The eternal knight

[Lyudmil Tsvetkov]

[d]1r3rk1/3q1ppp/3p4/p1pNpP2/PpP1P1P1/1P3Q2/6KP/5R2 w 0 1

What do your engines think about this position?
Black is an exchange and a pawn up, but it is very important that the excellently placed white knight on d5 is impossible to attack (making it possible for it to stay there as long as it deems necessary to, and in a way, immortalizing it).

Would appreciate any comments.

Lyudmil

[lucasart]

That typically looks like an artificially created position to make computers look stupid
I gave my engine DiscoCheck a shot at it, and it's not so pathetic actually. DiscoCheck, after a bit of search, starts to see that white has a lot of compensation for the lack of material. I also put the eval (without any search to see how the computer evaluates the position), to demonstrate that the advantage is actually found by the search, not the static evaluation (which in this case is pathetically materialistic showing white at -3.14 pawns)

Code: Select all

uci
id name DiscoCheck
id author Lucas Braesch
option name Hash type spin default 32 min 1 max 8192
option name Clear Hash type button
uciok
setoption name Hash value 256
position fen 1r3rk1/3q1ppp/3p4/p1pNpP2/PpP1P1P1/1P3Q2/6KP/5R2 w 0 1
eval
 . r . . . r k .
 . . . q . p p p
 . . . p . . . .
 p . p N p P . .
 P p P . P . P .
 . P . . . Q . .
 . . . . . . K P
 . . . . . R . .
1r3rk1/3q1ppp/3p4/p1pNpP2/PpP1P1P1/1P3Q2/6KP/5R2 w - - 0 1
eval = -314
go
info score cp -312 depth 1 nodes 38 time 0 pv f3d3
info score cp -310 depth 2 nodes 96 time 0 pv f3d3 f8c8
info score cp -310 depth 3 nodes 339 time 0 pv g4g5 f8c8 g2h1
info score cp -300 depth 4 nodes 1432 time 1 pv f5f6 g7g6 g4g5 g8h8
info score cp -295 depth 5 nodes 1713 time 2 pv f5f6 g7g6 g4g5 g8h8 g2h1
info score cp -293 depth 6 nodes 2355 time 3 pv f5f6 g7g6 g4g5 g8h8 g2h1 f8c8
info score cp -295 depth 7 nodes 7449 time 9 pv f5f6 g7g6 g4g5 g8h8
info score cp -279 depth 8 nodes 8689 time 10 lowerbound
info score cp -273 depth 8 nodes 11347 time 13 pv f5f6 g7g6 g4g5 g8h8 f3h3 d7c8 h3c8 f8c8
info score cp -286 depth 9 nodes 15688 time 16 pv f5f6 g7g6 g4g5 g8h8
info score cp -281 depth 10 nodes 19987 time 18 pv f5f6 g7g6 g4g5 g8h8 f3h3 d7c6 h3h6 f8g8 g2h1 c6d7
info score cp -282 depth 11 nodes 34388 time 25 pv f5f6 g7g6 g4g5 g8h8
info score cp -287 depth 12 nodes 59646 time 37 pv f5f6 g7g6 g4g5 g8h8
info score cp -284 depth 13 nodes 85150 time 49 pv f5f6 g7g6 g4g5 g8h8
info score cp -268 depth 14 nodes 155034 time 84 lowerbound
info score cp -252 depth 14 nodes 187230 time 100 lowerbound
info score cp -231 depth 14 nodes 392066 time 202 pv f5f6 d7e6 f6g7 g8g7
info score cp -226 depth 15 nodes 780214 time 400 pv f5f6 g7g6 f3h3 g8h8 d5e7 d7b7
info score cp -221 depth 16 nodes 1008484 time 509 pv f5f6 g7g6 f3h3 g8h8 d5e7 d7b7 h3h6
info score cp -205 depth 17 nodes 1298847 time 649 lowerbound
info score cp -189 depth 17 nodes 1661227 time 829 lowerbound
info score cp -157 depth 17 nodes 2301197 time 1150 lowerbound
info score cp -151 depth 17 nodes 2843119 time 1418 pv f5f6 d7e6 f3h3 g8h8 f6g7 h8g7 f1f6
info score cp -151 depth 18 nodes 3269099 time 1625 pv f5f6 d7e6 f3h3 g8h8 f6g7 h8g7
info score cp -167 depth 19 nodes 3587852 time 1771 upperbound
info score cp -135 depth 19 nodes 4621600 time 2293 lowerbound
info score cp -103 depth 19 nodes 5657717 time 2820 lowerbound
info score cp -85 depth 19 nodes 7008382 time 3490 pv f5f6 d7e6 f3h3 g8h8 f6g7 h8g7
info score cp -87 depth 20 nodes 8361390 time 4141 pv f5f6 d7e6 f3h3 g8h8 f6g7 h8g7
info score cp -71 depth 21 nodes 12460012 time 6173 lowerbound
info score cp -71 depth 21 nodes 13043229 time 6438 pv f5f6 d7e6 f3h3 g8h8
info score cp -66 depth 22 nodes 19256740 time 9411 pv f5f6 d7e6 f3h3 g8h8 f6g7 h8g7
info score cp -50 depth 23 nodes 51019893 time 24905 lowerbound
info score cp -45 depth 23 nodes 61555215 time 29967 pv f5f6 d7e6 f3h3 f8e8

What makes the position good for white is not to have a knight on d5. This is not an achievement per se, that is worth 3 pawns. It's the threats following f6! So it's more a problem of tactics rather than position knowledge for computers.

