Evaluation of Nf5 sacrifice in Sicilian....

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

Moderators: hgm, Rebel, chrisw

User avatar
PHALANX OF PAWNS
Posts: 10
Joined: Tue Sep 15, 2015 3:52 pm
Location: Ohio

Evaluation of Nf5 sacrifice in Sicilian....

Post by PHALANX OF PAWNS »

During a blitz game I came across this position. I personally feel that this sacrifice is sound and aggressive and white has the upper hand, but the engines say otherwise. Here it is...

1. e4 {0} c5 {0} 2. Nf3 d6 {0} 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Bc4 b5 7. Bb3 g6 8. f3 Bg7 9. Be3 O-O 10. Qd2 Nbd7 11. O-O-O Bb7 12. Kb1 Rc8 13. g4 Nc5 14. h4 Nxb3 15. axb3 e5 16. Nf5 *

2rq1rk1/1b3pbp/p2p1np1/1p2pN2/4P1PP/1PN1BP2/1PPQ4/1K1R3R b - - 0 16

Also could someone enlighten me on how to post the position graphically on this forum??!!

Thanks, Tony
"The moment you cease to explore the possibilities of every position, is the moment you fail to progress as a chess player!" (The thoughts of a pawn)
User avatar
Graham Banks
Posts: 41473
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: Evaluation of Nf5 sacrifice in Sicilian....

Post by Graham Banks »

[D] 2rq1rk1/1b3pbp/p2p1np1/1p2pN2/4P1PP/1PN1BP2/1PPQ4/1K1R3R b - - 0 16

Just put [D] in front of the fen.
gbanksnz at gmail.com
User avatar
PHALANX OF PAWNS
Posts: 10
Joined: Tue Sep 15, 2015 3:52 pm
Location: Ohio

Re: Evaluation of Nf5 sacrifice in Sicilian....

Post by PHALANX OF PAWNS »

Thanks Graham!
"The moment you cease to explore the possibilities of every position, is the moment you fail to progress as a chess player!" (The thoughts of a pawn)
User avatar
Eelco de Groot
Posts: 4567
Joined: Sun Mar 12, 2006 2:40 am
Full name:   

Re: Evaluation of Nf5 sacrifice in Sicilian....

Post by Eelco de Groot »

Hello Anthony,

What is playable for humans is not always playable for computers. I suppose Black has the bishoppair and then some material that White sacrifices. But maybe the compensation is partly because of a phalanx after 16...gxf5 17. gxf5. And the hole on g7. It only makes sense to detect such things if computers would also be able to play the attack or there is not enough real compensation apart from the theoretical. For the computer that is.

Here is some independant computer analysis from Stockfish relative:


[D]2rq1rk1/1b3pbp/p2p1np1/1p2pN2/4P1PP/1PN1BP2/1PPQ4/1K1R3R b - -

Engine: Rainbow Serpent Sf20150919_001 MOD MP (512 MB)
by Tord Romstad, Marco Costalba and Joona Kiiski

15/26 0:02 +1.14 16...gxf5 17.gxf5 Kh8 18.Qxd6 Re8
19.Bg5 h6 20.Qb4 Qc7 21.Qd6 Qxd6
22.Rxd6 Rc6 23.Rdd1 hxg5 24.hxg5+ Kg8
25.gxf6 Bxf6 26.Nd5 (848.673) 413
.
.
.
34/63 24:08 +0.93 16...gxf5 17.gxf5 Kh8 18.Bg5 Bc6
19.Qxd6 Qxd6 20.Rxd6 Ne8 21.Rd3 f6
22.Be3 Rg8 23.Bc5 Bf8 24.Bxf8 Rxf8
25.b4 Rf7 26.h5 h6 27.b3 Kh7 28.Kb2 Rfc7
29.Ra1 Bb7 (2.917.946.496) 2015

