Not logged inRybka Chess Community Forum
Up Topic The Rybka Lounge / Computer Chess / 300 game blitz match: Lc0 162.0 - Stockfish 138.0
- By MrKris (**) Date 2019-09-02 23:40
Stockfish maybe did well considering the Leela Ratio.

G/5'+2"  i7-4770K 8HT - RTX 2060  Leela Ratio ~1.8
1GBhash Sy.6 ; Net 42862 @3332 http://lczero.org/networks/
Openings: 150 Hert lowdraws reversed (sp-cc.de)
                                   Elo   +  -      Games     Score  Draws
1 Lc0-0.22.0_42862      162.0/300 : 28  18 17  (+42 =240 -18) 54%   80.0%
2 Stockfish_19082608c0  138.0/300 :  0  17 18  (+18 =240 -42) 46%   80.0%


Lc0's shortest win, only 24 moves (resign @ 3 moves 5.50).
17.O-O-O! Unusual offer of the a-pawn to 17...Nxa2+, best to retain the advantage.
19.Bxh7+! Classic bishop sacrifice.
21.Bxg7!! Classic 2nd bishop follow-up also unprotecting its own N on d6. [Event "Lc0-42862_Sf190826"]
[Site "?"]
[Date "2019.08.31"]
[Round "11"]
[White "Lc0-0.22.0_42862"]
[Black "Stockfish_19082608c0"]
[Result "1-0"]
[BlackElo "2200"]
[ECO "B45"]
[Opening "Sicilian"]
[Variation "Taimanov, Four Knights, 6.Nxc6 bxc6 7.e5 Nd5 8.Ne4 Qc7"]
[WhiteElo "2400"]
[TimeControl "300+2"]
[Termination "normal"]
[PlyCount "48"]
[WhiteType "human"]
[BlackType "program"]

1. e4 {book} c5 {book} 2. Nf3 {book} e6 {book} 3. d4 {book} cxd4 {book} 4.
Nxd4 {book} Nf6 {book} 5. Nc3 {book} Nc6 {book} 6. Nxc6 {book} bxc6 {book}
7. e5 {book} Nd5 {book} 8. Ne4 {+0.44/19 4.0s} Qc7 {-0.09/23 10s} 9. f4
{+0.50/20 4.3s} Qa5+ {-0.69/28 50s} 10. Bd2 {+0.56/15 9.1s} Qb6 {-0.74/26
3.2s} 11. Bd3 {+0.48/15 11s} Be7 {-0.61/26 4.8s} 12. Qe2 {+0.67/15 7.8s} f5
{-0.49/26 4.1s} 13. exf6 {+0.64/18 5.5s} Nxf6 {-0.55/26 5.2s} 14. Bc3
{+0.67/19 5.1s} Nd5 {-0.43/27 9.9s} 15. Be5 {+0.71/19 15s} O-O {0.00/22
3.3s} 16. Rf1 {+0.72/15 0.25s} Nb4 {-0.47/27 12s} 17. O-O-O {+1.02/19 12s
!} d5 {-0.46/27 3.3s Lc0 best, -0.96 at 2 mins., with:} (17. .. Nxa2+ {?
-2.10} 18. Kb1 Nb4 19. Qg4) (17. .. Rf7 {?! -1.14} 18. Rf3) (17. .. Rf5 {?!
-1.18} 18. Rf3) (17. .. Nxd3+ {?! -1.63} 18. Qxd3 Ba6 19. Qg3 g6 20. Rfe1)
18. Nd6 {+1.08/23 4.7s} c5 {-0.39/28 4.0s ?? 5th worst of PV6 2 mins.:
18...Bxd6, Nxa2+, Qd8, Bf6, , c5, Nxd3+ -0.95 to -1.51} 19. Bxh7+ {+1.91/23
13s !} Kxh7 {-3.73/30 12s} 20. Qh5+ {+2.13/23 6.0s} Kg8 {-3.73/1 0s} 21.
Bxg7 {+2.23/23 4.2s} Bxd6 {-4.89/29 8.9s} 22. Qg6 {+2.36/21 7.4s} Rf7
{-6.39/29 14s} 23. Bh6+ {+2.61/17 11s} Kh8 {-5.61/28 4.9s} 24. Rf3
{+2.81/15 5.7s} Nxa2+ {-9.29/28 12s, White wins by adjudication} 1-0

Round 19, game #19, Lc0 1-0 Stockfish after 15.Rhe1 Rb8 -0.19 16.f5 +2.37 b4?? 0.00 17.f6 -diagram- Stockfish played 17...g6 with its evaluation dropping all the way to -9.18 +-.
1rb2rk1/2qn1ppp/p3pP2/3pP3/1p1Q4/2NB4/PPP3PP/2KRR3 b - - 0 17


