Does your engine blunder here?

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

pilgrimdan
Posts: 405
Joined: Sat Jul 02, 2011 10:49 pm

Re: Does your engine blunder here?

Post by pilgrimdan »

Rodolfo Leoni wrote:
pilgrimdan wrote: Chessmaster 9000 database has 30 games with the position 16. Rg1 ...

8 games of which include the move 16. ...h5

with the following results...

+5 =2 -1
Of course there are many games with 16.Rg1. It's the standard Scid ECO B99y variation! :)

The fact is, most engines are stronger than the top human GMs now, and they could be used to build/tune opening books. But in some cases their task isn't so easy.

The question was: how much deep must they go to see 16. ..b4 is wrong? Stockfish finds 16. ..h5 rather quickly:

Code: Select all

FEN: r1b1kb1r/2q2p1p/p2ppP2/1pn5/3NP3/2N2Q2/PPP4P/2KR1BR1 b kq - 2 16

Stockfish_8_x64_popcnt:
 8/9	00:00	 19k	3,110k	+0.55	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Qb3 Ne5 20.Kb1
 9/11	00:00	 32k	3,156k	+0.79	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Qb3 Ne5 20.Qxb4 Bb7
 10/14	00:00	 57k	3,579k	+0.79	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Qe3+ Ne5 20.Nc6 Bd7 21.Nxb4 Qc5 22.Qxc5 dxc5
 11/17	00:00	 103k	3,956k	+0.31	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Re1+ Ne5 20.Nc6 Bb7 21.Bh3 Bh6+ 22.Kb1 Bxc6 23.dxc6 Kd8
 12/26	00:00	 199k	4,239k	+0.16	16. ... b4 17.Nce2 Bb7 18.Ng3 d5 19.exd5 Bxd5 20.Qe3 0-0-0 21.Kb1 Kb8 22.Re1 Rc8 23.Bh3
 13/26	00:00	 414k	4,404k	+0.36	16. ... b4 17.Nce2 Bb7 18.Ng3 d5 19.Kb1 dxe4 20.Qe2 Bd6 21.Bg2 Bxg3 22.hxg3 Rg8 23.Nb3 Nxb3 24.axb3 Qxg3 25.Bxe4 Qxg1 26.Bxb7
 14/26	00:00	 762k	4,591k	+0.48	16. ... b4 17.Nce2 Bb7 18.Ng3 Bh6+ 19.Kb1 0-0-0 20.Bc4 Kb8 21.Nb3 Nd7 22.Qe2 Qb6 23.Nd4 Rhg8 24.Rgf1 Ne5
 15/26	00:00	 1,508k	4,802k	+0.54	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Bb7 20.Bh3 Ne5 21.Rge1 a5 22.Kb1 Ra6 23.Bg2 Rg8 24.Rd4 Rg6 25.Rh4 Bh6
 16/30	00:00	 2,943k	4,865k	-0.13	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Bb7 20.Bh3 Ne5 21.Rge1 h5 22.Kb1 a5 23.Qd3 Bxc6 24.dxc6 Rh6 25.Bd7+ Kd8 26.Qd5 Rxf6
 17/30	00:01	 5,385k	4,778k	-0.31	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Rb8 20.Bh3 Ne5 21.Rge1 Bxh3 22.Qxh3 Rb5 23.Kb1 a5 24.Qd3 Rb6 25.Qc4 Qd7 26.Qf4 Rg8 27.Qe4 Rg6
 18/30	00:01	 5,530k	4,784k	-0.11	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Rb8 20.Bh3 Ne5 21.Rge1 Bxh3 22.Qxh3 Rg8 23.Kb1 Rb5 24.Qxh7 Rg6 25.Qh4 Rxd5 26.Rxd5 Qxc6 27.Rdd1 a5
 19/32	00:01	 8,355k	4,772k	-0.40	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Bb7 20.Bh3 Ne5 21.Rge1 h5 22.Kb1 a5 23.Qf4 Bh6 24.Qd4 Bxc6 25.dxc6 Kf8 26.Qxd6+ Qxd6 27.Rxd6 Nc4 28.Rd7 Nd2+ 29.Ka1
 20/32	00:02	 9,577k	4,734k	-0.41	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Rb8 20.Bh3 Ne5 21.Rge1 Bxh3 22.Qxh3 Rb6 23.Kb1 a5 24.Qf5 a4 25.Rd4 Rg8 26.Qe4 h5 27.h3 b3 28.Rc4
 21/38	00:02	 12,549k	4,707k	-0.34	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Rb8 20.Bh3 Ne5 21.Rge1 Bxh3 22.Qxh3 Rg8 23.Qxh7 Rg5 24.Qh4 Rg2 25.Qc4 Rc8 26.Kb1 Kd7 27.Qh4 Nxc6 28.Qh3+ Kd8 29.Qxg2 Ne5 30.h4 a5
 22/38-	00:03	 14,231k	4,675k	-0.41	16. ... b4 17.Nd5
 22/38-	00:03	 17,130k	4,654k	-0.49	16. ... b4 17.Nd5
 22/38-	00:04	 21,158k	4,610k	-0.60	16. ... b4 17.Nd5
 22/41+	00:05	 26,640k	4,595k	-0.50	16. ... b4
 22/41	00:05	 27,433k	4,591k	-0.47	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Rb8 20.Bh3 Ne5 21.Rge1 Rg8 22.Bxc8 Rxc8 23.Kb1 a5 24.Qf5 Rg2 25.h4 a4 26.Rd4 Qd7 27.Qf4 b3 28.axb3 axb3 29.cxb3 Qh3 30.Rxe5+ dxe5 31.Qxe5+ Kd7 32.Ra4 Qh1+ 33.Ka2
 23/41-	00:06	 31,198k	4,535k	-0.54	16. ... b4 17.Nd5
 23/41-	00:07	 32,820k	4,528k	-0.61	16. ... b4 17.Nd5
 23/41-	00:08	 40,285k	4,486k	-0.72	16. ... b4 17.Nd5
 23/41+	00:09	 41,555k	4,485k	-0.63	16. ... b4
 23/41+	00:09	 41,741k	4,486k	-0.42	16. ... b4
 23/41	00:09	 43,314k	4,486k	-0.20	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Bb7 20.Bh3 Ne5 21.Rge1 Rg8 22.Qh5 Bxc6 23.dxc6 Rg6 24.Kb1 Bh6 25.Qh4 Kf8 26.Bf5 Bg5 27.Qxh7 Rg8 28.Rxe5 Bxf6 29.Re6 Rh8 30.Rxf6 Rxh7 31.Bxh7 Qxc6 32.Rfxd6 Qc5
 24/41+	00:10	 45,581k	4,484k	-0.12	16. ... b4
 24/41-	00:10	 47,949k	4,477k	-0.20	16. ... b4 17.Nd5
 24/47-	00:12	 56,074k	4,449k	-0.46	16. ... b4 17.Nd5
 24/47	00:13	 60,444k	4,433k	-0.57	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Bb7 20.Bh3 Ne5 21.Rge1 Rg8 22.Qf4 Bxc6 23.dxc6 Rg6 24.Kb1 Rb8 25.Bd7+ Kd8 26.Bf5 Ke8 27.Rd5 Rb5 28.Rxb5 axb5 29.Bxg6 hxg6 30.Qe4 Bh6 31.Qxb4 Bg5 32.Qxb5 Bxf6 33.Re3 Be7
 25/47-	00:15	 66,264k	4,407k	-0.64	16. ... b4 17.Nd5
 25/47-	00:17	 78,120k	4,371k	-0.71	16. ... b4 17.Nd5
 25/47-	00:21	 92,792k	4,322k	-0.82	16. ... b4 17.Nd5
 25/47+	00:32	 132,126k	4,103k	-0.73	16. ... h5
 25/47+	00:39	 156,451k	4,003k	-0.24	16. ... h5
 25/47	01:22	 291,246k	3,540k	-0.02	16. ... h5 17.a3 Qb6 18.Rg7 b4 19.axb4 Qxb4 20.Nc6 Qb6 21.Ne7 Rb8 22.b3 Nd7 23.Na4 Qc7 24.Nb2 Nxf6 25.Qxf6 Qxe7 26.Rxf7 Qxf6 27.Rxf6 Ke7 28.Rf2 Bh6+ 29.Kb1 Bb7 30.Nc4 Rhf8 31.Rxf8 Rxf8 32.e5 dxe5 33.Nxe5 Rf5 34.Nc4 Rf4 35.Bd3 Rf6
 26/47	01:34	 328,064k	3,473k	 0.00	16. ... h5 17.a3 Qb6 18.Kb1 b4 19.axb4 Qxb4 20.Qe3 Bd7 21.e5 Rb8 22.b3 Qa3 23.Rg5 d5 24.Be2 Bh6 25.Rxh5 Nxb3 26.cxb3 Rxb3+ 27.Nxb3 Qxb3+ 28.Ka1 Qa3+ 29.Kb1 Qb3+
 27/47-	01:36	 334,847k	3,461k	-0.07	16. ... h5 17.a3
 27/47-	01:39	 342,713k	3,451k	-0.14	16. ... h5 17.a3
 27/47-	01:55	 390,687k	3,396k	-0.40	16. ... h5 17.a3
 27/47+	02:12	 442,659k	3,338k	-0.28	16. ... h5
 27/47	02:26	 483,300k	3,306k	-0.33	16. ... h5 17.a3 Qb6 18.Kb1 b4 19.axb4 Qxb4 20.Qe3 Bd7 21.e5 Rb8 22.b3 Rc8 23.Rg3 h4 24.Na2 Qa3 25.b4 Qxe3 26.Rxe3 Rb8 27.exd6 a5 28.Bg2 Rg8 29.Bf3 Rb6 30.Nc6 Bxc6 31.Bxc6+ Rxc6 32.bxc5 Rxc5
But if we give it the "blunder" and we continue some plies with the best moves from both sides, it thinks it's even.
(moves: 16. ..b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Bb7 20.Bh3 Ne5 21.Rge1)