35/63 26:57 +1.00++ 16...gxf5 (3.269.838.533) 2020

35/63 28:26 +1.00 16...gxf5 17.gxf5 Kh8 18.Bg5 Bc6
19.Qxd6 Qxd6 20.Rxd6 Ne8 21.Rd3 f6
22.Be3 Rg8 23.Bc5 Bf8 24.Bxf8 Rxf8
25.b4 Rf7 26.Rg1 Rg7 27.Rgd1 Ra7
28.Na2 Kg7 29.Ra3 Kh6 (3.451.390.285) 2022

36/63 34:26 +1.06++ 16...gxf5 (4.301.495.152) 2081

36/63 37:33 +1.00-- 16...gxf5 17.gxf5 (4.706.073.616) 2087

36/63 40:26 +1.03 16...gxf5 17.gxf5 Kh8 18.Bg5 Bc6
19.Qxd6 Qxd6 20.Rxd6 Ne8 21.Rd3 f6
22.Be3 Rg8 23.Bc5 Bf8 24.Bxf8 Rxf8
25.b4 Rf7 26.Rg1 Rg7 27.Rgd1 Ra7
28.Na2 Kg7 29.Ra3 Kh6 (5.060.534.453) 2085

37/77 63:09 +1.09++ 16...gxf5 (8.340.344.563) 2200

37/77 68:53 +1.03-- 16...gxf5 17.gxf5 (9.218.028.056) 2230

37/77 70:20 +1.09++ 16...gxf5 (9.383.400.443) 2223

37/77 108:30 +1.08 16...gxf5 17.gxf5 Kh8 18.Qxd6 Re8
19.Qa3 Qc7 20.Qd6 Re7 21.h5 Rd7
22.Qxc7 Rdxc7 23.h6 Bf8 24.Rhg1 Rb8
25.Rd2 Re8 26.Bg5 Ng8 27.Nd5 Bxd5
28.Rxd5 f6 29.Bh4 Rc6 (15.661.384.187) 2405

38/77 128:06 +1.02-- 16...gxf5 17.gxf5 (18.526.091.413) 2410

38/77 129:15 +1.08++ 16...gxf5 (18.666.024.692) 2406

38/77 132:05 +1.02-- 16...gxf5 17.gxf5 (19.017.688.184) 2399

38/77 137:44 +1.06 16...gxf5 17.gxf5 Kh8 18.Qxd6 Re8
19.Qa3 Qc7 20.Qd6 Qxd6 21.Rxd6 Rc6
22.Rd2 Bf8 23.Rg1 Rc7 24.Bg5 Nh5
25.Nd5 Bxd5 26.Rxd5 Rec8 27.c3 b4
28.c4 Bc5 29.Rg2 f6 (19.733.775.949) 2387


After for example 16...gxf5 17.gxf5 Kh8 18.Qxd6 Re8
19.Qa3 Qc7

[D]2r1r2k/1bq2pbp/p4n2/1p2pP2/4P2P/QPN1BP2/1PP5/1K1R3R w - -
.
.
.
32/56 7:03 -1.02 20.Qd6 Re7 21.h5 Rd7 22.Qxc7 Rdxc7
23.h6 Bf8 24.Bg5 Be7 25.Bh4 Bc6
26.Bg3 b4 27.Na4 Bxa4 28.bxa4 Rc5
29.Rd2 Ng8 30.Bf2 R5c6 31.Bg3 f6
32.Bf2 Rd6 33.Rxd6 (844.918.841) 1995

33/56 8:14 -1.08-- 20.Qd6 Re7 (985.769.672) 1993

33/56 8:47 -1.02++ 20.Qd6 (1.045.920.438) 1983

33/57 8:54 -1.08-- 20.Qd6 Re7 (1.058.233.607) 1979

33/57 10:52 -1.01++ 20.Qd6 (1.267.187.306) 1941

33/58 11:23 -1.05 20.Qd6 Re7 21.h5 Rd7 22.Qxc7 Rdxc7
23.h6 Bf8 24.Bg5 Be7 25.Bh4 Kg8
26.Bg3 Nd7 27.Nd5 Bxd5 28.Rxd5 f6
29.Rhd1 Nc5 30.Be1 Nb7 31.c3 Rd8
32.Rxd8+ Nxd8 33.Bd2 (1.321.702.583) 1935