Here Stockfish's massive-looking kingside attack loses to Lc0's counter attack in the center and queenside.
24...Qb5!! winning the b2 pawn and realizing its king is safe enough. Stockfish again suddenly got the point: 23.Rh3 -0.05 20 secs., 24.Kh1 0.00, 25.Bh4 -2.38 43 secs.
28...Qxa1! temporary queen sacrifice, the tactics do not end until 8 more moves.
33...Rbc8 white won the knight on c6 but there is a skewer.
38...Rxe8 the tactics end with black having a won bishop vs. pawn ending. [Event "Lc0-42862_Sf190826"]
[Site "?"]
[Date "2019.09.01"]
[Round "57"]
[White "Stockfish_19082608c0"]
[Black "Lc0-0.22.0_42862"]
[Result "0-1"]
[ECO "B24"]
[Opening "Sicilian"]
[Variation "Closed, 3.g3 g6 4.Bg2 Bg7 5.f4"]
[TimeControl "300+2"]
[Termination "adjudication"]
[PlyCount "79"]
[WhiteType "human"]
[BlackType "human"]

1. e4 {book} c5 {book} 2. Nc3 {book} Nc6 {book} 3. f4 {book} g6 {book} 4.
g3 {0.00/28 25s} Bg7 {+0.11/10 8.3s} 5. Bg2 {-0.15/26 5.7s} d6 {+0.10/10
5.5s} 6. Nf3 {-0.03/24 5.3s} e6 {+0.10/10 7.4s} 7. O-O {-0.06/24 4.1s} Nge7
{+0.14/10 4.6s} 8. Ne2 {0.00/27 25s} O-O {+0.21/10 6.6s} 9. c3 {+0.06/23
2.8s} b5 {+0.23/12 6.7s} 10. d4 {+0.03/25 6.1s} b4 {+0.27/15 7.8s} 11. Be3
{0.00/29 21s} Ba6 {+0.27/15 11s} 12. Re1 {-0.24/28 9.0s} Qa5 {+0.35/14
6.8s} 13. dxc5 {-0.07/25 12s} dxc5 {+0.56/13 7.6s} 14. Nd2 {-0.15/27 9.6s}
Rfd8 {+0.50/15 7.6s} 15. e5 {-0.32/26 8.3s} bxc3 {+0.46/18 12s} 16. Nxc3
{-0.39/27 17s} Rac8 {+0.48/19 1.8s} 17. a3 {-0.41/27 9.9s} Nf5 {+0.70/14
9.9s} 18. Bf2 {0.00/22 4.4s} Bf8 {+0.72/16 5.3s} 19. g4 {0.00/26 12s} Nfd4
{+0.95/15 11s} 20. Nce4 {0.00/24 2.7s} Bd3 {+0.99/18 5.5s} 21. Re3
{+0.20/21 3.5s} c4 {+1.02/14 10.0s} 22. Nf6+ {-0.21/26 9.9s} Kh8 {+0.87/18
11s} 23. Rh3 {-0.05/26 20s} h6 {+1.27/15 8.7s} 24. Kh1 {0.00/27 6.3s} Qb5
{+1.27/16 5.4s} 25. Bh4 {-2.38/29 43s} Qxb2 {+1.65/18 8.8s} 26. Bg5
{-2.96/30 16s} c3 {+1.77/19 4.1s} 27. Nde4 {-3.22/30 9.1s} c2 {+1.83/18
12s} 28. Qg1 {-3.23/30 2.4s} Qxa1 {+1.85/18 10s} 29. Qxa1 {-3.24/30 3.7s}
Rb8 {+1.86/19 0.22s} 30. Nc3 {-3.38/31 12s} Nb3 {+2.33/15 13s} 31. Qb2
{-3.12/32 3.3s} c1=Q+ {+2.61/15 9.3s} 32. Qxc1 {-3.13/34 2.0s} Nxc1
{+2.66/14 3.0s} 33. Bxc6 {-4.79/34 21s} Rbc8 {+2.76/13 6.2s} 34. Bxh6
{-4.89/28 2.8s} Bxh6 {+2.84/14 7.1s} 35. Rxh6+ {-5.00/31 4.9s} Kg7
{+2.75/11 0.010s} 36. Rh7+ {-5.01/29 1.9s} Kf8 {+2.67/11 0.026s} 37. Be8
{-4.89/30 11s} Rxe8 {+4.81/9 21s} 38. Nxe8 {-5.84/31 8.5s} Rxe8 {+9.26/9
33s} 39. Rh3 {-5.63/24 1.0s} Kg7 {+12.01/7 9.2s} 40. Re3 {-5.55/23 1.3s,
Black wins by adjudication} 0-1
Attachment: Lc0-42862_Sf190826_.pgn - Games (928k)
Notes:
Leela Ratio 850/(SfN/LcN) = (LcN/SfN)*850
init. pos.: 5 secs. 0.59 NO, 1.7 (for TC G/5'+2")
10 secs. 0.55 NO 1.82
FEN: 2bq1rk1/p1pp1ppp/2n2b2/1r2p3/8/P3P1P1/1P1PNPBP/R1BQK2R w KQ - 0 11
Move 11: 5 secs. 19k - sf 8209k = 1.97 (for TC G/5'+2")
10 sec. 20k - sf 8194k = 2.1

