Test Position: Nikolic Fischer revisited 19...Bg4!!

[zullil]

I had more or less the same line, albeit without bxc6 spliced in, which makes no key difference here.
Yes confirmed, in this position stockfish has same (relative) difficulties with the evaluation of hxg4
[d]r3r1k1/p2q1p2/2p2np1/P1Pp3p/Q2P1PP1/2N2pP1/1R1B1P1K/5R2 b - - 0 24

This one Stockfish solves, albeit not very quickly:

Code: Select all

info depth 39 seldepth 60 multipv 1 score cp 149 lowerbound nodes 3690338326 nps 16943857 time 217798 pv h5g4
info depth 39 currmove h5g4 currmovenumber 1
info depth 39 seldepth 60 multipv 1 score cp 158 lowerbound nodes 3746867623 nps 17004936 time 220340 pv h5g4
info depth 39 currmove h5g4 currmovenumber 1
info depth 39 seldepth 61 multipv 1 score cp 172 lowerbound nodes 3808608184 nps 17040676 time 223501 pv h5g4
info depth 39 currmove h5g4 currmovenumber 1
info depth 39 seldepth 61 multipv 1 score cp 193 lowerbound nodes 3873297263 nps 17099819 time 226511 pv h5g4
info depth 39 currmove h5g4 currmovenumber 1
info depth 39 seldepth 61 multipv 1 score cp 224 lowerbound nodes 3950608730 nps 17117170 time 230798 pv h5g4
info depth 39 currmove h5g4 currmovenumber 1
info depth 39 seldepth 61 multipv 1 score cp 271 lowerbound nodes 4060640826 nps 17256412 time 235312 pv h5g4
info depth 39 currmove h5g4 currmovenumber 1
info depth 39 seldepth 61 multipv 1 score cp 341 lowerbound nodes 4215958574 nps 17430463 time 241873 pv h5g4
info depth 39 currmove h5g4 currmovenumber 1
info depth 39 seldepth 61 multipv 1 score cp 446 lowerbound nodes 4472781908 nps 17664666 time 253205 pv h5g4
info depth 39 currmove h5g4 currmovenumber 1
info depth 39 currmove f6e4 currmovenumber 6
info depth 39 seldepth 61 multipv 1 score cp 539 nodes 4807305158 nps 18022505 time 266739 pv h5g4 f1h1 g8g7 h2g1 e8h8 h1h8 a8h8 a4c2 d7d8 c2b1 d8g8 b2b7 g8h7 b7f7 g7f7 b1b7 f7e6 b7h7 h8h7 d2e1 f6e4 f4f5 g6f5 a5a6 h7h8 c3a4 h8b8 a4b6 b8e8 b6a4 e8h8 e1a5 e4f2 g1f2 h8h2 f2g1 h2a2 a5b6 a2a4 b6a7 f5f4 a7b8 f4g3 b8g3 a4a6 g1f2 a6a2 f2e1 a2g2 g3e5 e6f5 e1f1
info depth 40 currmove h5g4 currmovenumber 1
info depth 40 seldepth 61 multipv 1 score cp 545 lowerbound nodes 4983871321 nps 18189776 time 273993 pv h5g4
info depth 40 currmove h5g4 currmovenumber 1
info depth 40 seldepth 61 multipv 1 score cp 551 lowerbound nodes 5123645290 nps 18314103 time 279765 pv h5g4
info depth 40 currmove h5g4 currmovenumber 1
info depth 40 seldepth 61 multipv 1 score cp 561 lowerbound nodes 5265730344 nps 18412803 time 285982 pv h5g4
info depth 40 currmove h5g4 currmovenumber 1
info depth 40 seldepth 61 multipv 1 score cp 575 lowerbound nodes 5404454997 nps 18541680 time 291476 pv h5g4
info depth 40 currmove h5g4 currmovenumber 1
info depth 40 seldepth 61 multipv 1 score cp 596 lowerbound nodes 5605715548 nps 18699556 time 299778 pv h5g4
info depth 40 currmove h5g4 currmovenumber 1
info depth 40 seldepth 63 multipv 1 score cp 627 lowerbound nodes 5942671673 nps 18967143 time 313314 pv h5g4
info depth 40 currmove h5g4 currmovenumber 1
info depth 40 currmove d7e6 currmovenumber 17