34/58 15:52 -1.09 20.Qd6 Re7 21.h5 Rd7 22.Qxc7 Rcxc7
23.Rxd7 Rxd7 24.h6 Bf8 25.Nd1 Rc7
26.Nf2 Nd7 27.c3 a5 28.Nd3 f6 29.Kc2 Rc8
30.Rg1 Bc6 31.Kd2 Ra8 32.b4 a4
33.Kc2 (1.900.258.507) 1994

35/58 20:59 -1.03++ 20.Qd6 (2.620.474.772) 2081

35/58 24:36 -1.09-- 20.Qd6 Bc6 (3.156.606.402) 2138

35/60 26:34 -1.03++ 20.Qd6 (3.409.303.830) 2138

35/60 27:35 -1.10-- 20.Qd6 Bc6 (3.542.866.719) 2139

35/60 28:49 -1.10 20.Qd6 Bc6 21.Qxc7 Rxc7 22.Rd6 Rec8
23.Rhd1 Rd7 24.Rxd7 Bxd7 25.Na2 Bc6
26.Nb4 Bf8 27.Nxc6 Rxc6 28.c3 Be7
29.b4 Kg7 30.Rg1+ Kf8 31.Bh6+ Ke8
32.Bd2 Bd8 33.Kc2 (3.705.687.182) 2142

36/60 31:35 -1.04++ 20.Qd6 (4.081.730.332) 2153

36/60 33:15 -1.10-- 20.Qd6 Bc6 (4.284.180.888) 2146

36/62 35:11 -1.05 20.Qd6 Bc6 21.Qxc7 Rxc7 22.Rd6 Rec8
23.Rhd1 Rd7 24.Rxd7 Bxd7 25.Na2 Bc6
26.Nb4 Bf8 27.Nxc6 Rxc6 28.c3 Be7
29.b4 Kg7 30.Rg1+ Kf8 31.Bh6+ Ke8
32.Bd2 Bd6 33.Rg7 (4.529.876.435) 2145

37/65 49:58 -1.03 20.Qd6 Bc6 21.Qxc7 Rxc7 22.Rd6 Rec8
23.Rhd1 Rd7 24.Rxd7 Bxd7 25.Na2 Bc6
26.Nb4 Bf8 27.Nxc6 Rxc6 28.c3 Bd6
29.b4 Kg7 30.Rg1+ Kf8 31.Bh6+ Ke8
32.Bd2 Be7 33.Bc1 (6.647.781.471) 2216

38/69 59:27 -1.09-- 20.Qd6 Bc6 (7.897.479.293) 2213

38/69 63:17 -1.07 20.Qd6 Bc6 21.Qxc7 Rxc7 22.Rd6 Rec8
23.Rhd1 Rd7 24.Rxd7 Bxd7 25.Na2 Bc6
26.Nb4 Bf8 27.Nxc6 Rxc6 28.c3 Bd6
29.b4 Kg7 30.Rg1+ Kf8 31.Bh6+ Ke8
32.Bg5 Bc7 33.Bd2 (8.424.769.250) 2218


Just adding a few more moves, there does not seem anything critical but I did not see something like moving the queen over to the kingside either, on the contrary White has swapped queens rather early and so this is no longer possible.