Part 1: Lc0 - Sf +30 Elo +/- 26; +20 -8 =110  54.3%. 138 games.
Arena 3.5.1: drawscore 0.00 (unchangable) set at move 40, 3 moves.
--Lc0 very rarely says 0.00, too many 150-200+ move draws.

Part 2: Score of lc0-v0.22.0_42862 vs Stockfish_19082608c0: 22 - 10 - 130 [0.537]
ELO difference: 25.78 +/- 23.58   162 of 162 games finished.
Cute Chess GUI: drawscore 0.16 set at move 50, 1 move,
Lc0 typically 0.25-0.65 in very drawish positions.
- By MrKris (**) Date 2019-09-02 23:52
19.Bf6!! bishop offer, accepting is bad and the only way to keep it attacking g7; there does not seem to be a defense.
22.g6!! no need to give black any chance.
25.Rxh6!! forcing through 30.Qxh6 when white's Q+N+f-pawn continue the winning attack vs. black's cramped R+R+B. [Event "Lc0-42862_Sf190826"]
[Site "?"]
[Date "2019.08.31"]
[Round "54"]
[White "Lc0-0.22.0_42862"]
[Black "Stockfish_19082608c0"]
[Result "1-0"]
[BlackElo "2400"]
[ECO "B43"]
[Opening "Sicilian"]
[Variation "Kan, 5.Nc3 Qc7 6.Bd3 Nf6"]
[WhiteElo "2400"]
[TimeControl "300+2"]
[Termination "normal"]
[PlyCount "80"]
[WhiteType "human"]
[BlackType "human"]