Code: Select all

FEN: r3kb1r/1bq2p1p/p1Np1P2/3Pn3/1p6/5Q1B/PPP4P/2KRR3 b kq - 6 21

Stockfish_8_x64_popcnt:
 8/9	00:00	 4k	1,368k	-0.08	21. ... a5 22.Qf4 Bxc6 23.dxc6 Ra6 24.Bd7+ Kd8 25.Bf5
 9/13	00:00	 5k	1,797k	-0.40	21. ... a5 22.Qf4 Bxc6 23.dxc6 Ra6 24.Rxe5+ dxe5 25.Bd7+ Kd8 26.Qc4
 10/14	00:00	 13k	2,602k	-0.87	21. ... Rg8 22.Qf4 Bxc6 23.dxc6 b3 24.axb3 Rg6 25.Kb1 Rb8 26.Bd7+ Kd8
 11/17	00:00	 38k	4,179k	-0.90	21. ... Rg8 22.Qf4 b3 23.cxb3 Bc8 24.Bxc8 Qxc8 25.Rd4 Qc7 26.Kb1 Rg2 27.Rc4
 12/17	00:00	 54k	4,151k	-0.91	21. ... Rg8 22.Qf4 b3 23.cxb3 Bc8 24.Bxc8 Qxc8 25.Rd4 Rg2 26.Kb1 Kd7 27.Rc1 Nxc6 28.dxc6+ Kc7
 13/19	00:00	 140k	4,681k	-0.89	21. ... Bxc6 22.dxc6 Rb8 23.Bd7+ Kd8 24.Rxe5 dxe5 25.Qd5 Bd6 26.Qxf7 b3 27.axb3 Rb4 28.Qg7 Rf8 29.c4
 14/21	00:00	 208k	4,959k	-1.09	21. ... Rg8 22.Qf4 b3 23.cxb3 Bc8 24.Bxc8 Qxc8 25.Rd4 h6 26.Kb1 Rg5 27.Nxe5 dxe5 28.Rxe5+ Kd7 29.Rxg5
 15/26	00:00	 323k	5,128k	-0.94	21. ... Rg8 22.Qf4 b3 23.cxb3 Bc8 24.Bxc8 Qxc8 25.Rd4 h6 26.Kb1 Rg5 27.Nxe5 dxe5 28.Rxe5+ Kd7 29.Rxg5
 16/35	00:00	 529k	4,902k	-0.68	21. ... Rg8 22.Qf4 b3 23.axb3 Bxc6 24.dxc6 Rg6 25.Kb1 Rb8 26.Bd7+ Kd8 27.Bf5 Ke8 28.Rd5 Rb5 29.Rxb5 axb5 30.Bxg6 hxg6 31.Qd4 Bh6
 17/35	00:00	 965k	4,755k	-0.92	21. ... Rg8 22.Qf4 b3 23.axb3 Bxc6 24.dxc6 Rg6 25.Kb1 Rb8 26.Bd7+ Kd8 27.Bf5 Ke8 28.Rd5 Rb5 29.Rxb5 axb5 30.Qb4 Qxc6 31.Bd3 Kd8 32.Rxe5 dxe5 33.Qxf8+ Qe8 34.Qa3 Rg1+ 35.Ka2
 18/37	00:00	 1,984k	4,636k	-1.03	21. ... Rg8 22.Qf4 b3 23.axb3 Bxc6 24.dxc6 Rg6 25.Kb1 Rb8 26.Bd7+ Kd8 27.Bf5 Ke8 28.Rd5 Rb5 29.Rxb5 axb5 30.Qb4 Qxc6 31.Bd3 Kd8 32.Bxb5 Qc5 33.Qxc5 dxc5 34.Rxe5 Bd6 35.Rh5 Rxf6
 19/37	00:00	 2,211k	4,577k	-1.01	21. ... Rg8 22.Qf4 b3 23.axb3 Bxc6 24.dxc6 Rg6 25.Kb1 Rb8 26.Bd7+ Kd8 27.Bf5 Ke8 28.Bxg6 hxg6 29.c4 Rb6 30.c5 Rxc6 31.b4 Qc8 32.Rd5 Qd7 33.cxd6 Bxd6 34.Qh4 Kd8 35.Rexe5 Bxe5 36.Rxd7+ Kxd7
 20/38	00:01	 4,754k	4,306k	-1.23	21. ... Rg8 22.Qf4 Bxc6 23.dxc6 Rg6 24.Kb1 Rb8 25.Bd7+ Kd8 26.Bf5 Ke8 27.Rd5 Rb5 28.Rxb5 axb5 29.Qxb4 Qxc6 30.Bd3 Kd8 31.Bxg6 hxg6 32.Rxe5 dxe5 33.Qxf8+ Qe8 34.Qd6+ Kc8 35.Qc5+ Kd8 36.a3 e4
 21/42	00:02	 9,982k	4,028k	-0.34	21. ... Rg8 22.Qf4 Bxc6 23.dxc6 Rg6 24.Kb1 Bh6 25.Qxb4 Rxf6 26.Bd7+ Kd8 27.Qh4 Ke7 28.Rf1 Bg7 29.Rde1 Rd8 30.Qg5 Bh8 31.c4 h6 32.Qh4 Rxd7 33.cxd7 Qxd7 34.Kc2 Qc6 35.Kb1 Qc5
 22/42-	00:03	 12,429k	3,943k	-0.34	21. ... Rg8 22.Qf4
 22/42+	00:03	 13,143k	3,929k	-0.27	21. ... Rg8
 22/42-	00:03	 14,946k	3,901k	-0.35	21. ... Rg8 22.Qf4
 22/42+	00:03	 15,337k	3,898k	-0.23	21. ... Rg8
 22/42	00:04	 17,443k	3,887k	 0.00	21. ... Rg8 22.Qf4 Bxc6 23.dxc6 Rg6 24.Kb1 Bh6 25.Qxb4 Rxf6 26.Bd7+ Kd8 27.Qh4 Ke7 28.Bf5 Rd8 29.Be4 Qb6 30.Rf1 Bg7 31.Bd5 Rb8 32.b3 h6 33.Rf4 Qc5 34.Bh1 Qc3 35.Rdf1 Ng6
 23/42	00:04	 18,082k	3,883k	 0.00	21. ... Rg8 22.Qf4 Bxc6 23.dxc6 Rg6 24.Kb1 Bh6 25.Qxb4 Rxf6 26.Bd7+ Kd8 27.Qh4 Ke7 28.Rf1 Bg7 29.Qg5 Bh8 30.Qh4 Bg7
 24/42	00:05	 19,922k	3,871k	 0.00	21. ... Rg8 22.Qf4 Bxc6 23.dxc6 Rg6 24.Kb1 Bh6 25.Qxb4 Rxf6 26.Bd7+ Kd8 27.Qh4 Ke7 28.Rf1 Bg7 29.Qg5 Bh8 30.Qh4 Bg7
 25/42	00:06	 25,092k	3,857k	 0.00	21. ... Rg8 22.Qf4 Bxc6 23.dxc6 Rg6 24.Kb1 Bh6 25.Qxb4 Rxf6 26.Bd7+ Kd8 27.Bf5 Ke8 28.Qh4 Rxf5 29.Qxh6 Qxc6 30.Rxd6 Qc5 31.Qxh7 Qxd6 32.Qxf5 f6 33.Qh5+ Ke7 34.Qh7+ Ke6 35.Qh3+ Kf7 36.Qh7+ Ke6
 26/42	00:07	 27,830k	3,841k	 0.00	21. ... Rg8 22.Qf4 Bxc6 23.dxc6 Rg6 24.Kb1 Bh6 25.Qxb4 Rxf6 26.Bd7+ Kd8 27.Bf5 Ke8 28.Bd7+ Kd8
 27/42	00:08	 33,376k	3,817k	 0.00	21. ... Rg8 22.Qf4 Bxc6 23.dxc6 Rg6 24.Kb1 Bh6 25.Qxb4 Rxf6 26.Bd7+ Kd8 27.Bf5 Ke8 28.Bd7+ Kd8
 28/42	00:13	 51,425k	3,774k	 0.00	21. ... Rg8 22.Qf4 Bxc6 23.dxc6 Rg6 24.Kb1 Bh6 25.Qxb4 Rxf6 26.Bd7+ Kd8 27.Bf5 Ke8 28.Bd7+ Kd8
 29/42	00:18	 68,678k	3,723k	 0.00	21. ... Rg8 22.Qf4 Bxc6 23.dxc6 Rg6 24.Kb1 Bh6 25.Qxb4 Rxf6 26.Bd7+ Kd8 27.Bf5 Ke8 28.Bd7+ Kd8
 30/42	00:20	 75,259k	3,705k	 0.00	21. ... Rg8 22.Qf4 Bxc6 23.dxc6 Rg6 24.Kb1 Bh6 25.Qxb4 Rxf6 26.Bd7+ Kd8 27.Bf5 Ke8 28.Bd7+ Kd8
 31/42	00:36	 134,201k	3,650k	 0.00	21. ... Rg8 22.Qf4 Bxc6 23.dxc6 Rg6 24.Kb1 Bh6 25.Qxb4 Rxf6 26.Bd7+ Kd8 27.Bf5 Ke8 28.Bd7+ Kd8
With correct play from both sides, 16. ..b4 should give us the following variation:

[pgn][Event "B99y analysys"]
[Site "UTENTE-PC"]
[Date "2017.07.04"]
[White "Leoni"]
[Black "Critter_1.6a"]
[Termination "unterminated"]
[PlyCount "61"]
[WhiteType "human"]
[BlackType "program"]

1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Bg5 e6 7. f4 Be7 8.
Qf3 Qc7 9. O-O-O Nbd7 10. g4 b5 11. Bxf6 Nxf6 12. g5 Nd7 13. f5 Nc5 14. f6
gxf6 15. gxf6 Bf8 16. Rg1 b4 17. Nd5 exd5 18. exd5 Nd7 19. Nc6 Bb7 20. Bh3
Ne5 21. Rge1 Rg8 22. Qf4 Bxc6 23. dxc6 Rg6 24. Kb1 Bh6 25. Qxb4 Rxf6 26.
Rxe5+ dxe5 27. Bd7+ Kd8 28. a4 Rb8 29. Qc5 Rd6 30. Qxd6 Qxd6 31. Rxd6 *

[/pgn]

Many engines would play 16. ..b4, thinking it's a good move and to be better.

Code: Select all

FEN: r1b1kb1r/2q2p1p/p2ppP2/1pn5/3NP3/2N2Q2/PPP4P/2KR1BR1 b kq - 2 16

Equinox 3.30-x64:
 10/22+	00:00	 33k	1,063k	-0.30	16. ... b4 17.Nce2 Bb7 18.Ng3 d5 19.Qh5 Qf4+ 20.Kb1 Qxf6 21.e5
 10/22+	00:00	 36k	1,157k	-0.22	16. ... b4 17.Nce2 Bb7 18.Ng3 d5 19.Qh5 Qf4+ 20.Kb1 Qxf6 21.e5
 10/22	00:00	 41k	1,333k	-0.20	16. ... b4 17.Nce2 Bb7 18.Ng3 d5 19.Kb1 Nxe4 20.Bd3 Nxg3 21.Rxg3 Bd6
 11/22	00:00	 60k	1,275k	-0.17	16. ... b4 17.Nce2 Bb7 18.Ng3 d5 19.Kb1 Nxe4 20.Bd3 Nxg3 21.Rxg3 Bd6 22.Rg7
 12/26	00:00	 109k	1,751k	-0.19	16. ... b4 17.Nce2 Bb7 18.Ng3 d5 19.Kb1 Bd6 20.exd5 Bxd5 21.Qg4 Na4 22.Ka1
 13/26+	00:00	 192k	2,457k	-0.11	16. ... b4 17.Nce2 Bb7 18.Ng3 d5 19.Kb1 Bd6 20.exd5 Bxd5 21.Qg4 Na4 22.Ka1
 13/26	00:00	 221k	2,352k	-0.04	16. ... b4 17.Nce2 Bb7 18.Ng3 d5 19.Kb1 dxe4 20.Qe3 Qe5 21.Bg2 Rd8 22.Rgf1 Rg8
 14/28+	00:00	 394k	2,817k	+0.04	16. ... b4 17.Nce2 Bb7 18.Ng3 d5 19.Kb1 dxe4 20.Qe3 Qe5 21.Bg2 Rd8 22.Rgf1 Rg8
 14/28+	00:00	 470k	3,012k	+0.12	16. ... b4 17.Nce2 Bb7 18.Ng3 d5 19.Kb1 dxe4 20.Qe3 Qe5 21.Bg2 Rd8 22.Rgf1 Rg8
 14/28	00:00	 605k	3,237k	+0.20	16. ... b4 17.Nce2 Bb7 18.Ng3 d5 19.Kb1 dxe4 20.Qe3 Qe5 21.Bg2 Rd8 22.Rgf1 Rg8 23.Rf4
 15/32+	00:00	 1,142k	3,331k	+0.28	16. ... b4 17.Nce2 Bb7 18.Ng3 d5 19.Kb1 dxe4 20.Qe3 Qe5 21.Bg2 Rd8 22.Rgf1 Rg8 23.Rf4
 15/32+	00:01	 2,287k	2,221k	+0.36	16. ... b4 17.Nce2 Bb7 18.Ng3 d5 19.Kb1 dxe4 20.Qe3 Qe5 21.Bg2 Rd8 22.Rgf1 Rg8 23.Rf4
 15/32	00:01	 2,455k	2,156k	+0.32	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bg2 Bh6+ 20.Kb1 0-0-0 21.Ka1 Nd7 22.Qb3 Qc5 23.Qf3
 16/32	00:01	 3,321k	2,047k	+0.36	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bg2 Bh6+ 20.Kb1 Rc8 21.Qf2 Rg6 22.Qe1 Rxf6 23.Qxb4
 17/33	00:03	 6,433k	2,031k	+0.40	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bc4 Rc8 20.Qe2 Nd7 21.Bb3 Rg6 22.Kb1 Nxf6 23.Ba4+
 18/34-	00:03	 7,800k	2,016k	+0.32	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bc4 Rc8 20.Qe2 Nd7 21.Bb3 Rg6 22.Kb1 Nxf6 23.Ba4+
 18/34-	00:05	 10,476k	2,054k	+0.24	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bc4 Rc8 20.Qe2 Nd7 21.Bb3 Rg6 22.Kb1 Nxf6 23.Ba4+
 18/34	00:06	 13,401k	2,060k	+0.30	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bc4 Rg6 20.Kb1 Qb6 21.Bb3 Rc8 22.h4 Nxb3 23.axb3
 19/34	00:08	 17,568k	2,137k	+0.31	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bc4 Rg6 20.Kb1 Qb6 21.Bb3 Rc8 22.h4 Nxb3 23.axb3
 20/45+	00:15	 33,172k	2,181k	+0.39	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bc4 Rg6 20.Kb1 Qb6 21.Bb3 Rc8 22.h4 Nxb3 23.axb3
 20/45	00:15	 34,103k	2,175k	+0.37	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bc4 0-0-0 20.Kb1 Kb8 21.Qf1 Bh6 22.Qe1 Qb6 23.e5
 21/45-	00:36	 84,159k	2,325k	+0.29	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bc4 0-0-0 20.Kb1 Kb8 21.Qf1 Bh6 22.Qe1 Qb6 23.e5
 21/45-	00:38	 89,158k	2,293k	+0.21	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bc4 0-0-0 20.Kb1 Kb8 21.Qf1 Bh6 22.Qe1 Qb6 23.e5
 21/45-	00:44	 118,449k	2,640k	-0.09	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bc4 0-0-0 20.Kb1 Kb8 21.Qf1 Bh6 22.Qe1 Qb6 23.e5
 21/45	00:45	 121,637k	2,677k	-0.05	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Bb7 20.Bh3 Ne5 21.Nxe5 dxe5 22.Qh5 Bd6 23.Rg7
 22/45	00:50	 148,905k	2,977k	 0.00	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Bb7 20.Bh3 Ne5 21.Nxe5 dxe5 22.Qh5 Bd6 23.Rg7
 23/45	00:59	 204,144k	3,419k	 0.00	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Bb7 20.Bh3 Ne5 21.Nxe5 dxe5 22.Qh5 Bd6 23.Rg7
 24/45	01:18	 303,761k	3,853k	 0.00	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Bb7 20.Bh3 Ne5 21.Nxe5 dxe5 22.Qh5 Bd6 23.Rg7