[Lyudmil Tsvetkov]

That typically looks like an artificially created position to make computers look stupid

Hi Lucas.
Believe me, I have spent an awful amount of time with this sole purpose in mind...
Thanks for the DiscoCheck check.
Your advantage is that you know much about programming and a bit about chess, but I only know an indefinite amount of chess.

[Ajedrecista]

Hello Lyudmil:

[d]1r3rk1/3q1ppp/3p4/p1pNpP2/PpP1P1P1/1P3Q2/6KP/5R2 w 0 1

What do your engines think about this position?
Black is an exchange and a pawn up, but it is very important that the excellently placed white knight on d5 is impossible to attack (making it possible for it to stay there as long as it deems necessary to, and in a way, immortalizing it).

Would appreciate any comments.

Lyudmil

This knight outpost looks good to my patzer eyes. This outpost plus the closed pawn chain is enough to compensate the lost exchange and the pawn down. Of course I am not a GM...

Analysis by Critter 1.4 w32 (using four cores) in an Intel i5-760 (2.8 GHz), with 64 MB of hash, running in console mode:

Code: Select all

Critter v1.4 32-bit, by Richard Vida
position fen 1r3rk1/3q1ppp/3p4/p1pNpP2/PpP1P1P1/1P3Q2/6KP/5R2 w 0 1
go infinite

[...]