1. e4 {book} c5 {book} 2. Nf3 {book} e6 {book} 3. d4 {book} cxd4 {book} 4.
Nxd4 {book} a6 {book} 5. Nc3 {book} Qc7 {book} 6. Bd3 {book} Nf6 {-0.17/25
45s} 7. f4 {+0.43/12 5.6s} d6 {-0.22/23 4.2s} 8. Qe2 {+0.47/12 5.8s} Be7
{-0.37/26 32s} 9. g4 {+0.39/13 11s} Nc6 {-0.32/20 3.0s} 10. Nxc6 {+0.50/13
5.3s} bxc6 {-0.38/23 5.6s} 11. g5 {+0.50/13 15s} Nd7 {-0.23/21 2.8s} 12.
Be3 {+0.53/14 0.78s} Rb8 {-0.49/26 26s} 13. O-O-O {+0.48/14 3.8s} O-O
{-0.32/24 7.1s} 14. Kb1 {+0.60/13 14s} d5 {0.00/21 2.1s} 15. h4 {+0.65/13
3.4s} Nc5 {-0.13/26 17s} 16. b3 {+0.60/14 6.5s} Rd8 {0.00/24 6.3s} 17. Bd4
{+0.81/15 10s} Nxd3 {-0.01/22 2.8s} 18. Rxd3 {+0.85/16 8.1s} Bd6 {0.00/25
4.1s} 19. Bf6 {+1.68/12 11s !!} Re8 {-0.63/25 17s} (19. .. Be7 {? Lc0 1
min.:} 20. Bxe7 Qxe7 21. exd5 cxd5 22. Nxd5 {Just a pawn up but a much
better position, Nxf6+ is also threatened, so:} Rxd5 23. Rxd5) 20. e5
{+1.82/12 9.3s} Bf8 {-0.61/24 3.1s} 21. Rg3 {+1.76/14 6.3s} a5 {-1.91/28
38s ; Lc0 PV6 2 mins., alternatives are all the same, 0.04 'better' max.:}
(21. .. Rb7 22. h5 Qb6 23. g6) (21. .. Rb4 22. Qf3 Qa7 23. h5 Rb7 24. g6)
(21. .. Qb6 22. h5 Rb7 23. g6) (21. .. Kh8 22. g6) (21. .. Qa7 22. g6) 22.
g6 {+2.00/12 12s !!} fxg6 {-1.95/23 2.6s relatively best, Lc0:} (22. .. h6
{-3.10} 23. gxf7+ Qxf7 24. Rhg1 Rb7 25. Bxg7 Bxg7 26. Rxg7+ Qxg7 27. Rxg7+
Rxg7 28. Qe3 Rf8 29. Kb2 Bd7 (29. .. Rg4 {?!} 30. Na4 Rfxf4 31. Qc5 Rg2 32.
Qxc6 Rff2 33. Qxc8+ Kg7 34. Qc7+ Kg6 35. Qc6 Kf5 36. Qc3 d4 37. Qd3+ Kxe5
38. Nb6 {+5.79 1 min.}) 30. Na4 Bc8 31. Nc5) (22. .. hxg6 {-3.35} 23. h5
gxf6 24. hxg6 Bg7 25. exf6 Bxf6 26. gxf7+ Kxf7 27. Rh7+) (22. .. gxf6 {?}
23. gxh7+ Kxh7 24. Qh5+ Bh6 25. Rhg1 Ba6 26. Rg7+ Kh8 27. Qxh6#) (22. ..
Kh8 {?} 23. gxh7 gxf6 24. Rhg1 Bh6 25. Rg8+ Kxh7 26. Rxe8) (22. .. a4 {?}
23. gxh7+ Kh8 24. Rhg1 axb3 25. cxb3 Ba6 26. Qh5 Bd3+ 27. Rxd3) 23. h5
{+2.04/12 4.3s} Rb7 {-1.81/26 9.7s} 24. hxg6 {+2.43/16 12s} h6 {-2.01/28
5.1s} 25. Rxh6 {+2.47/17 6.8s !!} gxh6 {-2.69/27 9.6s} (25. .. gxf6 {??}
26. g7 Bxg7 (26. .. Qxg7 27. Qh5 Qxg3 28. Rh8+ Kg7 29. Qh7#) 27. Qh5 Qf7
28. Rh8#) 26. g7 {+2.46/16 0.41s} Bxg7 {-2.90/27 5.7s} 27. Rxg7+ {+2.54/15
12s} Qxg7 {-3.14/28 7.8s} 28. Bxg7 {+2.55/13 0.28s} Rxg7 {-2.83/27 28s} 29.
Qh5 {+2.67/13 10.0s} Bd7 {-3.46/27 14s} 30. Qxh6 {+2.60/14 11s} Ree7
{-3.69/28 17s} 31. Ne2 {+2.63/12 25s} Rh7 {-4.28/26 10s} 32. Qg5+ {+2.70/12
9.7s} Rhg7 {-4.22/23 1.2s} 33. Qh4 {+2.75/11 4.1s} c5 {-4.46/27 14s} 34.
Kb2 {+3.03/11 13s} a4 {-4.30/26 11s} 35. Ng3 {+3.13/12 5.5s} Be8 {-4.85/27
9.9s} 36. f5 {+3.13/12 5.6s} Rh7 {-5.19/24 4.5s} 37. Qf6 {+3.11/12 22s}
Rhg7 {-5.44/25 3.8s} 38. fxe6 {+3.29/11 2.9s} Bg6 {-7.03/23 2.1s} 39. Ne2
{+3.10/10 7.9s} a3+ {-7.15/23 2.0s} 40. Kc1 {+3.26/11 7.5s} Re8 {-7.85/23
2.0s, White wins by adjudication} 1-0

Stockfish gives a good lesson in classic, or Tal-Fischer for inventor-practitioner, minor piece sacrifices in the Sicilian.
11.e5! a pawn sacrifice, about +/ had black played better. The human game reference in the viewer follows this game through 14.Bxf6 and had a similar 16.Nd5 knight sacrifice.
15.Nd5!! wins already. Extremely complicated, some variations requiring a follow-up sacrifice, but black's lack of development and centralized king should lose. [Event "Lc0-42862_Sf190826"]
[Site "?"]
[Date "2019.08.31"]
[Round "45"]
[White "Stockfish_19082608c0"]
[Black "Lc0-0.22.0_42862"]
[Result "1-0"]
[BlackElo "2400"]
[ECO "B94"]
[Opening "Sicilian"]
[Variation "Najdorf, 6.Bg5 Nbd7"]
[WhiteElo "2400"]
[TimeControl "300+2"]
[Termination "normal"]
[PlyCount "44"]
[WhiteType "human"]
[BlackType "human"]