info depth 40 currmove g6g5 currmovenumber 29
info depth 40 currmove a7a6 currmovenumber 30
info depth 40 seldepth 63 multipv 1 score cp 653 nodes 6211915217 nps 19139497 time 324560 pv h5g4 f1h1 g8g7 h2g1 e8h8 h1h8 a8h8 a4c2 d7d8 c2b1 d8g8 b2b7 g8h7 b7f7 g7f7 b1b7 f7e6 b7h7 h8h7 a5a6 f6e4 c3b1 h7h8 f4f5 g6f5 d2f4 h8e8 g1f1 e6d7 f1g1 e4c5 f4e3 c5a6 b1d2 a6c7 d2b3 c7e6 g1f1 e8b8 b3c1 c6c5 d4c5 d5d4 c5c6 d7c6 e3d2 b8b2 f1e1 c6d6 e1d1 e6c5 d2f4 d6d5 d1e1
info depth 41 currmove h5g4 currmovenumber 1
info depth 41 seldepth 66 multipv 1 score cp 660 lowerbound nodes 6509080283 nps 19314377 time 337007 pv h5g4
info depth 41 currmove h5g4 currmovenumber 1
info depth 41 currmove f6g4 currmovenumber 2
info depth 41 seldepth 66 multipv 1 score cp 661 nodes 6733525025 nps 19420474 time 346723 pv h5g4 f1h1 g8g7 h2g1 e8h8 h1h8 a8h8 a4c2 d7d8 c2b1 d8g8 b2b7 g8h7 b7f7 g7f7 b1b7 f7e6 b7h7 h8h7 d2e3 f6e4 c3a2 h7b7 a2c1 b7b1 g1f1 b1a1 f1e1 a1a5 e3d2 a5a3 d2b4 a3a4 b4d2 a4d4 d2e3 d4b4 c1d3 b4b3 d3c1 b3c3 e3d2 c3c5 c1b3 c5c4 d2e3 c6c5 e1d1 d5d4 b3d2 d4e3 d2c4 e3f2
info depth 42 currmove h5g4 currmovenumber 1
info depth 42 seldepth 66 multipv 1 score cp 667 lowerbound nodes 7084006932 nps 19624320 time 360981 pv h5g4
info depth 42 currmove h5g4 currmovenumber 1
info depth 42 currmove f6g4 currmovenumber 2
info depth 42 seldepth 66 multipv 1 score cp 661 upperbound nodes 7090611483 nps 19614144 time 361505 pv h5g4 f1h1
info depth 42 currmove h5g4 currmovenumber 1
info depth 42 currmove d7g4 currmovenumber 3
info depth 42 currmove f6e4 currmovenumber 21
info depth 42 currmove d7b7 currmovenumber 22
info depth 42 currmove e8e5 currmovenumber 30
info depth 42 currmove e8f8 currmovenumber 32
info depth 42 currmove f6h7 currmovenumber 33
info depth 42 seldepth 66 multipv 1 score cp 651 upperbound nodes 7093030741 nps 19609662 time 361711 pv h5g4 f1h1
info depth 42 currmove h5g4 currmovenumber 1
info depth 42 currmove f6g4 currmovenumber 2
info depth 42 seldepth 66 multipv 1 score cp 655 nodes 7149939409 nps 19625653 time 364316 pv h5g4 f1h1 g8g7 h2g1 e8h8 h1h8 a8h8 a4c2 d7d8 c2b1 d8g8 b2b7 g8h7 b7f7 g7f7 b1b7 f7e6 b7h7 h8h7 a5a6 f6e4 d2e1 e4c3 e1c3 e6f5 c3a5 f5e4 a5d8 e4d4 d8h4 d4c5 g1f1 d5d4 f1e1 c5b5 e1d2 c6c5 d2d3 h7d7 h4f6 c5c4 d3d2 b5a6 f6e5 a6b5 e5f6 a7a5 d2c2 a5a4 f6e5
info depth 43 currmove h5g4 currmovenumber 1
info depth 43 seldepth 66 multipv 1 score cp 661 lowerbound nodes 7347564733 nps 19721141 time 372573 pv h5g4
info depth 43 currmove h5g4 currmovenumber 1
info depth 43 seldepth 66 multipv 1 score cp 667 lowerbound nodes 7580995617 nps 19810429 time 382677 pv h5g4
info depth 43 currmove h5g4 currmovenumber 1
info depth 43 currmove e8e1 currmovenumber 8
info depth 43 seldepth 66 multipv 1 score cp 667 nodes 7811957366 nps 19856987 time 393411 pv h5g4 f1h1 g8g7 h2g1 e8h8 h1h8 a8h8 a4c2 d7d8 c2b1 d8g8 b2b7 g8h7 b7f7 g7f7 b1b7 f7e6 b7h7 h8h7 a5a6 f6e4 c3b1 h7h8 f4f5 g6f5 d2f4 h8e8 g1f1 e6d7 b1d2 e4d2 f4d2 e8b8 f1g1 d7e6 g1h2 b8b1 d2c3 b1f1 c3a5 f1a1 a5c3 a1a3 c3b2 a3a2 b2c1 a2a6 c1d2 a6a4 d2e3 a4a2 e3f4 a7a5 f4e5 a2f2 h2g1 f2a2

[zullil]

I explored this with stockfish and can testify that this is an immedeate win for black in any case after 19..Bg4.
Since if 20.hxg4 hxg4 21.Rh1 (what else?) Nf3+ 22.Bxf3 gxf3 there is no escape
[d]r3r1k1/pp1q1p2/2p2npb/PPPp4/3PpN2/2N1PpP1/1R1B1P1K/3Q3R w - - 0 23

A very long search (still going) by SF suggests that White try Nxg6 here.

Code: Select all

info depth 51 seldepth 101 multipv 1 score cp -61 nodes 1764788220248 nps 21473567 time 82184213 pv h3g4 h5g4 f1h1 g5f3 g2f3 g4f3 f4g6 f7g6 h2g1 h6g5 d1f1 g8f7 f1h3 d7h3 h1h3 a8b8 h3h1 e8c8 b2b1 f7e6 b1b4 f6d7 c3a4 g5d8 b4b2 a7a6 b5c6 b7c6 a4b6 d8g5 d2c3 c8g8 h1h7 g5e7 h7h1 d7f6 b6a4 b8b2 c3b2 e7f8 a4b6 f8g7 b2c3 g8b8 c3b4 b8h8 h1h8 g7h8 b6a8 f6e8 g1h2 g6g5 h2h3 e6f5 g3g4 f5e6 a8b6 h8f6 b6c8 f6d8 c8d6 e8d6 c5d6 e6d7 h3g3

[zullil]

After your continuation by Stockfish "19. ... Bg4 20. hxg4 hxg4 21. Rh1 Nf3+ 22. Bxf3 gxf3 23. Nd3"

23.Nd3 loses to 23...Kg7! as black is not obliged to capture the knight as Stockfish suggests in the continuation you posted.

How about 23. Nxg6, as suggested by a deep SF search (after abandoning the Nd3 line)?

[acase]

After your continuation by Stockfish "19. ... Bg4 20. hxg4 hxg4 21. Rh1 Nf3+ 22. Bxf3 gxf3 23. Nd3"

23.Nd3 loses to 23...Kg7! as black is not obliged to capture the knight as Stockfish suggests in the continuation you posted.

How about 23. Nxg6, as suggested by a deep SF search (after abandoning the Nd3 line)?

Hello Louis,


23.Ng6 loses also the continuation is 23...fxg6 24.Kg1 Bg5 25.Qf1 Kg7! and not 25...Kf7? as suggested in your Stockfish search, this is where my own move is better than the computer's move


[d]r3r3/pp1q2k1/2p2np1/PPPp2b1/3Pp3/2N1PpP1/1R1B1P2/5QKR w - - 0 26

[zullil]

After your continuation by Stockfish "19. ... Bg4 20. hxg4 hxg4 21. Rh1 Nf3+ 22. Bxf3 gxf3 23. Nd3"

23.Nd3 loses to 23...Kg7! as black is not obliged to capture the knight as Stockfish suggests in the continuation you posted.

How about 23. Nxg6, as suggested by a deep SF search (after abandoning the Nd3 line)?

Hello Louis,


23.Ng6 loses also the continuation is 23...fxg6 24.Kg1 Bg5 25.Qf1 Kg7! and not 25...Kf7? as suggested in your Stockfish search, this is where my own move is better than the computer's move


[d]r3r3/pp1q2k1/2p2np1/PPPp2b1/3Pp3/2N1PpP1/1R1B1P2/5QKR w - - 0 26

Will look at your line. In reply to Nxg6, SF at depth 50 says:

Code: Select all

info depth 50 seldepth 71 multipv 1 score cp 83 nodes 55396716122 nps 21027986 time 2634428 pv f7g6 h2g1 h6f8 a5a6 e8e7 b5c6 b7c6 b2b7 d7e6 d1a4 a8d8 g1f1 e6c8 b7b2 d8d7 f1e1 f6g4 h1f1 e7h7 d2c1 h7h2 c3d1 d7c7 e1d2 g4f6 d2c3 c8h3 f1e1 f6g4 c3b3 h3h7 b3a2 h7d7 a2b1 d7c8 e1f1 f8e7 d1c3 e7f6 c3d1 c7f7 c1d2 f6g7 d2c1 f7e7 d1c3 e7c7 c3d1 c7d7 d1c3 g7f6 c3d1 f6g5 d1c3

So SF hasn't found your winning line for Black yet!

[acase]

Will look at your line. In reply to Nxg6, SF at depth 50 says:

Code: Select all

info depth 50 seldepth 71 multipv 1 score cp 83 nodes 55396716122 nps 21027986 time 2634428 pv f7g6 h2g1 h6f8 a5a6 e8e7 b5c6 b7c6 b2b7 d7e6 d1a4 a8d8 g1f1 e6c8 b7b2 d8d7 f1e1 f6g4 h1f1 e7h7 d2c1 h7h2 c3d1 d7c7 e1d2 g4f6 d2c3 c8h3 f1e1 f6g4 c3b3 h3h7 b3a2 h7d7 a2b1 d7c8 e1f1 f8e7 d1c3 e7f6 c3d1 c7f7 c1d2 f6g7 d2c1 f7e7 d1c3 e7c7 c3d1 c7d7 d1c3 g7f6 c3d1 f6g5 d1c3

So SF hasn't found your winning line for Black yet!

Louis,


It makes no sense to search here at this point as the recapture of the knight by black is forced, let me suggest searching after black recaptures the knight to see if white (Stockfish) plays 24.Kg1

[zullil]

Will look at your line. In reply to Nxg6, SF at depth 50 says:

Code: Select all

info depth 50 seldepth 71 multipv 1 score cp 83 nodes 55396716122 nps 21027986 time 2634428 pv f7g6 h2g1 h6f8 a5a6 e8e7 b5c6 b7c6 b2b7 d7e6 d1a4 a8d8 g1f1 e6c8 b7b2 d8d7 f1e1 f6g4 h1f1 e7h7 d2c1 h7h2 c3d1 d7c7 e1d2 g4f6 d2c3 c8h3 f1e1 f6g4 c3b3 h3h7 b3a2 h7d7 a2b1 d7c8 e1f1 f8e7 d1c3 e7f6 c3d1 c7f7 c1d2 f6g7 d2c1 f7e7 d1c3 e7c7 c3d1 c7d7 d1c3 g7f6 c3d1 f6g5 d1c3

So SF hasn't found your winning line for Black yet!

Louis,


It makes no sense to search here at this point as the recapture of the knight by black is forced, let me suggest searching after black recaptures the knight to see if white (Stockfish) plays 24.Kg1

Of course it will. But OK.

Switches to Kg1 very quickly:

Code: Select all

info depth 29 seldepth 38 multipv 1 score cp -6 lowerbound nodes 122863988 nps 8856977 time 13872 pv h2g1
info depth 29 currmove h2g1 currmovenumber 1
info depth 29 currmove b5c6 currmovenumber 2
info depth 29 currmove h1f1 currmovenumber 28
info depth 29 seldepth 38 multipv 1 score cp -13 upperbound nodes 124906646 nps 8881934 time 14063 pv h2g1 h6g5
info depth 29 currmove h2g1 currmovenumber 1
info depth 29 seldepth 39 multipv 1 score cp -3 lowerbound nodes 145000135 nps 9361491 time 15489 pv h2g1
info depth 29 currmove h2g1 currmovenumber 1
info depth 29 seldepth 39 multipv 1 score cp 0 nodes 152680163 nps 9558042 time 15974 pv h2g1 h6g5 d1f1 a8b8 a5a6 b7a6 b5c6 d7c6 b2a2 e8e6 a2a3 b8b2 d2c1 b2b7 c1d2 b7b8 f1a1 c6b7 a3a2 b7c6 a2a3
info depth 30 currmove h2g1 currmovenumber 1
info depth 30 currmove d2e1 currmovenumber 4
info depth 30 seldepth 39 multipv 1 score cp 0 nodes 162622483 nps 9778274 time 16631 pv h2g1 h6g5 d1f1 a8b8 a5a6 b7a6 b5c6 d7c6 b2a2 e8e6 a2a3 b8b2 d2c1 b2b7 c1d2 b7b8 f1a1 c6b7 a1f1 g8f7 h1h3 f7g7 h3h1 g7f7

[acase]

Will look at your line. In reply to Nxg6, SF at depth 50 says:

Code: Select all

info depth 50 seldepth 71 multipv 1 score cp 83 nodes 55396716122 nps 21027986 time 2634428 pv f7g6 h2g1 h6f8 a5a6 e8e7 b5c6 b7c6 b2b7 d7e6 d1a4 a8d8 g1f1 e6c8 b7b2 d8d7 f1e1 f6g4 h1f1 e7h7 d2c1 h7h2 c3d1 d7c7 e1d2 g4f6 d2c3 c8h3 f1e1 f6g4 c3b3 h3h7 b3a2 h7d7 a2b1 d7c8 e1f1 f8e7 d1c3 e7f6 c3d1 c7f7 c1d2 f6g7 d2c1 f7e7 d1c3 e7c7 c3d1 c7d7 d1c3 g7f6 c3d1 f6g5 d1c3

So SF hasn't found your winning line for Black yet!

Louis,


It makes no sense to search here at this point as the recapture of the knight by black is forced, let me suggest searching after black recaptures the knight to see if white (Stockfish) plays 24.Kg1

Of course it will. But OK.

OK, so you do agree on my line then right?

[zullil]

Will look at your line. In reply to Nxg6, SF at depth 50 says:

Code: Select all

info depth 50 seldepth 71 multipv 1 score cp 83 nodes 55396716122 nps 21027986 time 2634428 pv f7g6 h2g1 h6f8 a5a6 e8e7 b5c6 b7c6 b2b7 d7e6 d1a4 a8d8 g1f1 e6c8 b7b2 d8d7 f1e1 f6g4 h1f1 e7h7 d2c1 h7h2 c3d1 d7c7 e1d2 g4f6 d2c3 c8h3 f1e1 f6g4 c3b3 h3h7 b3a2 h7d7 a2b1 d7c8 e1f1 f8e7 d1c3 e7f6 c3d1 c7f7 c1d2 f6g7 d2c1 f7e7 d1c3 e7c7 c3d1 c7d7 d1c3 g7f6 c3d1 f6g5 d1c3

So SF hasn't found your winning line for Black yet!

Louis,


It makes no sense to search here at this point as the recapture of the knight by black is forced, let me suggest searching after black recaptures the knight to see if white (Stockfish) plays 24.Kg1

Of course it will. But OK.

OK, so you do agree on my line then right?

Yes, it appears that White is lost after Kg7. So Nxg6 doesn't work either. Thanks.

[BeyondCritics]

As Andrew has insinuated, this very likely looses to the same plan: Kg7!, Rh8 and mate attack. And even if not, white is firmly nailed into the coffin, it is positionally a lost game.
Nevertheless even in lost positions the defender should play with the greatest precision, to reveal the inner beauty of the position.
Therefore i like the heroic try, albeit it seems futile here.