16/42  00:01   5502339 4698837  +0.00 f6 Rfe8 Qh3 h6 fxg7 Kxg7 Nf6 Qb7 Nh5+ Kf8 Nf6 Kg7
17/42+ 00:01   8899950 4985966  +0.08 f6 Rfe8 Qh3 h6 fxg7 Kxg7 Nf6 Qb7 Nh5+ Kf8 Nf6 Kg7
17/42  00:01   9527794 5004093  +0.00 f6 Rfe8 Qh3 Qd8 g5 Re6 Qh4 g6 Nc7 Qc8 Nxe6 Qxe6 Qh6 Qg4+ Kf2 Qf4+ Kg2 Qg4+
18/42+ 00:02  10947061 5070431  +0.08 f6 Rfe8 Qh3 Qd8 g5 Re6 Qh4 g6 Nc7 Qc8 Nxe6 Qxe6 Qh6 Qg4+ Kf2 Qf4+ Kg2 Qg4+
18/53+ 00:02  13611173 5165530  +0.17 f6 Rfe8 Qh3 Qd8 g5 Re6 Qh4 g6 Nc7 Qc8 Nxe6 Qxe6 Qh6 Qg4+ Kf2 Qf4+ Kg2 Qg4+
18/53  00:02  14169526 5182708  +0.04 f6 Rfe8 Qh3 Qd8 g5 Re6 Qh4 g6 Nc7 Qc8 Nxe6 Qxe6 Rf3 Qe8 Rh3 h5 gxh6/ep Kh7 Rd3 Qc6 Rd5 Re8 Qg5 Rh8 h3 Rd8 Kg3
19/53  00:03  16435249 5255915  +0.04 f6 Rfe8 Qh3 Qd8 g5 Re6 Qh4 g6 Nc7 Qc8 Nxe6 Qxe6 Rf3 Qe8 Rh3 h5 gxh6/ep Kh7 Rd3 Qc6 Rd5 Re8 Qg5 Rh8 h3 Rd8 Kg3
20/53  00:03  21553735 5460789  +0.05 f6 Rfe8 Qh3 Qd8 g5 Re6 Qh4 g6 Nc7 Qc8 Nxe6 Qxe6 Rf3 Qe8 Rh3 h5 gxh6/ep Kh7 Rd3 Qc6 Rd5 Re8 Qg5 Rh8 Qd2 Qd7
21/53  00:06  38099264 5815793  +0.04 f6 Rfe8 Qh3 Qd8 g5 Re6 Qh4 g6 Nc7 Qc8 Nxe6 Qxe6 Rf3 Qe8 Rh3 h5 gxh6/ep Kh7 Rd3 Qc6 Rd5 Re8 Qg5 Rh8 h3 Rd8 Kg3
22/54  00:11  68214992 6081393  +0.03 f6 Rfe8 Qh3 Qd8 g5 Re6 Qh4 g6 Nc7 Qc8 Nxe6 Qxe6 Rf3 h5 gxh6/ep Kh7 Rg3 Qd7 Qg5 Re8 Rh3 Re6 Rd3 Qd8 Rf3 Qa8 Re3
23/54  00:18 112997013 6167959  +0.05 f6 Rfe8 Qh3 Qd8 g5 Re6 Qh4 g6 Nc7 Qc8 Nxe6 Qxe6 Rf3 h5 gxh6/ep Kh7 Rg3 Qd7 Qg5 Re8 Rh3 Rd8 Rd3 Qc6 Rd5 Rh8 Qd2 Qd7
24/56+ 00:36 225571498 6235735  +0.14 f6 Rfe8 Qh3 Qd8 g5 Re6 Qh4 g6 Nc7 Qc8 Nxe6 Qxe6 Rf3 h5 gxh6/ep Kh7 Qg5 Qd7 Rd3 Qc6 Rd5 Qd7 Rd1 Re8 h3 Re6 Rd2 Qd8 Rf2
24/56+ 00:49 307420763 6204128  +0.22 f6 Rfe8 Qh3 Qd8 g5 Re6 Qh4 g6 Nc7 Qc8 Nxe6 Qxe6 Rf3 h5 gxh6/ep Kh7 Qg5 Qd7 Rd3 Qc6 Rd5 Qd7 Rd1 Re8 h3 Re6 Rd2 Qd8 Rf2
24/56+ 01:32 575248965 6219647  +0.35 f6 Rfe8 Qh3 Qd8 g5 Re6 Qh4 g6 Nc7 Qc8 Nxe6 Qxe6 Rf3 h5 gxh6/ep Kh7 Qg5 Qd7 Rd3 Qc6 Rd5 Qd7 Rd1 Re8 h3 Re6 Rd2 Qd8 Rf2
24/56+ 02:09 809745337 6244132  +0.54 f6 Rfe8 Qh3 Qd8 g5 Re6 Qh4 g6 Nc7 Qc8 Nxe6 Qxe6 Rf3 h5 gxh6/ep Kh7 Qg5 Qd7 Rd3 Qc6 Rd5 Qd7 Rd1 Re8 h3 Re6 Rd2 Qd8 Rf2
24/80+ 03:11 1189360161 6221804  +0.83 f6 Rfe8 Qh3 Qd8 g5 Re6 Qh4 g6 Nc7 Qc8 Nxe6 Qxe6 Rf3 h5 gxh6/ep Kh7 Qg5 Qd7 Rd3 Qc6 Rd5 Qd7 Rd1 Re8 h3 Re6 Rd2 Qd8 Rf2
24/80  05:08 1929417872 6261782  +1.12 f6 g5 Qh3 Kh8 Qh5 Rg8 Rf3 Rg6 Rh3 h6 Qxh6+ Rxh6 Rxh6+ Kg8 Ne7+ Qxe7 fxe7 Re8 Rxd6 Rxe7 Kf2 Re6 Rd5 Rf6+ Kg3 Rf1 Rxe5 f6 Rxc5 Rb1 Rxa5 Rxb3+ Kf2 Kf7 Rb5 Ke6 a5 Kd6 Ke2 Ra3 Rxb4 Rxa5 Rb6+ Ke7 Rb2 Kd6 Rd2+ Kc6 Ke3 Ra3+ Rd3
25/80  05:33 2085420512 6262016  +1.07 f6 g5 Qh3 Kh8 Qh5 Rg8 Rf3 Rg6 Rh3 h6 Qxh6+ Rxh6 Rxh6+ Kg8 Ne7+ Qxe7 fxe7 Re8 Rxd6 Rxe7 Rd5 f6 Rxc5 Ra7 Kf3 Kf7 Rb5 Ke7 Rd5 Ke6 h3 Rh7 Kg2 Rd7 Rxd7 Kxd7 Kf3 Kc6 Kf2 Kc5 Ke3 Kd6 Kd2 Kc6 Kd3 Kc5 Ke3
stop
bestmove f6 ponder g5

time: 499584 nodes: 3137180047 evals: 2286308309 knps: 6279
phash: 85.50% evalcache: 15.60% lazy: 49.66% eg_recog: 6974954
splits: 191406

Critter 1.4 evaluates this position around +1 for black at depth 25/80. It would be an unpleasant position for myself with either side.

Regards from Spain.

Ajedrecista.