1. e4 {book} c5 {book} 2. Nf3 {book} d6 {book} 3. d4 {book} cxd4 {book} 4.
Nxd4 {book} Nf6 {book} 5. Nc3 {book} a6 {book} 6. Bg5 {book} Nbd7 {book} 7.
Qe2 {+0.05/27 29s} h6 {-0.20/11 5.1s} 8. Bh4 {+0.60/25 7.3s} Qc7 {-0.24/14
9.5s} 9. O-O-O {+0.61/22 2.9s} e6 {-0.25/15 11s} 10. f4 {+0.96/23 4.2s} b5
{-0.25/14 0.22s} 11. e5 {+1.72/27 4.0s} dxe5 {-0.72/21 7.7s} 12. fxe5
{+2.18/29 5.7s} Qxe5 {-0.87/19 5.4s ?} (12. .. Nxe5 {Sf: best, 2 mins. PV3
-1.93 showing white winning the exchange for a pawn} 13. Bg3 Nfd7 14. Nf3
f6 15. Nxe5 Nxe5 16. Qh5+ Qf7 17. Be2 Qxh5 18. Bxh5+ Nf7 19. Bf3 Ra7 20.
Bc6+ Ke7 21. Bf2 Ne5 22. Bxa7 Nxc6 23. Bc5+) 13. Qf3 {+2.36/27 4.5s} Qb8
{-1.18/16 3.3s} 14. Bxf6 {+2.63/28 6.1s} gxf6 {-1.97/17 23s} (14. .. Nxf6
{Stockfish has this about .50 better.} 15. Nc6 Bb7 16. Nd5 {!} Qc8 (16. ..
Nxd5 17. Nxb8 Rxb8 18. Qg3 {European Individual Chess Championship
2019-03-28, Skopie, Csiki E. 2251 1-0 Drnovsek G. 2402, 25 moves; Stockfish
+4.36 in 32 secs.}) 17. Nxf6+ gxf6 18. Qxf6) (14. .. Bb7 {?} 15. Ne4 gxf6
16. Nxe6 {!} fxe6 17. Nxf6+ Nxf6 18. Qxf6) 15. Nd5 {+3.01/25 4.2s !!} Bg7
{-0.94/17 6.9s} (15. .. Bb7 {?!} 16. Nxe6 {!} Bd6 17. Ng7+ Kf8 18. Nf5 Bc5
19. Qg4 Rh7 20. Nde7 Bxe7 (20. .. Ke8 21. Bxb5 {!} axb5 22. Rhe1 Qe5 23.
Rxe5) 21. Nxe7) 16. Nc6 {+5.10/26 4.2s} Qb7 {-1.07/19 5.1s} 17. Nxf6+
{+5.82/26 3.6s} Kf8 {-3.82/10 12s ?!} (17. .. Bxf6 {About -6.62, Sf.} 18.
Qxf6 Qxc6 19. Qxh8+ Nf8 20. Qxh6) 18. Nh5 {+10.05/23 5.0s} Qc7 {-4.10/10
8.6s ? but better are all in the -11.00 range, Sf 1 min.:} (18. .. a5 19.
Bd3 Ne5 20. Be4 Nxc6 21. Bxc6 Qe7 22. Bxa8) (18. .. Ra7 19. Be2 Kg8 20. Qg4
Rh7 21. Bd3 f5 22. Bxf5 exf5 23. Ne7+ Kh8 24. Nf6 {!} Nf8 25. Ng6+ Nxg6 26.
Rd8+ Bf8 27. Rxf8+ Kg7 28. Rg8+ Kxf6 29. Qxg6+ Ke5 30. Re8+ Re7 31. Qg7+
Ke6 32. Re1+ Qe4 33. Qxh6+ Ke5 34. Rxe4+ Kxe4 35. Rxe7+ Rxe7 36. Qh4+) (18.
.. Be5 19. Be2 Rg8 20. Rhf1 f5 21. g4 Rxg4 22. Nxe5 Qxf3 23. Nxd7+ Bxd7 24.
Bxf3) 19. Qa3+ {+20.84/29 5.0s} Kg8 {-5.23/11 13s} 20. Ne7+ {+23.19/26
4.2s} Kf8 {-6.24/9 9.7s} 21. Nxc8+ {+25.51/26 4.1s} b4 {-7.09/8 9.9s} 22.
Qxb4+ {+32.99/27 4.0s} Nc5 {-9.30/7 12s, White wins by adjudication} 1-0

