f4b8 b4c3
2020-08-13 02:23:09,628<--1:info depth 7 seldepth 10 multipv 1 score cp 119 nodes 3867 nps 1289000 tbhits 0 time 3 pv c5b4 d2e1 d7d6 f5c8 e5c4 e1d1 b4b3 c8c5
2020-08-13 02:23:09,632<--1:info depth 8 seldepth 13 multipv 1 score cp 105 nodes 8550 nps 1425000 tbhits 0 time 6 pv c5b4 d2e1 d7d6 f5f4 e5c4 e1d1 b4b3 f4c1 f6f5 c1b1 b3c3 b1c1 c3d4
2020-08-13 02:23:09,636<--1:info depth 9 seldepth 20 multipv 1 score cp 178 nodes 13017 nps 1627125 tbhits 0 time 8 pv c5b4 f5f4 e5c4 d2d1 d3d2 f4e4 f6f5 e4b1 b4c3 b1c2 c3d4
2020-08-13 02:23:09,641<--1:info depth 10 seldepth 17 multipv 1 score cp 213 nodes 19691 nps 1640916 tbhits 0 time 12 pv c5b4 d2e1 d7d6 f5e4 e5c4 e4b7 b4c3 b7c7 d3d2 e1d1
2020-08-13 02:23:09,646<--1:info depth 11 seldepth 20 multipv 1 score cp 289 nodes 27131 nps 1695687 tbhits 0 time 16 pv c5b4 d2e1 d7d6 f5e4 e5c4 e4b7 b4c3 b7f3 d6e6 e1f2
2020-08-13 02:23:09,679<--1:info depth 12 seldepth 29 multipv 1 score cp 196 nodes 151953 nps 1973415 tbhits 0 time 77 pv c5b4 d2d1 d7d6 f5f4 e5c4 f4c1 d3d2 c1b1 b4c3 b1c2 c3d4 c2a4 d6e6
2020-08-13 02:23:09,697<--1:info depth 13 seldepth 27 multipv 1 score cp 93 nodes 189027 nps 1989757 tbhits 0 time 95 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d6 e4b7 b3c3 b7g7 d6d4 g7g3 d4e4 d1c1 e4e5 c1b1 e5b5 b1a1 b5f5 a1b1 f5e5 g3g7
2020-08-13 02:23:09,704<--1:info depth 14 seldepth 30 multipv 1 score cp 166 nodes 196656 nps 1966560 tbhits 0 time 100 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d6 e4b7 b3c3 b7g7 d6d4 g7g3 d4e4 d1c1 e4e5 c1b1 e5e7 g3g2
2020-08-13 02:23:09,709<--1:info depth 15 seldepth 31 multipv 1 score cp 205 nodes 213138 nps 1973500 tbhits 0 time 108 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d6 e4b7 b3c3 b7g7 d6d4 g7g3 d4e4 g3g7 e4e5
2020-08-13 02:23:09,796<--1:info depth 16 seldepth 34 multipv 1 score cp 157 nodes 387408 nps 1996948 tbhits 0 time 194 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d6 e4f3 b3c3 d1e1 d6e6 e1f2 e6e5 f3h3 e5e3 h3h8 c3b3 h8b8 b3c2 b8c8 e3e2 f2g1 e2e4 g1f2 d3d2 c8f5 d2d1q
2020-08-13 02:23:09,824<--1:info depth 17 seldepth 38 multipv 1 score cp 172 nodes 440334 nps 1983486 tbhits 0 time 222 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d6 e4f3 b3c3 d1e1 d6e6 e1f2 e6e5 f3h3 e5e2 f2g1 e2e3 h3h8 c3c2 h8c8 e3e1 g1f2 e1e2 f2g1 c2b3 c8b8 b3c3 b8g3
2020-08-13 02:23:09,886<--1:info depth 18 seldepth 43 multipv 1 score cp 179 nodes 572403 nps 2015503 tbhits 0 time 284 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d6 e4f3 b3c3 d1e1 d6e6 e1f2 e6e3 f3f6 c3b3 f6f5 e3e2 f2g3 b3c3 f5f1 e2e3 g3f4 d3d2 f1a1 c3c2 a1a4 c2d3 a4d7 d3c3 d7g4
2020-08-13 02:23:09,904<--1:info depth 19 seldepth 40 multipv 1 score cp 217 nodes 612895 nps 2029453 tbhits 0 time 302 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d6 e4f3 b3c3 d1e1 d6e6 e1f2 e6e3 f3f6 c3b3 f6f5 e3e2 f2g3 b3c3 f5f1 e2e3 g3f4 d3d2 f1a1 c3c2 a1a4 c2d3 a4d7 d3c3 d7g4
2020-08-13 02:23:10,026<--1:info depth 20 seldepth 50 multipv 1 score cp 337 nodes 875201 nps 2064153 tbhits 0 time 424 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d6 e4f3 b3c3 d1e1 d6e6 e1f2 e6e3 f3f6 c3b3 f6f5 e3e2 f2g3 b3c3 f5f1 e2e3 g3g4 d3d2 g4f5 e3e5 f5g6 e5e1 f1f3 e1e3
2020-08-13 02:23:10,207<--1:info depth 21 seldepth 46 multipv 1 score cp 390 nodes 1279879 nps 2115502 tbhits 0 time 605 pv c5b4 d2d1 d7d6 f5e4 e5c4 e4e1 b4b3 e1g3 d6d5 g3f3 b3c3 d1e1 d5e5 e1f2 e5e2 f3e2 d3e2 f2e2 c3d4 e2f3 d4e5 f3f2 e5f4 f2f1 f4e3 f1e1 f6f5
2020-08-13 02:23:10,415<--1:info depth 22 seldepth 54 multipv 1 score cp 518 nodes 1756651 nps 2160702 tbhits 0 time 813 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d6 e4f3 b3c3 d1e1 d6e6 e1f2 e6e3 f3f6 c3b3 f6d4 b3b4 d4a1 d3d2 a1b1 b4c5 b1a1 e3e5 a1d1 c5b4 d1f3 e5e1 f3b7 b4c3 b7f3 e1e3 f3d1 c4b2 f2e3
2020-08-13 02:23:10,564<--1:info depth 23 seldepth 43 multipv 1 score cp 571 nodes 2101171 nps 2184169 tbhits 0 time 962 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d6 e4f3 b3c3 d1e1 d6e6 e1f2 e6e3 f3f6 c3b3 f6d8 e3e2 f2f1 b3c3 d8a8 e2e3 a8h8 e3e5 h8h3 c4e3 f1f2 d3d2 h3g3 e5e4 g3f3
2020-08-13 02:23:10,774<--1:info depth 24 seldepth 53 multipv 1 score cp 594 nodes 2589486 nps 2207575 hashfull 1 tbhits 0 time 1173 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d6 e4f3 b3c3 d1e1 d6e6 e1f2 e6e3 f3f6 c3b3 f6d8 e3e2 f2f1 b3c3 d8a8 e2e3 a8a1 c3b4 a1d1 d3d2 f1f2 e3e7 d1b1 b4c5 b1a1 e7e6 a1d1 e6e5 d1a4 e5e1 a4a7 c5b4 a7b7 b4c3 b7f3 e1e3 f3d1
2020-08-13 02:23:11,352<--1:info depth 25 seldepth 49 multipv 1 score cp 147 nodes 3878816 nps 2216466 hashfull 3 tbhits 0 time 1750 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d6 e4f3 b3c3 d1e1 d6e6 e1f2 e6e5 f3h3 e5e3 h3h8 c3c2 h8d4 c2b3 d4d8 e3e2 f2g1 b3c3 d8h8 e2e5 h8h3 c3c2 h3h7 e5e3 g1f2 e3e6 h7d7 e6e2 f2f1 c2c3 d7g7 e2e5 g7g3 e5e7 g3b8
2020-08-13 02:23:11,566<--1:info depth 26 seldepth 60 multipv 1 score cp 138 nodes 4339811 nps 2208555 hashfull 3 tbhits 0 time 1965 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d6 e4f3 b3c3 d1e1 d6e6 e1f2 e6e5 f3h3 e5e2 f2g1 e2e3 h3h8 c3c2 h8g8 e3e1 g1f2 e1e4 g8h7 e4e2 f2g1 e2e5 h7h4 c4b2 h4g3 d3d2 g3e5 d2d1q g1h2 b2d3
2020-08-13 02:23:11,729<--1:info depth 27 seldepth 53 multipv 1 score cp 138 nodes 4711706 nps 2215188 hashfull 3 tbhits 0 time 2127 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d6 e4f3 b3c3 d1e1 d6a6 e1f2 a6a2 f2g1 a2a1 g1h2 a1e1 f3g3 e1e3 g3g7 c3c2 g7c7 e3e4 c7h7 e4e5 h2g1 c2c3 h7h3 e5e3 h3h8 c3c2 h8g8 e3e1 g1f2 e1e4 g8h7 e4e3
2020-08-13 02:23:12,048<--1:info depth 28 seldepth 63 multipv 1 score cp 168 nodes 5426964 nps 2217803 hashfull 4 tbhits 0 time 2447 pv c5b4 f5f6 e5c4 d2d1 d7d5 f6e7 b4b3 e7e4 d5d6 e4f3 b3c3 d1e1 d6a6 e1f2 a6a2 f2g1 a2a1 g1h2 a1e1 f3g3 e1e3 g3g7 c3c2 g7h7 e3e2 h2g1 e2e1 g1f2 e1e5 h7g6 c2c3 g6g3 e5e7 f2g1 c4e3 g3d6 d3d2 d6a3 c3c2 a3c5 c2d3 c5d6 d3e2 d6e7 d2d1q g1h2
2020-08-13 02:23:12,624<--1:info depth 29 seldepth 63 multipv 1 score cp 174 lowerbound nodes 6699511 nps 2216912 hashfull 5 tbhits 0 time 3022 pv c5b4
2020-08-13 02:23:12,632<--1:info depth 27 currmove c5b4 currmovenumber 1
2020-08-13 02:23:12,681<--1:info depth 29 seldepth 63 multipv 1 score cp 209 lowerbound nodes 6828558 nps 2217784 hashfull 5 tbhits 0 time 3079 pv c5b4

[Kieren Pearson]

To be fair on Myrddin, neither Halogen 5 (which I presume played in this match) nor Halogen 6 nor even my development version (+20 elo higher again) can find the winning move. All versions favour Rd6 with a score of about +1.10.

What kind of search techniques would allow an engine to spot this move? It might be time to try and implement EGTB in my engine.