[Adam Hair]

The latest beta of Gaviota agrees with DiscoCheck:

Code: Select all

Gaviota_0.86_beta3:
   1	00:00	          39	39	-1.66	1.h4
   2	00:00	         299	299	-1.49	1.h4 h6
   3	00:00	         742	742	-1.37	1.f6 g6 2.h3
   4	00:00	       1.003	1.003	-1.29	1.f6 g6 2.h3 Kh8
   5	00:00	       1.372	1.372	-1.33	1.f6 g6 2.h3 Kh8 3.Qe3
   6	00:00	       3.539	3.539	-1.29	1.f6 g5 2.Ne7+ Kh8 3.h3 Rfe8 4.Qe3
   7	00:00	       7.875	7.875	-1.41	1.f6 g6 2.h3 h6 3.Qe3 Kh7 4.Kh2
   8	00:00	       9.712	971.200	-1.32	1.f6 g6 2.h3 h6 3.Kh2 Kh7 4.Re1 Qc6
   8	00:00	      11.566	1.156.600	-1.32	1.f6 g6 2.h3 h6 3.Kh2 Kh7 4.Re1 Qc6
   9	00:00	      24.009	800.300	-1.37	1.f6 g6 2.g5 Kh8 3.h4 Rg8 4.Rd1 Rge8 5.Ne7
   9	00:00	      64.729	462.350	-1.23	1.g5 Kh8 2.h4 Rg8 3.Kh2 Rgd8 4.Rg1 Rf8 5.Re1
   9	00:00	      65.024	382.494	-1.23	1.g5 Kh8 2.h4 Rg8 3.Kh2 Rgd8 4.Rg1 Rf8 5.Re1
  10	00:00	     115.486	461.944	-1.17	1.g5 Kh8 2.h4 Qc6 3.f6 g6 4.Kh2 Qd7 5.Re1 Rg8 <EMPTY> [TT]
  10	00:00	     120.340	462.846	-1.17	1.g5 Kh8 2.h4 Qc6 3.f6 g6 4.Kh2 Qd7 5.Re1 Rg8 <EMPTY> [TT]
  11	00:00	     216.188	469.973	-1.12	1.g5 Kh8 2.h4 Qc6 3.Kh2 Rfe8 4.Re1 Qd7 5.Rg1 Red8 [TT]
  11	00:00	     222.063	435.417	-1.12	1.g5 Kh8 2.h4 Qc6 3.Kh2 Rfe8 4.Re1 Qd7 5.Rg1 Red8 [TT]
  12	00:00	     466.322	490.865	-1.07	1.g5 Kh8 2.h4 Rfe8 3.Kh2 Rbd8 4.Rd1 Kg8 5.f6 g6 6.h5 Kh8
  12	00:00	     468.144	492.783	-1.07	1.g5 Kh8 2.h4 Rfe8 3.Kh2 Rbd8 4.Rd1 Kg8 5.f6 g6 6.h5 Kh8
  13	00:01	     532.964	512.465	-1.08	1.g5 Kh8 2.h4 Rfe8 3.Kh2 Rbd8 4.Rd1 Rf8 5.Nb6 Qc6 6.Nd5 h6 7.Ne7
  13	00:01	     569.658	517.870	-1.08	1.g5 Kh8 2.h4 Rfe8 3.Kh2 Rbd8 4.Rd1 Rf8 5.Nb6 Qc6 6.Nd5 h6 7.Ne7
  14	00:01	     733.597	495.673	-1.08	1.g5 Kh8 2.h4 Rfe8 3.Kh2 Rbd8 4.Rd1 Rf8 5.Nb6 Qc6 6.Nd5 [>]
  14	00:01	     752.771	498.523	-1.08	1.g5 Kh8 2.h4 Rfe8 3.Kh2 Rbd8 4.Rd1 Rf8 5.Nb6 Qc6 6.Nd5 [>]
  15	00:03	   1.851.825	484.770	-1.05	1.g5 Kh8 2.h4 Rfe8 3.Kh2 Rbd8 4.Rd1 Rf8 5.f6 g6 6.Ne7 Rb8 7.h5 Rb6 8.h6
  15	00:03	   1.873.852	486.714	-1.05	1.g5 Kh8 2.h4 Rfe8 3.Kh2 Rbd8 4.Rd1 Rf8 5.f6 g6 6.Ne7 Rb8 7.h5 Rb6 8.h6
  16	00:06	   3.326.848	513.402	-1.01	1.g5 Kh8 2.h4 Rfe8 3.Kh2 Rg8 4.Re1 Rgd8 5.h5 Rg8 6.f6 g6 7.Ne7 Rge8 8.h6 Rf8
  16	00:06	   3.441.238	511.328	-1.01	1.g5 Kh8 2.h4 Rfe8 3.Kh2 Rg8 4.Re1 Rgd8 5.h5 Rg8 6.f6 g6 7.Ne7 Rge8 8.h6 Rf8
  17	00:10	   5.451.237	532.867	-1.01	1.g5 Kh8 2.h4 Rg8 3.Kh2 Rbd8 4.Qh5 Rdf8 5.Qf3 Rb8 6.Re1 [TT]
  17	00:10	   5.516.233	539.221	-1.01	1.g5 Kh8 2.h4 Rg8 3.Kh2 Rbd8 4.Qh5 Rdf8 5.Qf3 Rb8 6.Re1 [TT]
  18	00:11	   6.611.829	551.905	-1.01	1.g5 Kh8 2.h4 Rg8 3.Kh2 Rbd8 4.Qh5 Rdf8 5.Qf3 Rb8 6.Re1 [TT]
  18	00:12	   6.727.898	550.114	-1.01	1.g5 Kh8 2.h4 Rg8 3.Kh2 Rbd8 4.Qh5 Rdf8 5.Qf3 Rb8 6.Re1 [TT]
  19	00:16	   9.393.125	553.187	-1.01	1.g5 Kh8 2.h4 Rg8 3.Kh2 Rbd8 4.Qh5 Rdf8 5.Qf3 Rb8 6.Re1 [TT]
  19	00:29	  15.390.578	521.537	-0.62	1.f6 :-)
  19	01:35	  50.556.008	532.168	 0.00	1.f6 g6 2.Ne7+ Kh8 3.Qh3 Qb7 4.Re1 Qd7 5.Rf1
  19	01:35	  50.593.945	532.567	 0.00	1.f6 g6 2.Ne7+ Kh8 3.Qh3 Qb7 4.Re1 Qd7 5.Rf1

[lucasart]

That typically looks like an artificially created position to make computers look stupid

Hi Lucas.
Believe me, I have spent an awful amount of time with this sole purpose in mind...
Thanks for the DiscoCheck check.
Your advantage is that you know much about programming and a bit about chess, but I only know an indefinite amount of chess.

Your position is definitely very interesting. The more I analyze it, and try variations, the more I understand the beauty of it.

[Lyudmil Tsvetkov]

Muchas gracias, ajedrecista.
Now I think I did not set up the right position, as I did it in a hurry.
But the other positions I posted should be more relevant, as similar
structures have frequently occurred in my engine games.

En rio revuelto - fortuna de pescadores

[Lyudmil Tsvetkov]

Hi Lucas.
Now I think I got the wrong position.

[Mdannib]

Very interesting position....
This is the analysis done with the current beta of BlackMamba

Code: Select all

FEN: 1r3rk1/3q1ppp/3p4/p1pNpP2/PpP1P1P1/1P3Q2/6KP/5R2 w - - 0 1 

BlackMamba_1.1.295 beta x64:
  1/4	00:00	          88	0	-1,49	1.Rd1
  2/9	00:00	         284	0	-1,39	1.Rd1 h6
  3/9	00:00	         475	0	-1,49	1.Rd1 h6 2.Qe3
  4/11	00:00	         848	0	-1,39	1.Rd1 h6 2.f6 g6
  4/11	00:00	         930	0	-1,39	1.Rd1 h6 2.f6 g6
  5/12	00:00	       1.541	0	-1,49	1.Rd1 f6 2.Qe3 Kh8 3.g5
  5/12	00:00	       1.643	0	-1,49	1.Rd1 f6 2.Qe3 Kh8 3.g5
  6/12	00:00	       2.128	0	-1,41	1.Rd1 f6 2.Qe3 Kh8 3.g5 fxg5 4.Qxg5
  6/12	00:00	       2.231	0	-1,41	1.Rd1 f6 2.Qe3 Kh8 3.g5 fxg5 4.Qxg5
  7/12	00:00	       3.216	0	-1,46	1.Rd1 f6 2.Qe3 Kh8 3.Qe2 Rg8 4.Qe3
  7/15	00:00	       9.184	0	-1,39	1.g5 Rfe8 2.Rd1 Rb7 3.Qh5 Qc6 4.f6 g6
  7/15	00:00	      12.233	815.000	-1,39	1.g5 Rfe8 2.Rd1 Rb7 3.Qh5 Qc6 4.f6 g6
  8/15	00:00	      15.085	1.005.000	-1,33	1.g5 Rfd8 2.Qe3 Rb7 3.Rd1 Rf8 4.Qe2
  8/15	00:00	      15.842	1.056.000	-1,33	1.g5 Rfd8 2.Qe3 Rb7 3.Rd1 Rf8 4.Qe2
  9/17	00:00	      20.040	1.336.000	-1,39	1.g5 Rb7 2.Rd1 Re8 3.Rd2 Qc6 4.f6 g6
  9/19+	00:00	      28.078	905.000	-1,13	1.f6
  9/19	00:00	      31.606	1.019.000	-1,07	1.f6 g6 2.Qh3 Kh8 3.Qh4 Rfe8 4.Rd1 Rb7
  9/19	00:00	      32.569	1.050.000	-1,07	1.f6 g6 2.Qh3 Kh8 3.Qh4 Rfe8 4.Rd1 Rb7
 10/21	00:00	      48.139	1.024.000	-1,17	1.f6 g6 2.Qh3 Kh8 3.Qh4 Rg8 4.Rd1 g5 5.Qh3 Rg6 6.Rd2 h6 7.Qf3
 10/21	00:00	      49.501	1.053.000	-1,17	1.f6 g6 2.Qh3 Kh8 3.Qh4 Rg8 4.Rd1 g5 5.Qh3 Rg6 6.Rd2 h6 7.Qf3
 11/22	00:00	      54.784	883.000	-1,17	1.f6 g6 2.Qh3 Kh8 3.Qh4 Rg8 4.Rd1 g5 5.Qh3 Rg6 6.Rd2 h6 7.Qf3
 11/22	00:00	      57.720	930.000	-1,17	1.f6 g6 2.Qh3 Kh8 3.Qh4 Rg8 4.Rd1 g5 5.Qh3 Rg6 6.Rd2 h6 7.Qf3
 12/22	00:00	      65.359	1.054.000	-1,17	1.f6 g6 2.Qh3 Kh8 3.Qh4 Rg8 4.Rd1 g5 5.Qh3 Rg6 6.Rd2 h6 7.Qf3
 12/22	00:00	      72.314	927.000	-1,17	1.f6 g6 2.Qh3 Kh8 3.Qh4 Rg8 4.Rd1 g5 5.Qh3 Rg6 6.Rd2 h6 7.Qf3
 13/27	00:00	     119.468	955.000	-1,15	1.f6 g6 2.Qh3 Kh8 3.Qh4 Rg8 4.Rd1 g5 5.Qh3 Rg6 6.Rd2 Rbg8 7.Nb6 Qb7 8.Nd5 h6
 13/27	00:00	     134.073	957.000	-1,15	1.f6 g6 2.Qh3 Kh8 3.Qh4 Rg8 4.Rd1 g5 5.Qh3 Rg6 6.Rd2 Rbg8 7.Nb6 Qb7 8.Nd5 h6
 14/27	00:00	     154.045	1.100.000	-1,15	1.f6 g6 2.Qh3 Kh8 3.Qh4 Rg8 4.Rd1 g5 5.Qh3 Rg6 6.Rd2 Rbg8 7.Nb6 Qb7 8.Nd5 h6
 14/27	00:00	     180.295	1.054.000	-1,15	1.f6 g6 2.Qh3 Kh8 3.Qh4 Rg8 4.Rd1 g5 5.Qh3 Rg6 6.Rd2 Rbg8 7.Nb6 Qb7 8.Nd5 h6
 15/33	00:00	     450.263	1.254.000	-1,09	1.f6 g6 2.Rd1 Rfe8 3.Qh3 Kh8 4.Qh4 Red8 5.Ne7 Rb7 6.Nd5 Rf8 7.h3 Qc6 8.Qh6 Rg8 9.Rd2
 15/33	00:00	     510.836	1.261.000	-1,09	1.f6 g6 2.Rd1 Rfe8 3.Qh3 Kh8 4.Qh4 Red8 5.Ne7 Rb7 6.Nd5 Rf8 7.h3 Qc6 8.Qh6 Rg8 9.Rd2
 16/36+	00:00	   1.135.521	1.400.000	-0,89	1.f6 g6 2.Rd1 Kh8 3.Rd3 h6 4.Qh3 Kh7 5.Qh4 Rh8 6.Ne7 Qb7 7.g5 h5 8.Rxd6 Rbd8
 16/40+	00:01	   1.609.158	1.412.000	-0,69	1.f6 g6 2.Ne7+ Kh8 3.Qh3 Rbe8 4.Rf3 Rxe7 5.Qh6 Qxg4+ 6.Kf2 Rg8 7.fxe7 Qe6 8.Rh3 Rg7 9.Qg5 Kg8 10.Rf3 f6 11.Qxf6 Qxe7 12.Ke3 g5 13.Qxe7 Rxe7
 16/47+	00:02	   2.924.952	1.442.000	-0,49	1.f6 g6 2.Ne7+ Kh8 3.Qh3 Rfe8 4.Qh4 Rxe7 5.fxe7 Kg8 6.Rf3 Re8 7.Rh3 h6 8.Qxh6 Qxg4+ 9.Kf2 Qf4+ 10.Qxf4 exf4 11.Rd3 Rxe7 12.Kf3 Re6 13.Kxf4 Kg7 14.h4 Kf6 15.Rd1 Ke7 16.Rd2 Re5 17.Kf3 Ke6
 16/47	00:02	   3.748.570	1.465.000	-0,43	1.f6 g6 2.Ne7+ Kh8 3.Qh3 Qb7 4.Re1 Rg8 5.Nxg8 Rxg8 6.Qh4 Qd7 7.h3 g5 8.Qf2 Rg6 9.Rd1 Qe6 10.Rf1 Kg8
 16/47	00:02	   3.774.140	1.457.000	-0,43	1.f6 g6 2.Ne7+ Kh8 3.Qh3 Qb7 4.Re1 Rg8 5.Nxg8 Rxg8 6.Qh4 Qd7 7.h3 g5 8.Qf2 Rg6 9.Rd1 Qe6 10.Rf1 Kg8
 17/47	00:05	   7.255.626	1.431.000	-0,27	1.f6 g5 2.Qh3 Rbe8 3.Rf5 Kh8 4.Rxg5 Rg8 5.Rf5 Rxg4+ 6.Qxg4 Rg8 7.Qxg8+ Kxg8 8.Ne7+ Kh8 9.Rg5 Qd8 10.Kf3 Qf8 11.Rg7 h5 12.Ke3 h4 13.Rg8+ Qxg8 14.Nxg8 Kxg8
 17/47	00:05	   7.321.595	1.430.000	-0,27	1.f6 g5 2.Qh3 Rbe8 3.Rf5 Kh8 4.Rxg5 Rg8 5.Rf5 Rxg4+ 6.Qxg4 Rg8 7.Qxg8+ Kxg8 8.Ne7+ Kh8 9.Rg5 Qd8 10.Kf3 Qf8 11.Rg7 h5 12.Ke3 h4 13.Rg8+ Qxg8 14.Nxg8 Kxg8
 18/47	00:07	  10.489.326	1.418.000	-0,21	1.f6 Rfe8 2.fxg7 Qe6 3.h3 Qg6 4.Qd3 Rbd8 5.Nf6+ Kxg7 6.Nxe8+ Rxe8 7.Rf5 Rg8 8.Qd5 Qe6 9.Qb7
 18/47	00:07	  10.580.175	1.418.000	-0,21	1.f6 Rfe8 2.fxg7 Qe6 3.h3 Qg6 4.Qd3 Rbd8 5.Nf6+ Kxg7 6.Nxe8+ Rxe8 7.Rf5 Rg8 8.Qd5 Qe6 9.Qb7
 19/47+	00:13	  17.395.411	1.322.000	-0,01	1.f6 Rfe8 2.g5 g6 3.Kg3
 19/63	00:21	  29.912.756	1.362.000	 0,00	1.f6 Rfe8 2.fxg7 Qe6 3.Nf6+ Kxg7 4.Qe3 h6 5.Nh5+ Kh7 6.Nf6+ Kg7
 19/63	00:22	  30.022.846	1.363.000	 0,00	1.f6 Rfe8 2.fxg7 Qe6 3.Nf6+ Kxg7 4.Qe3 h6 5.Nh5+ Kh7 6.Nf6+ Kg7
 20/63	00:24	  34.140.476	1.374.000	 0,00	1.f6 Rfe8 2.fxg7 Qe6 3.Nf6+ Kxg7 4.Qe3 h6 5.Nh5+ Kh7 6.Nf6+ Kg7
 20/63	00:24	  34.256.611	1.375.000	 0,00	1.f6 Rfe8 2.fxg7 Qe6 3.Nf6+ Kxg7 4.Qe3 h6 5.Nh5+ Kh7 6.Nf6+ Kg7
 21/63	00:32	  45.040.307	1.390.000	 0,00	1.f6 Rfe8 2.fxg7 Qe6 3.Nf6+ Kxg7 4.Qe3 h6 5.Nh5+ Kh7 6.Nf6+ Kg7
 21/63	00:32	  45.231.570	1.390.000	 0,00	1.f6 Rfe8 2.fxg7 Qe6 3.Nf6+ Kxg7 4.Qe3 h6 5.Nh5+ Kh7 6.Nf6+ Kg7
 22/63	00:40	  56.459.968	1.407.000	 0,00	1.f6 Rfe8 2.fxg7 Qe6 3.Nf6+ Kxg7 4.Qe3 h6 5.Nh5+ Kh7 6.Nf6+ Kg7
 22/63	00:40	  56.858.909	1.408.000	 0,00	1.f6 Rfe8 2.fxg7 Qe6 3.Nf6+ Kxg7 4.Qe3 h6 5.Nh5+ Kh7 6.Nf6+ Kg7
 23/63	00:55	  78.105.214	1.415.000	 0,00	1.f6 Rfe8 2.fxg7 Qe6 3.Nf6+ Kxg7 4.Qe3 h6 5.Nh5+ Kh7 6.Nf6+ Kg7
 23/63	00:55	  78.722.253	1.415.000	 0,00	1.f6 Rfe8 2.fxg7 Qe6 3.Nf6+ Kxg7 4.Qe3 h6 5.Nh5+ Kh7 6.Nf6+ Kg7
 24/63	01:22	 115.958.730	1.410.000	 0,00	1.f6 Rfe8 2.fxg7 Qe6 3.Nf6+ Kxg7 4.Qe3 h6 5.Nh5+ Kh7 6.Nf6+ Kg7
 24/63	01:23	 117.454.866	1.411.000	 0,00	1.f6 Rfe8 2.fxg7 Qe6 3.Nf6+ Kxg7 4.Qe3 h6 5.Nh5+ Kh7 6.Nf6+ Kg7
 24/63	01:58	 166.147.020	1.407.000	 0,00	1.f6 Rfe8 2.fxg7 Qe6 3.Nf6+ Kxg7 4.Qe3 h6 5.Nh5+ Kh7 6.Nf6+ Kg7

[acase]

[d]1r3rk1/3q1ppp/3p4/p1pNpP2/PpP1P1P1/1P3Q2/6KP/5R2 w 0 1

What do your engines think about this position?
Black is an exchange and a pawn up, but it is very important that the excellently placed white knight on d5 is impossible to attack (making it possible for it to stay there as long as it deems necessary to, and in a way, immortalizing it).

Would appreciate any comments.

Lyudmil

This is an interesting position, however it looks like Houdini 3 in tactical mode may have found a way to win, but I am not sure it is choosing the best continuation for black in it's analysis (after 1.f6 it looks like 1...Rfe8 is better than 1...h6, that way the rook can take the knight if it moves to e7).

29/86+ 45:39 20,309,108,646 7,413,000 +2.15 f5-f6

29/86+ 1:00:37 26,934,336,341 7,405,000 +2.36 f5-f6

29/86 1:17:45 34,595,081,301 7,415,000 +2.41 f5-f6 h7-h6 Nd5-e7+ Kg8-h7 Ne7-f5 d6-d5 f6xg7 Rf8-g8 e4xd5 Rb8-b6 Qf3-e4 Rb6-g6 h2-h3 Qd7-d8 d5-d6 Qd8-g5 Nf5-e7 Qg5-d2+ Rf1-f2 Qd2xd6 Rf2xf7 Qd6-e6 Ne7xg6 Qe6xg6 Qe4xg6+ Kh7xg6 Rf7-c7 Rg8-d8 Kg2-f3 Kg6-h7 Kf3-e4 Rd8-d4+ Ke4xe5 Rd4-d3 Rc7xc5 Kh7xg7 Rc5xa5 Rd3xh3 Ra5-a7+ Kg7-f8 c4-c5 Rh3xb3 a4-a5 Rb3-c3 Ke5-d4 Rc3-g3 c5-c6 Rg3xg4+ Kd4-c5 Kf8-e8 a5-a6 b4-b3 Ra7-b7

30/86- 1:22:02 36,618,214,352 7,439,000 +2.29 f5-f6 h7-h6

30/86 1:46:43 47,818,290,145 7,467,000 +2.38 f5-f6 h7-h6 Nd5-e7+ Kg8-h7 Ne7-f5 d6-d5 f6xg7 Rf8-g8 e4xd5 Rb8-b6 Qf3-e4 Rb6-g6 h2-h3 Rg8-e8 Kg2-g3 Re8-g8 Nf5-h4 Rg8xg7 Rf1-f5 Qd7-e7 Rf5xe5 Qe7-d6 Nh4-f5 Qd6-f6 Nf5xg7 Kh7xg7 Re5-f5 Qf6-d6+ Qe4-f4 Qd6-e7 Kg3-f3 Kg7-f8 Qf4-b8+ Kf8-g7 Qb8-e5+ Qe7xe5 Rf5xe5 Rg6-f6+ Kf3-e4 Kg7-f8 h3-h4 Rf6-g6 Ke4-f4 Kf8-g7 Re5-f5 Rg6-a6 g4-g5 Kg7-f8 Rf5-f6 Ra6xf6+ g5xf6 Kf8-e8 h4-h5 Ke8-d7

31/86 2:14:17 60,406,505,491 7,496,000 +2.38 f5-f6 h7-h6 Nd5-e7+ Kg8-h7 Ne7-f5 d6-d5 f6xg7 Rf8-g8 e4xd5 Rb8-b6 Qf3-e4 Rb6-g6 h2-h3 Rg8-e8 Kg2-g3 Re8-g8 Nf5-h4 Rg8xg7 Rf1-f5 Qd7-e7 Rf5xe5 Qe7-d6 Nh4-f5 Qd6-f6 Nf5xg7 Kh7xg7 Re5-f5 Qf6-d6+ Qe4-f4 Qd6-e7 Kg3-f3 Kg7-f8 Qf4-b8+ Kf8-g7 Qb8-e5+ Qe7xe5 Rf5xe5 Rg6-f6+ Kf3-e4 Kg7-f8 h3-h4 Rf6-g6 Ke4-f4 Kf8-g7 Re5-f5 Rg6-a6 g4-g5 Kg7-f8 Rf5-f6 Ra6xf6+ g5xf6 Kf8-e8

32/86- 2:29:48 67,883,319,355 7,552,000 +2.27 f5-f6 h7-h6

32/86+ 3:51:52 105,528,962,460 7,585,000 +2.49 f5-f6

32/86+ 5:35:48 153,763,907,594 7,631,000 +2.72 f5-f6