Another Stockfish Sicilian lesson.
17...Rg8?? lost at once to 18.Nd5!! a 'classic' varient offering white's Q (if taken there is a quick mate), the knight can not be accepted however.
Then a 'classic' 20.Nxe6!! with a mate threat if not accepted. [Event "Lc0-42862_Sf190826"]
[Site "?"]
[Date "2019.08.31"]
[Round "26"]
[White "Stockfish_19082608c0"]
[Black "Lc0-0.22.0_42862"]
[Result "1-0"]
[BlackElo "2400"]
[ECO "B97"]
[Opening "Sicilian"]
[Variation "Najdorf, Poisoned Pawn, 10.e5 dxe5"]
[WhiteElo "2400"]
[TimeControl "300+2"]
[Termination "normal"]
[PlyCount "72"]
[WhiteType "human"]
[BlackType "human"]

1. e4 {book} c5 {book} 2. Nf3 {book} d6 {book} 3. d4 {book} cxd4 {book} 4.
Nxd4 {book} Nf6 {book} 5. Nc3 {book} a6 {book} 6. Bg5 {book} e6 {book} 7.
f4 {book} Qb6 {book} 8. Qd2 {book} Qxb2 {book} 9. Rb1 {book} Qa3 {book} 10.
e5 {book} dxe5 {book} 11. fxe5 {book} h6 {book} 12. Bh4 {book} g5 {book}
13. exf6 {book} gxh4 {book} 14. Be2 {-0.41/24 11s} Qa5 {-0.07/12 5.1s} 15.
O-O {-0.41/26 11s} h3 {-0.20/14 6.2s} 16. Bf3 {+0.38/23 4.4s} Nd7 {-0.10/14
8.6s} 17. Kh1 {+0.42/25 4.9s} (17. Nd5 {? Stockfish has as 0.00, for
example:} Qc5 18. Rbe1 Bd6 19. Kh1 Ne5 20. Nb3 Qc6 21. Na5 Qc5 22. Nb3) 17.
.. Rg8 {+0.11/18 12s ??} (17. .. Ne5 {would have been 0.00/40 2.5 mins., a
whole different game.}) 18. Nd5 {+5.01/25 4.6s !!} Qc5 {-1.89/14 9.6s} (18.
.. Qxd2 {??} 19. Nc7+ Kd8 20. Ncxe6+ fxe6 (20. .. Ke8 21. Nc7+ Kd8 22.
Nde6+ fxe6 23. Nxe6+ Ke8 24. Bh5+ Rg6 25. Bxg6#) 21. Nxe6+ Ke8 22. Bh5+ Rg6
23. Bxg6#) 19. Rbe1 {+5.36/24 4.3s} Rg6 {-2.46/15 17s} 20. Nxe6 {+6.44/29
6.7s !!} fxe6 {-2.48/19 2.6s} 21. Rxe6+ {+6.58/27 5.7s} Kf7 {-2.63/16 6.2s}
22. Qe2 {+7.06/29 4.4s} b5 {-2.53/12 17s} 23. Nf4 {+7.22/28 4.7s} Rg5
{-3.13/11 13s} 24. Bxa8 {+7.28/28 4.9s} Ne5 {-3.29/11 7.3s} 25. Nd3
{+7.76/30 6.9s} Nxd3 {-3.63/11 15s} 26. Re7+ {+8.43/30 5.2s} Kg8 {-3.97/11
12s} 27. Qxd3 {+9.04/30 5.8s} Bxe7 {-3.28/13 11s} 28. Qb3+ {+9.77/29 6.3s}
Kh8 {-3.66/13 6.5s} 29. fxe7 {+10.48/30 6.6s} Qxe7 {-4.24/13 15s} 30. Qc3+
{+10.82/29 8.0s} Kh7 {-4.56/11 4.3s} 31. Qxc8 {+11.08/30 6.8s} hxg2+
{-5.02/10 9.6s} 32. Bxg2 {+12.48/25 6.2s} Rxg2 {-4.69/10 6.1s} 33. Kxg2
{+56.37/26 5.5s} Qg5+ {-5.10/10 13s} 34. Kf3 {+148.91/40 5.3s} Qd5+
{-5.96/10 20s} 35. Ke3 {+148.94/49 5.8s} Qe5+ {-6.48/8 7.3s} 36. Kd3
{+M47/49 52s} Qd5+ {-8.21/8 15s, White wins by adjudication} 1-0
- By MrKris (**) Date 2019-09-03 00:05
12.Qc2!? At FIDE 2500+ I found 94 games: 92 games with 12.O-O, 2 with 12.f3.
At under that I found 4 games with 12.Qc2!?, 3 had 12...Be7 1/2-1/2, 12...Nbd7 1-0 and 12...b4 0-1, ratings 2286 to 2578.
12...Qxd4 (?! or so probably) Only 1 of the above, the lowest rateds, had this: in 2008, De Rosa F. 1935 1-0 Pinto G. 2061, in 31 moves.
13...Qb6? allowed 14.Ng6!
Black's game was much worse, then 23...Qxa2? was lost.
However to win Stockfish had to find the exchange sacrifice 26.Rxd5, 2nd best would have lost almost all of the advantage. [Event "Lc0-42862_Sf190826"]
[Site "?"]
[Date "2019.08.29"]
[Round "42"]
[White "Stockfish_19082608c0"]
[Black "Lc0-0.22.0_42862"]
[Result "1-0"]
[BlackElo "2400"]
[ECO "D43"]
[Opening "Semi-Slav"]
[Time "19:53:32"]
[Variation "Moscow, 6.Bh4 Gambit, 7.e4"]
[WhiteElo "2400"]
[TimeControl "300+2"]
[Termination "normal"]
[PlyCount "70"]
[WhiteType "human"]
[BlackType "human"]