Code: Select all

FEN: r1b1kb1r/2q2p1p/p2ppP2/1pn5/3NP3/2N2Q2/PPP4P/2KR1BR1 b kq - 2 16

Arasanx-64-popcnt:
 8	00:00	 16k	885k	+0.08	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Re1+ Ne5 20.Nc6 Qb6
 9-	00:00	 24k	977k	-0.09	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Re1+ Ne5 20.Nc6 Qb6
 9	00:00	 33k	961k	-0.35	16. ... b4 17.Nce2 Bb7 18.Ng3 d5 19.exd5 Bxd5 20.Qe3 0-0-0
 10	00:00	 69k	1,097k	-0.46	16. ... b4 17.Nce2 Bb7 18.Ng3 Bh6+ 19.Kb1 0-0-0 20.a3 d5 21.e5 Nd7 22.axb4 Nxe5
 11	00:00	 122k	1,282k	-0.46	16. ... b4 17.Nce2 Bb7 18.Ng3 Bh6+ 19.Kb1 0-0-0 20.a3 d5 21.e5 Nd7 22.axb4 Nxe5
 12+	00:00	 131k	1,313k	-0.27	16. ... b4 17.Nce2
 12	00:00	 165k	1,420k	-0.12	16. ... b4 17.Nce2 Bb7 18.Ng3 Bh6+ 19.Kb1 0-0-0 20.a3 d5 21.e5 Qxe5 22.axb4 Be3
 13	00:00	 250k	1,623k	-0.05	16. ... b4 17.Nce2 Bb7 18.Ng3 h5 19.h4 d5 20.exd5 Bxd5 21.Qf2 Qa5 22.Qd2 Qxa2 23.Qxb4
 14	00:00	 324k	1,772k	+0.07	16. ... b4 17.Nce2 Bb7 18.Ng3 h5 19.h4 d5 20.exd5 Bxd5 21.Qf2 Qa5 22.Ndf5 Qxa2 23.Ng7+ Bxg7 24.fxg7
 15+	00:00	 573k	1,996k	+0.26	16. ... b4 17.Nce2
 15	00:00	 645k	2,017k	+0.13	16. ... b4 17.Nce2 Bb7 18.Ng3 h5 19.h4 d5 20.Nxh5 0-0-0 21.Kb1 dxe4 22.Qe2 Na4 23.Bg2 Bc5
 16+	00:00	 1,442k	2,185k	+0.32	16. ... b4 17.Nce2
 16	00:00	 1,623k	2,118k	+0.21	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bh3 Bh6+ 20.Kb1 Rg6 21.Bg2 Bg5 22.Nh5 Nxe4 23.Qe2 Nxf6 24.Bxb7 Qxb7
 17	00:00	 1,853k	2,125k	+0.29	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bh3 Bh6+ 20.Kb1 Rg6 21.Bg2 Qd8 22.Nh5 Bxe4 23.Qe2 Rxg2 24.Rxg2 Bxg2 25.Qxg2
 18	00:01	 2,830k	2,179k	+0.30	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bg2
 19-	00:01	 3,359k	2,216k	+0.11	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bg2
 19	00:01	 4,275k	2,220k	+0.06	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bg2 Bh6+ 20.Kb1 Rg6 21.Qf2 0-0-0 22.Qe1 Qb6 23.e5 Bxg2 24.Rxg2 dxe5 25.Qxe5 Nd7 26.Qe2 Nxf6
 20-	00:02	 6,729k	2,303k	-0.12	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bg2 Bh6+ 20.Kb1 Rg6 21.Qf2 0-0-0 22.Qe1 Qb6 23.e5 Bxg2 24.Rxg2 dxe5 25.Qxe5 Nd7 26.Qe2 Nxf6
 20-	00:04	 9,682k	2,349k	-0.50	16. ... b4 17.Nce2 Bb7 18.Ng3 Rg8 19.Bg2 Bh6+ 20.Kb1 Rg6 21.Qf2 0-0-0 22.Qe1 Qb6 23.e5 Bxg2 24.Rxg2 dxe5 25.Qxe5 Nd7 26.Qe2 Nxf6
 20	00:04	 9,865k	2,331k	-0.50	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Re1+ Ne5 20.Nc6 Bh6+ 21.Kb1 Bb7 22.Qg3 Bxc6 23.dxc6 Bf8 24.Qf3 Kd8 25.Qh3 Qc8 26.Qb3 Qxc6 27.Bg2
 21+	00:04	 10,806k	2,357k	-0.31	16. ... b4 17.Nd5
 21	00:05	 12,245k	2,357k	-0.34	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Re1+ Ne5 20.Nc6 Bd7 21.Nxe5 dxe5 22.d6 Qa7 23.Rxe5+ Kd8 24.Rg7 Bxd6 25.Re1 Kc8 26.Rxf7 Rf8 27.Rxh7 Rb8
 22+	00:05	 13,675k	2,382k	-0.15	16. ... b4 17.Nd5
 22	00:06	 14,589k	2,391k	-0.31	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Re1+ Ne5 20.Nc6 Bb7 21.Kb1 Bxc6 22.dxc6 h5 23.Bh3 Rb8 24.Qd5 Bh6 25.Rxe5+ dxe5 26.Bd7+ Kd8 27.Qxf7 Qd6 28.h4
 23	00:07	 18,816k	2,419k	-0.32	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Re1+ Ne5 20.Nc6 Bb7 21.Kb1 Bxc6 22.dxc6 h5 23.Bh3 Rb8 24.Qd5 Bh6 25.Rxe5+ dxe5 26.Bd7+ Kd8 27.Qxf7 Qd6 28.Bf5 Be3
 24	00:09	 23,889k	2,451k	-0.31	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Re1+ Ne5 20.Nc6 Bb7 21.Kb1 Bxc6 22.dxc6 h5 23.Bh3 Rb8 24.Qd5 Bh6 25.Rxe5+ dxe5 26.Bd7+ Kd8 27.Qxf7 Qd6 28.Bf5 Rb6 29.Bd7
 25	00:15	 38,645k	2,447k	-0.26	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Re1+ Ne5 20.Nc6 Bb7 21.Kb1 Bxc6 22.dxc6 h5 23.Bh3 Rb8 24.Qd5 Bh6 25.Rxe5+ dxe5 26.Bd7+ Kd8 27.Qxf7 Qd6 28.Bf5 Rb6 29.Qa7 Be3
 26	00:33	 81,529k	2,414k	-0.43	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Qh5 Ne5 20.Ne6 Qb6
 27	00:43	 103,298k	2,351k	-0.36	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6
 28	00:59	 136,211k	2,290k	-0.19	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Bb7 20.Kb1 Ne5 21.Re1 Bxc6 22.dxc6 a5 23.Qd5 a4 24.Bh3 Ra5 25.Bd7+ Kd8 26.Rxe5 dxe5 27.Qxf7 Qa7 28.Rd1 Kc7 29.Bf5+ Kb6 30.Rd7 Qa8
 29-	01:23	 185,797k	2,217k	-0.38	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Bb7 20.Kb1 Ne5 21.Re1 Bxc6 22.dxc6 a5 23.Qd5 a4 24.Bh3 Ra5 25.Bd7+ Kd8 26.Rxe5 dxe5 27.Qxf7 Qa7 28.Rd1 Kc7 29.Bf5+ Kb6 30.Rd7 Qa8
 29	02:01	 263,247k	2,171k	-0.30	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Bb7 20.Qe2+ Ne5 21.Kb1 Bxc6 22.dxc6 a5 23.Bh3 a4 24.Rge1 Ra5 25.Qe4 a3 26.b3 Kd8 27.Re2 Qxc6 28.Qxb4 Qc7 29.Red2 h5 30.c4 Rh6 31.Rf2
 30	02:04	 270,213k	2,166k	-0.49	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Bb7 20.Qe2+ Ne5 21.Kb1 Bxc6 22.dxc6 a5 23.Bh3 a4 24.Rge1 Ra5 25.Qe4 a3 26.b3 Kd8 27.Re2 Qxc6 28.Qxb4 Qc7 29.Red2 h5 30.c4 Rh6 31.Rf2
 30	02:04	 270,213k	2,166k	-0.49	16. ... b4 17.Nd5 exd5 18.exd5 Nd7 19.Nc6 Bb7 20.Qe2+ Ne5 21.Kb1 Bxc6 22.dxc6 a5 23.Bh3 a4 24.Rge1 Ra5 25.Qe4 a3 26.b3 Kd8 27.Re2 Qxc6 28.Qxb4 Qc7 29.Red2 h5 30.c4 Rh6 31.Rf2
The question is still: Is your engine able to avoit the weak move 16. ..b4?
[d]r1b1kb1r/2q2p1p/p2ppP2/1pn5/3NP3/2N2Q2/PPP4P/2KR1BR1 b kq - 0 16
top 3 moves by CM9...

Time Depth Score Positions Moves
5:28 3/12 -0.47 124452235 16...b4 17.Nb1 Bb7 18.Nd2 Bh6 19.Qh3
Bf4 20.Bg2 Be5 21.Nc4 Bxf6 22.Ne2 Rd8
14:06 3/12 -0.26 294608000 16...Bb7 17.a3 O-O-O 18.Kb1 Kb8
19.Be2 Rc8 20.Qg2 Nd7 21.Qf2 Ne5 22.Nf3
9:33 3/12 -0.19 219675213 16...h5 17.a3 Nd7 18.Bh3 Ne5 19.Qf2
Nc4 20.Rd3 Bb7 21.Nf3 Bh6+ 22.Kb1 Qc5
Top
Rodolfo Leoni
Posts: 545
Joined: Tue Jun 06, 2017 4:49 pm
Location: Italy

Re: Does your engine blunder here?

Post by Rodolfo Leoni »

pilgrimdan wrote:
[d]r1b1kb1r/2q2p1p/p2ppP2/1pn5/3NP3/2N2Q2/PPP4P/2KR1BR1 b kq - 0 16
top 3 moves by CM9...

Time Depth Score Positions Moves
5:28 3/12 -0.47 124452235 16...b4 17.Nb1 Bb7 18.Nd2 Bh6 19.Qh3
Bf4 20.Bg2 Be5 21.Nc4 Bxf6 22.Ne2 Rd8
14:06 3/12 -0.26 294608000 16...Bb7 17.a3 O-O-O 18.Kb1 Kb8
19.Be2 Rc8 20.Qg2 Nd7 21.Qf2 Ne5 22.Nf3
9:33 3/12 -0.19 219675213 16...h5 17.a3 Nd7 18.Bh3 Ne5 19.Qf2
Nc4 20.Rd3 Bb7 21.Nf3 Bh6+ 22.Kb1 Qc5
That's quite interesting. CM9 thinks at 16. ..b4 as a the best move, possibly because it misses 17.Nd5. It misses 16. ..Bb7 17.Bh3! too...

Writing the analyses for Peter now... :)
F.S.I. Chess Teacher
Top
Rodolfo Leoni
Posts: 545
Joined: Tue Jun 06, 2017 4:49 pm
Location: Italy

Re: Does your engine blunder here?

Post by Rodolfo Leoni »

peter wrote:
Rodolfo Leoni wrote: I just think performances aren't so relevant. 16. ..h5 is playable, and 16. ..b4 is a bad move. An engine tournament isn't necessary. Just some analyses, possibly with the help of an engine with persistent hashes. A pity no top engine has PH option, which would be extremely useful here.

I spent a lot of time on this opening variation, but Critter was the best engine with PH option, so it was not so easy to analyze it carefully.
I agree ...h5 is better, but the question of "playabiltiy" of ...b4 compared to it isn't so easy to be answered yet and that's a question of performance only of course, because Black can lose with ...h5 as well as you see in the not so little number of games in corr.- database, even from recent years only too.
You'd have to prove ...b4 to lose for sure and ...h5 to hold the draw for sure not to have performance (of a certain high level of full games) as the only possible answer if one move is much better playable than the other.

And the most difficult question is, how exactly to use the position as a test position for engines.

Again I agree with you that good or bad llines in output after Forward- Backward of really good test- lines should do it, but that's not the way output- lines are seen normally (which might be a pity too, but again, my personal pov isnt' normal here neither :))

BTW persistent hash is a feature of asmFish, komodo and Shredder13 too
Hi Peter,

this is a short summary.

1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Bg5 e6 7. f4 Be7 8.
Qf3 Qc7 9. O-O-O Nbd7 10. g4 b5 11. Bxf6 Nxf6 12. g5 Nd7 13. f5 Nc5 14. f6
gxf6 15. gxf6 Bf8 16. Rg1 b4 17. Nd5 exd5 18. exd5 Nd7 (18. .. Bb7 19. Re1+
Kd8 20. Rg7 b3 (20. .. Qa5 21. Qh5) 21. a3 Qa5 22. c3 Kc7 23. Rxf7+ Kb6 24.
Kb1 Rb8 25. Qg3 Bxd5 26. Rxf8 Rhxf8 27. Qxd6+) (18. .. Bd7 19. Rg7 O-O-O
20. Rxf7 Kb8 (20. .. Bh6+ 21. Kb1 Rdf8 22. Rg7) 21. Nc6+ Ka8 22. Bh3) 19.
Nc6 Bb7 (19. .. Ne5 20. Nxe5 dxe5 21. d6 Bb7 (21. .. Qa7 22. Bc4 Bd7 23.
Bxf7+ Kxf7 24. Qh5+ Ke6 25. Rge1 Bxd6 26. Qg4+ Kf7 27. Qg7+ Ke6 28. f7)
(21. .. Qb8 22. Bc4 Ra7 23. Qe3 Kd8 24. Qc5) 22. dxc7 Bxf3 23. Bxa6) 20.
Bh3 Ne5 21. Rge1 Rg8 22. Qf4 Bxc6 23. dxc6 Rg6 24. Kb1 Bh6 25. Qxb4 Rxf6
26. Rxe5+ dxe5 27. Bd7+ Kd8 28. a4 Rb8 (28. .. Rf4 29. Qc5 Rd4 30. Rg1) 29.
Qc5 Rd6 (29. .. Rb6 30. b4 Bg7 (30. .. Rf4 31. c3 Bg7 32. Kc2) 31. c3 h6
(31. .. e4 32. Kc2 h6 33. Rd4) 32. Kc2 e4 33. Rd4 Rbxc6 (33. .. Rf2+ 34.
Kb3 Rb2+ 35. Kxb2 Qxh2+ 36. Ka3) 34. Bxc6+ Kc8 35. Rc4) 30. Qxd6 Qxd6 31.
Rxd6 Bg5 32. c3 Kc7 33. Rd5 f6 34. b4 e4 35. b5 axb5 36. axb5 *

In every line White has a decisive advantage.

You can also look at Jeroen Noomen opinion here:

http://www.talkchess.com/forum/viewtopic.php?t=64513
F.S.I. Chess Teacher
Top
Rodolfo Leoni
Posts: 545
Joined: Tue Jun 06, 2017 4:49 pm
Location: Italy

Re: Does your engine blunder here?

Post by Rodolfo Leoni »

MikeB wrote:
Rodolfo Leoni wrote:Sicilian variation:

[pgn][Event "B99y analysis"]
[Site "UTENTE-PC"]
[Date "2017.06.16"]
[White "Leoni"]
[Black "Critter_1.6a_64bit"]
[ECO "B99"]
[Opening "Sicilian"]

1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Bg5 e6 7. f4 Be7 8.
Qf3 Qc7 9. O-O-O Nbd7 10. g4 b5 11. Bxf6 Nxf6 12. g5 Nd7 13. f5 Nc5 14. f6 gxf6 15. gxf6 Bf8 16. Rg1
[/pgn]

Best move is 16. .. h5 here. I noticed many engines tend to play 16. .. b4, which can be considered a blunder because of 17.Nd5! many top engines too! :)

What move your engine would play here?

Rodolfo
Both Stockfish and SF-McB switched to h5 almost at the same time and depth. SF was a little quicker to make the change to h5.

[d]r1b1kb1r/2q2p1p/p2ppP2/1pn5/3NP3/2N2Q2/PPP4P/2KR1BR1 b kq - 2 16

Code: Select all

dep	score	nodes	time	(not shown:  tbhits	knps	seldep)
 32	+0.12 	3.93G	3:20.86	h5 a3 Qb6 Bh3 b4 axb4 Qxb4 Kb1 Qb7 Rg7 Bd7 Qg3 Rb8 b3 a5 Rxf7 Kxf7 Rg1 Ke8 f7+ Kd8 Qg8 Nxe4 Nxe4 Qxe4 Qxh8 Ke7 Rf1 Qe5 Qxe5 dxe5 Nf3 Kf6 Re1 e4 Rxe4 Bd6 Nd4 Kxf7 Bxe6+ Bxe6 Rxe6 Bxh2 Ra6 Rh8 Nf3 Bf4 Rxa5 h4 Ra7+ Kg6 
 32	+0.17!	3.73G	3:10.36	h5! 
 32	+0.28!	3.61G	3:04.71	h5! 
 32	+0.35!	3.59G	3:03.61	h5! 
 31	+0.43 	3.49G	2:58.47	h5 a3 Qb6 Kb1 b4 axb4 Qxb4 Bh3 Qb7 Rg7 Bd7 Qg3 Rb8 b3 a5 Rxf7 Kxf7 Rg1 Ke8 f7+ Kd8 Qg8 Kc7 Qxh8 a4 Rg3 Qa6 Kb2 axb3 cxb3 Ra8 Kc2 Rb8 Qf6 h4 Qxh4 Qa3 Kd1 Rb6 Ke1 Qc1+ Nd1 
 31	+0.28!	3.02G	2:34.91	h5! 
 31	+0.51?	2.84G	2:25.60	h5 a3? 
 31	+0.22?	2.40G	2:02.81	h5 a3? 
 31	+0.10!	2.00G	1:42.36	h5! 
 31	+0.12!	1.93G	1:38.81	h5! 
 31	+0.20!	1.92G	1:38.39	h5! 
 30	+0.27 	1.85G	1:34.74	h5 a3 Qb6 Kb1 b4 axb4 Qxb4 e5 Bb7 Qf2 O-O-O Bg2 d5 Bf3 Qb6 Be2 Kb8 Rg5 Qc7 Rxh5 Rxh5 Bxh5 Ne4 Nxe4 dxe4 Qe3 Rc8 Qb3 Ka8 Bxf7 Qxf7 Nxe6 Bh6 c3 
 30	+0.12?	1.47G	1:15.31	h5 a3? 
 30	+0.03!	1.38G	1:11.00	h5! 
 30	+0.14!	1.25G	1:04.35	h5! 
 30	+0.21?	1.19G	1:01.46	h5 Qe2? 
 29	+0.14 	1.08G	0:55.70	h5 Qe2 Nd7 e5 Nxe5 Bg2 Bb7 Ncxb5 axb5 Qxb5+ Kd8 Bxb7 Rxa2 Kb1 Qa5 c4 Ra1+ Kc2 Qxb5 Nxb5 Rxd1 Rxd1 Rh6 c5 Rxf6 c6 Rf2+ Kb3 Nxc6 Bxc6 d5 h4 Bg7 Nc3 Bf6 Ra1 Ke7 Bb5 Bxh4 Ra7+ Kf6 
 29	+0.22!	953.5M	0:49.40	h5! 
 29	+0.33!	944.7M	0:48.95	h5! 
 29	+0.40?	897.4M	0:46.52	h5 Qe2? 
 28	+0.33 	830.2M	0:43.02	h5 Qe2 Nd7 e5 Nxe5 Bg2 Bb7 Ncxb5 axb5 Qxb5+ Kd8 Bxb7 Rxa2 Kb1 Qa5 c4 Ra1+ Kc2 Qxb5 Nxb5 Rxd1 Rxd1 Nxc4 Ra1 Ne5 Ra7 d5 b4 Rg8 Kb3 Ng4 Bc6 Nxf6 Rxf7 
 28	+0.25?	779.6M	0:40.40	h5 Qe2? 
 28	+0.18?	724.9M	0:37.54	h5 Qe2? 
 27	+0.10 	661.7M	0:34.22	h5 Qe2 Nd7 e5 Nxe5 Bg2 Bb7 Ncxb5 axb5 Qxb5+ Qd7 Qxb7 Qxb7 Bxb7 Rxa2 Kb1 Ra7 Be4 Kd7 Rg5 Kc8 Nc6 Nxc6 Bxc6 Rc7 Bf3 d5 Rdg1 Bd6 Rxh5 Rxh5 Bxh5 Be5 Rg8+ Kb7 Rf8 Bxf6 
 26	+0.12 	637.3M	0:32.93	h5 Qe2 Nd7 e5 Nxe5 Bg2 Bb7 Ncxb5 axb5 Qxb5+ Kd8 Bxb7 Rxa2 Kb1 Qa5 c4 Ra1+ Kc2 Qxb5 Nxb5 Rxd1 Rxd1 Rh6 c5 Rxf6 c6 Rf2+ Kb3 Nxc6 Bxc6 d5 h4 Bg7 Nd4 Rf4 Nb5 Rxh4 Ra1 Ke7 Ra7+ Kf6 
 26	+0.18?	615.9M	0:31.80	h5 Qe2? 
 26	+0.11?	574.4M	0:29.62	h5 Qe2? 
 25	+0.04!	397.3M	0:20.46	h5! 
 25	+0.33!	279.9M	0:14.37	h5! 
 25	+0.54!	187.3M	0:09.66	b4! 
 25	+0.76?	184.4M	0:09.51	b4 Nd5? 
 25	+0.69?	171.8M	0:08.88	b4 Nd5? 
 24	+0.62 	152.0M	0:07.89	b4 Nd5 exd5 exd5 Nd7 Nc6 Bb7 Bh3 Ne5 Rge1 Rg8 Qf4 Bxc6 dxc6 Rg6 Kb1 Rb8 Bd7+ Kd8 Bf5 Ke8 Bxg6 hxg6 Rd5 Rb5 Rxb5 axb5 Qd4 Bh6 Qxb4 Bg5 Qxb5 Kd8 Rd1 Bxf6 Rf1 Bg5 Qb7 Qxc6 Qxc6 Nxc6 Rxf7 
 24	+0.49?	142.0M	0:07.38	b4 Nd5? 
 24	+0.27!	135.3M	0:07.04	b4! 
 24	+0.43?	133.9M	0:06.97	b4 Nd5? 
 24	+0.22?	129.1M	0:06.72	b4 Nd5? 
 24	  0.00!	123.6M	0:06.44	b4! 
 24	+0.07?	121.4M	0:06.32	b4 Nd5? 
 23	  0.00 	117.9M	0:06.14	b4 Nd5 exd5 exd5 Nd7 Nc6 Bb7 Bh3 Ne5 Rge1 Rg8 Qh5 Bxc6 dxc6 Rg6 Bd7+ Kd8 Bf5 Rh6 Qe2 Ke8 Bd7+ Kd8 Bf5 
 22	  0.00 	109.7M	0:05.72	b4 Nd5 exd5 exd5 Nd7 Nc6 Bb7 Bh3 Ne5 Rge1 Rg8 Qh5 Bxc6 dxc6 Rg6 Bd7+ Kd8 Bf5 Rh6 Qe2 Ke8 Bd7+ Kd8 Bf5 
 22	+0.28!	106.7M	0:05.56	b4! 
 22	+0.45?	105.4M	0:05.49	b4 Nd5? 
 22	+0.24?	97.7M  	0:05.10	b4 Nd5? 
 22	+0.22?	94.9M  	0:04.96	b4 Nd5? 
 22	+0.15?	88.8M  	0:04.65	b4 Nd5? 
 21	+0.08 	84.0M  	0:04.39	b4 Nd5 exd5 exd5 Nd7 Nc6 Bb7 Bh3 Ne5 Rge1 Rg8 Qh5 Bxc6 dxc6 Rg6 Bd7+ Kd8 Bf5 Ke8 Kb1 Rb8 Bd7+ Kd8 Qxh7 Rxf6 Qh4 Ke7 Rf1 Bg7 Rf4 Rh8 Qg5 Nxd7 cxd7 Rxh2 
 21	+0.22!	82.7M  	0:04.32	b4! 
 21	+0.33!	82.0M  	0:04.28	b4! 
 21	+0.40?	81.6M  	0:04.27	b4 Nd5? 
 20	+0.33 	77.8M  	0:04.07	b4 Nd5 exd5 exd5 Nd7 Nc6 Bb7 Bh3 Ne5 Rge1 Rg8 Qf4 Bxc6 dxc6 Rg6 Kb1 Bh6 Qxb4 Rxf6 Bd7+ Kd8 Qh4 Ke7 b3 Bg7 Rf1 h6 a3 Rb8 Rd5 Rg8 Kb2 Qb6 Ka1
Hi Michael,

I was missing your post. :)

I had no doubt about McBrain. It's an engine I like very much because of all of its options.

Did you plan about including a persistent hash option? Not just as asmFish or Komodo, but something similar to Critter one. That would be a great help for analyses.

Thanks!

Rodolfo
F.S.I. Chess Teacher
Top
peter
Posts: 3619
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: Does your engine blunder here?

Post by peter »

Rodolfo Leoni wrote: Writing the analyses for Peter now... :)
I just did let McBrain play a game from 16...h5 with 12 threads and 8G hash each at 180'+30" against asmFish:
[pgn]
[Event "SCHWARZERBLITZ, Lang 180m+30s"]
[Site "SchwarzerBlitz"]
[Date "2017.07.04"]
[Round "1"]
[White "SF-McBrain 2.6 64 POPCNT"]
[Black "asmFishW_2017-06-28_popcnt"]
[Result "1-0"]
[ECO "B99"]
[Annotator "0.30;0.37"]
[PlyCount "73"]
[TimeControl "10800+30"]

{Intel(R) Xeon(R) CPU X5650 @ 2.67GHz 2666 MHz W=40.8 plies; 11.
932kN/s B=40.9 plies; 14.664kN/s} 1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5.
Nc3 a6 6. Bg5 e6 7. f4 Be7 8. Qf3 Qc7 9. O-O-O Nbd7 10. g4 b5 11. Bxf6 Nxf6 12.
g5 Nd7 13. f5 Nc5 14. f6 gxf6 15. gxf6 Bf8 16. Rg1 h5 17. a3 {0.30/38 888} Qb6
{0.37/36 549} 18. Rg7 {0.40/38 8 (Kb1)} Rb8 {0.48/36 1023 (b4)} 19. Bh3 {0.72/
39 848} b4 {0.45/42 51} 20. Nd5 {0.31/35 2} exd5 {0.34/37 78} 21. exd5 {0.53/
40 95} Nd7 {0.91/41 839} 22. Bxd7+ {1.07/42 50} Bxd7 {1.06/38 25} 23. a4 {1.37/
41 316} Kd8 {1.26/41 286} 24. Rxf7 {1.40/42 0} Rb7 {2.19/44 1870} 25. Qe4 {1.
29/45 0} Kc8 {2.14/42 324} 26. b3 {1.94/40 297} Bh6+ {2.45/45 442 (Kb8)} 27.
Kb1 {2.84/39 245} Bg4 {2.08/44 80} 28. a5 {2.92/41 22} Qc5 {2.65/39 31} 29. Ne6
{3.07/44 232} Qxa5 {2.72/45 182} 30. Rxb7 {3.07/48 0} Kxb7 {2.99/40 91} 31. Rd4
{3.32/46 292} Rc8 {2.86/46 76 (Ka8)} 32. Qh7+ {4.71/39 266 (h3)} Kb8 {5.64/41
622} 33. f7 {6.12/43 668} Qb5 {6.40/44 388} 34. Qxh6 {6.61/40 0} Qf1+ {6.90/41
468} 35. Qc1 {6.78/41 0} Qxf7 {6.99/34 46} 36. Rxb4+ {7.03/38 400} Ka8 {7.09/
41 7} 37. Rb6 {6.95/37 93} 1-0
[/pgn]
You will easily find moves to enhance the black side, but maybe there are such for White too
:?:
Peter.
Top
peter
Posts: 3619
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: Does your engine blunder here?

Post by peter »

Ok, White's chances after ...b4 are good to win that, but did you have a look at this game here already too?
http://www.talkchess.com/forum/viewtopi ... 08&t=64510

In the meantime I'll have to go through your lines still too, if this is a forced win alfer ...b4 already for sure
Last edited by peter on Tue Jul 04, 2017 11:33 pm, edited 1 time in total.
Peter.
Top
Rodolfo Leoni
Posts: 545
Joined: Tue Jun 06, 2017 4:49 pm
Location: Italy

Re: Does your engine blunder here?

Post by Rodolfo Leoni »

peter wrote:
Rodolfo Leoni wrote: Writing the analyses for Peter now... :)
I just did let McBrain play a game from 16...h5 with 12 threads and 8G hash each at 180'+30" against asmFish:
[pgn]
[Event "SCHWARZERBLITZ, Lang 180m+30s"]
[Site "SchwarzerBlitz"]
[Date "2017.07.04"]
[Round "1"]
[White "SF-McBrain 2.6 64 POPCNT"]
[Black "asmFishW_2017-06-28_popcnt"]
[Result "1-0"]
[ECO "B99"]
[Annotator "0.30;0.37"]
[PlyCount "73"]
[TimeControl "10800+30"]

{Intel(R) Xeon(R) CPU X5650 @ 2.67GHz 2666 MHz W=40.8 plies; 11.
932kN/s B=40.9 plies; 14.664kN/s} 1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5.
Nc3 a6 6. Bg5 e6 7. f4 Be7 8. Qf3 Qc7 9. O-O-O Nbd7 10. g4 b5 11. Bxf6 Nxf6 12.
g5 Nd7 13. f5 Nc5 14. f6 gxf6 15. gxf6 Bf8 16. Rg1 h5 17. a3 {0.30/38 888} Qb6
{0.37/36 549} 18. Rg7 {0.40/38 8 (Kb1)} Rb8 {0.48/36 1023 (b4)} 19. Bh3 {0.72/
39 848} b4 {0.45/42 51} 20. Nd5 {0.31/35 2} exd5 {0.34/37 78} 21. exd5 {0.53/
40 95} Nd7 {0.91/41 839} 22. Bxd7+ {1.07/42 50} Bxd7 {1.06/38 25} 23. a4 {1.37/
41 316} Kd8 {1.26/41 286} 24. Rxf7 {1.40/42 0} Rb7 {2.19/44 1870} 25. Qe4 {1.
29/45 0} Kc8 {2.14/42 324} 26. b3 {1.94/40 297} Bh6+ {2.45/45 442 (Kb8)} 27.
Kb1 {2.84/39 245} Bg4 {2.08/44 80} 28. a5 {2.92/41 22} Qc5 {2.65/39 31} 29. Ne6
{3.07/44 232} Qxa5 {2.72/45 182} 30. Rxb7 {3.07/48 0} Kxb7 {2.99/40 91} 31. Rd4
{3.32/46 292} Rc8 {2.86/46 76 (Ka8)} 32. Qh7+ {4.71/39 266 (h3)} Kb8 {5.64/41
622} 33. f7 {6.12/43 668} Qb5 {6.40/44 388} 34. Qxh6 {6.61/40 0} Qf1+ {6.90/41
468} 35. Qc1 {6.78/41 0} Qxf7 {6.99/34 46} 36. Rxb4+ {7.03/38 400} Ka8 {7.09/
41 7} 37. Rb6 {6.95/37 93} 1-0
[/pgn]
You will easily find moves to enhance the black side, but maybe there are such for White too
:?:
Interesting game. I must precise, I was only a CM with italian ELO when I retired from chess, 17 years ago. So I can't compete with engines.. :)

But my impression is, 18. ..Rb8 isn't a good move. You can see score drops in favor of W within few plies. 18. ..b4 could have been better. E.g. 19.axb4 Qxb4 20.e5 Bb7 21.Qf4 d5

Blindfold analysis... There could be a huge mistake. But I guess Black avoided the sac Nd5 and is a bit less troubled than it was in game.
F.S.I. Chess Teacher
Top
Rodolfo Leoni
Posts: 545
Joined: Tue Jun 06, 2017 4:49 pm
Location: Italy

Re: Does your engine blunder here?

Post by Rodolfo Leoni »

A propòs, there's another version of asmFishW, released just today. :)
F.S.I. Chess Teacher
Top
peter
Posts: 3619
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: Does your engine blunder here?

Post by peter »

Rodolfo Leoni wrote: 19.axb4 Qxb4 20.e5 Bb7 21.Qf4 d5
Missed this one your posting before mine above answering to your ...b4-lines and couldn't delete it anymore.

Have to leave for today and come back to you tomorrow when I have looked at both of your postings thoroughly.
But just to be clear about this: I don't doubt ...h5 to be better than ...b4, it's just a matter of how much better and how to prove or at least quantify it in a way, cause that yould be necessary to judge the eval and output of the one and the other engine tested with the position.
:arrow:
Peter.
Top
Rodolfo Leoni
Posts: 545
Joined: Tue Jun 06, 2017 4:49 pm
Location: Italy

Re: Does your engine blunder here?

Post by Rodolfo Leoni »

peter wrote: it's just a matter of how much better and how to prove or at least quantify it in a way
I posted a summary of my analyses few hours ago...

16. ..b4 is definitely losing.
16. ..h5 is playable, W could win, B could win, it could be a draw. W is better but the game is open to any result.

See you tomorrow! :)
F.S.I. Chess Teacher
Top

Return to “Computer Chess Club: General Topics”