[pgn][Event "?"]
[Site "?"]
[Date "2015.10.01"]
[Round "?"]
[White "?"]
[Black "?"]
[Result "*"]

1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Bc4
b5 7. Bb3 g6 8. f3 Bg7 9. Be3 O-O 10. Qd2 Nbd7 11. O-O-O
Bb7 12. Kb1 Rc8 13. g4 Nc5 14. h4 Nxb3 15. axb3 e5 16. Nf5
gxf5 17. gxf5 Kh8 18. Qxd6 Re8 19. Qa3 Qc7 20. Qd6 Bc6
21. Qxc7 Rxc7 22. Rd6 *[/pgn]


[D]4r2k/2r2pbp/p1bR1n2/1p2pP2/4P2P/1PN1BP2/1PP5/1K5R b - -

33/55 9:08 +1.08 22...Rec8 23.Rhd1 Rd7 24.Rxd7 Bxd7
25.Na2 Bc6 26.Nb4 Bf8 27.Nxc6 Rxc6
28.c3 Be7 29.b4 Kg7 30.Rg1+ Kf8
31.Bh6+ Ke8 32.Bc1 Rc8 33.Be3 Bf8
34.Bg5 Nh5 35.Be3 Be7 (977.935.955) 1782

34/55 9:57 +1.08 22...Rec8 23.Rhd1 Rd7 24.Rxd7 Bxd7
25.Na2 Bc6 26.Nb4 Bf8 27.Nxc6 Rxc6
28.c3 Be7 29.b4 Kg7 30.Rg1+ Kf8
31.Bh6+ Ke8 32.Bc1 Rc8 33.Be3 Kd7
34.Rd1+ Kc6 35.Kc2 Nh5 (1.078.914.886) 1806

35/61 16:42 +1.09 22...Rec8 23.Rhd1 Rd7 24.Rxd7 Bxd7
25.Na2 Bc6 26.Nb4 Bf8 27.Nxc6 Rxc6
28.c3 Be7 29.b4 Kg7 30.Rg1+ Kf8
31.Bh6+ Ke8 32.Be3 Rc8 33.Rg7 Bf8
34.Rg1 Kd7 35.Rd1+ Bd6 (2.049.183.671) 2043

36/61 35:39 +1.03-- 22...Rec8 23.Rhd1 (4.753.580.953) 2221

36/61 36:24 +1.09++ 22...Rec8 (4.848.658.846) 2219

36/61 37:22 +1.03-- 22...Rec8 23.Rhd1 (4.970.701.379) 2217

36/64 40:58 +1.10++ 22...Rec8 (5.518.867.417) 2245

36/64 44:21 +1.11 22...Rec8 23.Rhd1 Ne8 24.R6d3 Bf8
25.Bb6 Rb7 26.Be3 Be7 27.Nd5 Bxh4
28.Nb4 Be7 29.Nxa6 Ra8 30.Nc5 Rba7
31.c3 Nf6 32.R3d2 b4 33.cxb4 Ra1+
34.Kc2 Rxd1 35.Rxd1 Bb5 (5.908.117.028) 2219


best move: Re8-c8 time: 49:37.825 min n/s: 2.219.930 nodes: 6.627.852.583
Debugging is twice as hard as writing the code in the first
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan
User avatar
PHALANX OF PAWNS
Posts: 10
Joined: Tue Sep 15, 2015 3:52 pm
Location: Ohio

Re: Evaluation of Nf5 sacrifice in Sicilian....

Post by PHALANX OF PAWNS »

You are absolutely correct Eelco, I have observed that after the capture gxf5, Nh5 is even more demoralizing for white, than Kh8. I believe that with such strong computer chess programs, we will eventually need to rewrite(rethink) many of these sacrificial motifs and also gambit lines! Chess is evolving fast my friend....
"The moment you cease to explore the possibilities of every position, is the moment you fail to progress as a chess player!" (The thoughts of a pawn)
whereagles
Posts: 565
Joined: Thu Nov 13, 2014 12:03 pm

Re: Evaluation of Nf5 sacrifice in Sicilian....

Post by whereagles »

Most Tal-like sacs will still be playable in human games. Only those well-known from theoretical lines might get unplayable, because opponent may have studied it.