1. d4 Nf6 2. c4 e6 3. Nf3 d5 4. Nc3 c6 5. Bg5 h6 6. Bh4 dxc4 7. e4 g5 8.
Bg3 b5 9. Ne5 Bb7 10. h4 g4 11. Be2 h5 12. Qc2 {+0.40/25 25} Qxd4 {-0.48/13
5 ?!} (12. .. Bg7 {Stockfish -0.09/29, just 1 min.}) 13. Rd1 {+0.20/24 6}
Qb6 {-0.56/15 9 ?} (13. .. Qc5 {Stockfish -0.52/32 1 min.} 14. Ng6 {?? does
not work as it does in the game:} fxg6 15. e5 Nd5 16. Qxg6+ Kd7 17. Qf6
{and the black bishop on f8 is protected, not forked as in the next note.})
14. Ng6 {+1.18/24 5} Rg8 {-1.36/15 7} (14. .. fxg6 {??} 15. e5 {Stockfish
+3.00/34 1 min.; if} Nd5 16. Qxg6+ Kd7 (16. .. Ke7 {?} 17. Nxd5+ cxd5 18.
Qf6+) 17. Qf6) 15. Nxf8 {+1.14/24 5} Rxf8 {-1.36/18 5} 16. e5 {+1.50/26 8}
Nd5 {-1.49/17 5} 17. Ne4 {+1.73/23 3} Qa5+ {-1.42/17 7} 18. Rd2 {+1.79/23
4} Ba6 {-1.39/14 6} 19. Nd6+ {+2.08/26 19} Kd7 {-1.13/16 18} 20. O-O
{+2.13/25 2} Kc7 {-1.13/19} 21. Rfd1 {+2.23/27 5} Nd7 {-1.16/16 7} 22. Qh7
{+2.26/25 6} Nc5 {-1.18/15 5} 23. Qxh5 {+2.41/24 6} Qxa2 {-1.29/15 12 ?}
(23. .. Rad8 {Stockfish 1 min.}) 24. b4 {+4.00/26 4} Qa3 {-1.71/16 16} 25.
bxc5 {+5.44/24 5} c3 {-2.09/14 13} 26. Rxd5 {+6.61/25 4} (26. Rc2 {?
Stockfish's 2nd best only +1.21/33 1 min., not necessarily winning}) 26. ..
cxd5 {-2.17/11 3} 27. Qg5 {+6.80/26 7} Qxc5 {-2.88/11 17} 28. Qe7+
{+9.23/25 6} Kb6 {-2.66/11 12} 29. Ra1 {+9.53/27 6} c2 {-3.29/11 15} 30.
Bf4 {+13.73/29 8} Rfe8 {-3.76/10 17} 31. Rxa6+ {+M13/54 4} Kxa6 {-1.81/13}
32. Qb7+ {+M12/64 5} Ka5 {-2.06/10} 33. Bd2+ {+M11/67 4} Ka4 {-6.19/8 15}
34. Bxb5+ {+M10/70 4} Ka3 {-8.06/5 9} 35. Bc1+ {+M9/74 5} Kb4 {-8.86/5 7
Arena Adjudication} 1-0

Lc0 was worse when its 27...Bxe4?? lost immediately.
It allowed 29.Bc6!! a bishop sacrifice line clearing for 31.R(f2)-h2.
30.Qxh7+ temporary queen sacrifice, and why the R needed h2, the rest was essentially forced. [Event "Lc0-42862_Sf190826"]
[Site "?"]
[Date "2019.09.01"]
[Round "75"]
[White "Stockfish_19082608c0"]
[Black "Lc0-0.22.0_42862"]
[Result "1-0"]
[BlackElo "2400"]
[ECO "B43"]
[Opening "Sicilian"]
[Variation "Kan, 5.Nc3 Qc7 6.g3 Nf6"]
[WhiteElo "2400"]
[TimeControl "300+2"]
[Termination "normal"]
[PlyCount "78"]
[WhiteType "human"]
[BlackType "human"]

1. e4 {book} c5 {book} 2. Nf3 {book} e6 {book} 3. d4 {book} cxd4 {book} 4.
Nxd4 {book} a6 {book} 5. Nc3 {book} Qc7 {book} 6. g3 {book} Nf6 {-0.28/11
6.0s} 7. Bg2 {+0.15/26 15s} Be7 {-0.25/12 7.0s} 8. f4 {+0.28/21 2.5s} d6
{-0.33/14 9.0s} 9. g4 {+0.25/24 6.7s} Nc6 {-0.33/14 6.5s} 10. g5 {+0.77/20
4.0s} Nd7 {-0.38/15 2.8s} 11. Be3 {+0.69/22 5.4s} b5 {-0.40/13 10s} 12.
Nxc6 {+0.53/24 24s} Qxc6 {-0.43/14 2.7s} 13. h4 {+0.76/24 8.3s} b4
{-0.47/14 13s} 14. Ne2 {+1.00/19 2.5s} O-O {-0.46/14 17s} 15. O-O {+0.94/22
8.3s} Bb7 {-0.47/14 0.23s} 16. Nd4 {+1.17/26 21s} Qc4 {-0.44/12 13s} 17. a3
{+0.88/26 14s} Rfe8 {-0.47/14 7.8s} 18. axb4 {+0.92/24 3.8s} Qxb4 {-0.47/14
3.5s} 19. f5 {+0.89/26 23s} exf5 {-0.40/13 9.9s} 20. Nxf5 {+1.03/25 4.0s}
Bf8 {-0.35/13 3.8s} 21. Qg4 {+0.85/28 26s} Ne5 {-0.42/13 12s} 22. Qf4
{+0.60/29 6.6s} Rac8 {-0.40/16 17s} 23. Rf2 {+0.88/27 17s} Qxb2 {-0.35/13
7.1s} 24. Bd4 {+1.30/24 2.4s} Qb5 {-0.34/15 6.9s} 25. h5 {+0.53/26 14s} Rc7
{-0.39/17 5.9s} 26. h6 {+1.23/23 4.4s} Ng6 {-0.41/20 6.6s} 27. Qh2
{+1.38/28 20s} Bxe4 {+0.06/16 10s ??} (27. .. Re5 {Stockfish -1.96/37 3
mins.}) 28. hxg7 {+5.84/25 2.0s} Bxf5 {+0.06/15 0.39s} (28. .. Bxg7 {?} 29.
Nxg7 Re5 (29. .. Bxg2 30. Qxd6 Rc6 31. Qd7 Rf8 32. Nf5 Qxf5 (32. .. Ne5 33.
Ne7+ Kg7 34. Rxa6 {Mate in 12} Bh3 {Stockfish_19082608c0} 35. Qxc6 Qxc6 36.
Nxc6 Bc8 37. Bxe5+ f6 38. Ra7+ Kg6 39. gxf6 Rg8 40. f7 Rf8 41. Ne7+ Kh5 42.
Bf6 Rxf7 43. Ra5+ Bf5 44. Raxf5+ Kg4 45. Rg5+ Kh4 46. Rh2#) 33. Qxf5) 30.
Bxe5 Qxe5 31. Qxe5 dxe5 32. Ne8) 29. Bc6 {+6.46/27 4.8s !!} Qxc6 {-2.78/17
24s} 30. Qxh7+ {+18.24/29 6.8s} Kxh7 {-2.93/17 0.040s} 31. Rh2+ {+23.13/28
12s} Bh3 {-3.10/17 5.4s} 32. Rxh3+ {+23.91/26 3.1s} Nh4 {-3.23/17 4.8s} 33.
Rxh4+ {+27.75/25 5.4s} Kg6 {-3.50/16 3.4s} 34. g8=Q+ {+28.80/26 5.2s} Kf5
{-3.67/14 4.7s} 35. g6 {+29.41/24 3.4s} Ke6 {-4.69/8 38s} (35. .. Re2 {Sf}
36. gxf7) 36. gxf7 {+148.99/31 3.8s} Rb8 {-5.05/8 6.5s} 37. Qg4+ {+M21/39
2.9s} Ke7 {-6.58/8 5.8s} 38. Rh8 {+M19/51 3.5s} Rcb7 {-10.46/5 13s} 39.
Re1+ {+M11/75 3.1s} Qe4 {-25.76/4 8.0s, White wins by adjudication} 1-0
Up Topic The Rybka Lounge / Computer Chess / 300 game blitz match: Lc0 162.0 - Stockfish 138.0

Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill