Old position I posted a while back

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

Moderators: hgm, Rebel, chrisw

Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Old position I posted a while back

Post by Lyudmil Tsvetkov »

acase wrote:
Lyudmil Tsvetkov wrote:
hi Andrew.

this is some precomputed output.

how did you get to +500cps in 0.0.00 seconds?

this is an impossible score for this position, something is going wrong with your search.

yeah, take the time to mind your business, I am also very busy now with my game with Louis's SF, I will have to mate it at any cost. :)

later we will certainly discuss more, but this is not a large advantage position for white.
if anything, even if a miraculous win there, white will struggle and struggle for very very long.

PS. sorry for the quick position, just don't have another opportunity, as am doing much stuff at the same time.

did you run some matches with this position?


Hi Lyudmil,

I'm back from work now.

Of course there may have been some pre-computed output as I am running my computer in analysis mode and giving it the moves you suggested, so it's really no surprise that it might be in the Hash, so I can assure you there is nothing wrong with my computer's search

Here is the rest of the output from the search after almost 11-1/2 hours:


New game
2r1q2k/5p1p/5P2/2pRB3/1b6/p4QP1/5PKP/8 w - - 0 1

Analysis by Stockfish 190817 64 BMI2:

3
+- (5.52 ++) Depth: 53/95 00:30:34 69623MN
31.Qf5 Ra8 32.Rd7 Qg8 33.Qh5 Qg6 34.Qf3 Rf8 35.Qd5 a2 36.Qxa2 Kg8 37.Bd6 Rc8 38.Qd5 h5 39.Qb7 Re8 40.Re7 Rd8 41.Bc7 Rf8 42.h4 Bc3 43.Be5 Bd2 44.Rd7 Bh6 45.Bd6 Re8 46.Re7 Rd8 47.Qc7 Ra8 48.Qc6 Rd8 49.Bxc5 Qd3 50.Re5 Kh7 51.Rxh5 Kg6 52.Re5 Kh7 53.Qb7 Rd7 54.Qc8 Rd8 55.Qc6 Kg6 56.Qc7 Qd7 57.h5+ Kh7 58.Qxd7 Rxd7 59.Re7 Rd5 60.Rxf7+ Kg8 61.Rc7 Rf5 62.Bd4 Bf8 63.f7+ Rxf7 64.Rxf7
+- (5.52) Depth: 53/95 00:37:21 85308MN
31.Qf5 Qg8
+- (5.44 --) Depth: 54/67 00:38:39 88240MN
31.Qf5
+- (5.52 ++) Depth: 54/76 00:39:49 90914MN
31.Qf5
+- (5.63 ++) Depth: 54/76 00:42:54 98070MN
31.Qf5
+- (5.79 ++) Depth: 54/87 00:57:47 132905MN
31.Qf5 Ra8 32.Rd7 c4 33.Qf3 Rd8 34.Re7 Qg8 35.Qe4 Rf8 36.Qxc4 Bd2 37.Rc7 Ba5 38.Rd7 h6 39.Bf4 Qg6 40.Qc5 Ra8 41.Qd5 Rg8 42.Qxa5 Qe4+ 43.f3 Qxf4 44.Rxf7 Qc4 45.Re7 Qc2+ 46.Kh3 Qg6 47.f7 Rf8 48.Qe5+ Qg7 49.Qe6 a2 50.Qxa2 Kh7 51.Qc2+ Qg6 52.Qc8 Rxf7 53.Rxf7+ Qxf7 54.Qg4 Kh8 55.Qd4+ Kh7 56.Qf4 Qe6+ 57.Kh4 Qe7+ 58.Kg4 Kg7 59.Qd4+ Kg8 60.Kh3 Kf7 61.Qf4+ Kg7 62.Qg4+ Kf8 63.f4 Qd6 64.Qc8+ Ke7 65.Qf5 Kd8 66.Qf7 Kc8 67.Qe8+ Kb7 68.f5 Qa6 69.Qe7+ Kb8 70.Qf8+ Kc7
+- (5.89) Depth: 54/87 01:05:49 151437MN
31.Qf5
+- (5.96 ++) Depth: 55/86 01:09:09 159229MN
31.Qf5 c4
+- (5.81 --) Depth: 55/90 01:20:51 186547MN
31.Qf5 c4
+- (5.70 --) Depth: 55/90 01:21:30 188019MN
31.Qf5 c4
+- (5.54 --) Depth: 55/90 01:21:59 189126MN
31.Qf5 c4
+- (5.32 --) Depth: 55/90 01:23:37 192818MN
31.Qf5 c4
+- (5.03 --) Depth: 55/90 01:24:36 195042MN
31.Qf5
+- (5.27 ++) Depth: 55/90 01:25:24 196849MN
31.Qf5
+- (5.77 ++) Depth: 55/90 01:26:21 199107MN
31.Qf5 Qg8 32.Bf4 Ra8 33.Bh6 Qg6 34.Bg7+ Kg8 35.Qf3 a2 36.Rd1 a1Q 37.Rxa1 Rxa1 38.Qd5 Qxg7 39.fxg7 Ra6 40.h4 h6 41.h5 Kxg7 42.g4 Re6 43.f4 c4 44.Qxc4 Bf8 45.Qd4+ Kg8 46.g5 Rd6 47.Qc4 Kg7 48.Qb5 Kh7 49.Kg3 Kg8 50.Kg4 Bg7 51.Kf3 Re6 52.Kg3 Kf8 53.Qc5+ Kg8 54.Qc8+ Bf8 55.Qc3 Bd6 56.Kg4 hxg5 57.fxg5 Re4+ 58.Kf3 Re5 59.Qc8+ Kh7 60.Qd8 Rf5+ 61.Kg4 Rd5 62.Qe8 Rd4+ 63.Kf3 Rf4+ 64.Ke3 Rf1 65.Qb5 Re1+ 66.Kf3 Bf8 67.Qd5 Bg7 68.Kg4 Rg1+ 69.Kf3 Rh1
+- (6.12) Depth: 55/95 01:31:44 211875MN
31.Qf5 Qg8
+- (6.02 --) Depth: 56/93 01:33:03 215047MN
31.Qf5
+- (6.12 ++) Depth: 56/93 01:33:50 216948MN
31.Qf5 Ra8 32.Rd7 Qg8 33.Qh5 Qg6 34.Qf3 Rf8 35.Qd5 Qg8 36.Bd6 Rc8 37.Rxf7 Bc3 38.Qe6 Re8 39.Qd7 Rc8 40.Rxh7+ Qxh7 41.Qxc8+ Qg8 42.Qxc5 Qa8+ 43.f3 Bxf6 44.Qxa3 Qe8 45.Bc5 Kg7 46.Bf2 Qb5 47.Qa7+ Kg8 48.Qc5 Qd3 49.Qe3 Qg6 50.Qf4 Qf7 51.Qf5 Kh8 52.h4 Qe7 53.Qc5 Qe6 54.Qe3 Qd6 55.h5 Be5 56.Qe4 Qf6 57.Be3 Ba1 58.g4 Qb2+ 59.Kh3 Qe5 60.Qxe5+ Bxe5 61.Kg2 Bf6 62.f4 Kg8 63.h6 Kh8
+- (6.16) Depth: 56/93 01:36:04 222221MN
31.Qf5
+- (6.23 ++) Depth: 57/86 01:42:38 237983MN
31.Qf5
+- (6.31 ++) Depth: 57/86 01:43:54 240910MN
31.Qf5
+- (6.42 ++) Depth: 57/86 02:03:07 285383MN
31.Qf5 Ra8 32.Rd7 c4 33.Qf3 Rd8 34.Re7 Qg8 35.Qe4 Rf8 36.Qxc4 Bd2 37.Ra7 Bg5 38.h4 Bd2 39.Rxa3 Rc8 40.Qd4 Rd8 41.Qb6 Re8 42.Qc6 Rc8 43.Qf3 Qe8 44.Qe4 h6 45.Ra7 Qc6 46.Qxc6 Rxc6 47.Rxf7 Kg8 48.Rg7+ Kf8 49.Rh7 Bc3 50.Bxc3 Rxc3 51.Rxh6 Rc4 52.Rh7 Rc6 53.Ra7 Rxf6 54.g4 Rh6 55.Kg3 Rc6 56.g5 Rc1 57.Ra8+ Ke7 58.h5 Rg1+ 59.Kf4 Rg2 60.Ra7+ Kd6 61.h6 Rh2 62.Kg3 Rh1 63.Ra6+ Kc5 64.Ra7
+- (6.47) Depth: 57/89 02:06:43 293822MN
31.Qf5
+- (6.55 ++) Depth: 58/82 02:16:01 315577MN
31.Qf5
+- (6.62 ++) Depth: 58/95 02:26:10 339697MN
31.Qf5 Qg8 32.Bf4 Ra8 33.Bh6 Qg6 34.Bg7+ Kg8 35.Qf3 a2 36.Rd1 a1Q 37.Rxa1 Rxa1 38.Qd5 Qxg7 39.fxg7 Ra6 40.h4 h6 41.h5 Re6 42.g4 Bc3 43.Qxc5 Bxg7 44.Qc8+ Bf8 45.f4 Rd6 46.Qc4 Bg7 47.g5 Rd8 48.Kf3 Rf8 49.Qc5 Kh8 50.Qc6 f6 51.gxh6 Bxh6 52.Qc2 Bg7 53.Qg6 f5 54.h6 Bc3 55.Qc6 Ba1 56.Qc5 Rf7 57.Qa5 Bf6 58.Qxf5 Rf8 59.Qc5 Rg8 60.Qd6 Bd8 61.Qe6 Rf8 62.Qg4 Rg8 63.Qf5 Rg1 64.Qf8+ Rg8 65.Qf7 Rg1 66.h7 Bh4 67.Qe8+ Kxh7 68.Qh5+ Kg7
+- (6.66) Depth: 58/98 02:32:52 355473MN
31.Qf5
+- (6.73 ++) Depth: 59/92 02:42:47 379163MN
31.Qf5 c4 32.Rd4 a2 33.Rh4 Qg8 34.Rxh7+ Qxh7 35.Qxc8+ Qg8 36.Qxc4 Bd6 37.Bc3 Qg6 38.Qxa2 Qd3 39.Qa4 Kh7 40.Qh4+ Kg8 41.Qg4+ Qg6 42.Qc8+ Kh7 43.Qe8 Bc5 44.Qe5 Bf8 45.Qd5 Bh6 46.h4 Kh8 47.Qd4 Bc1 48.Qe5 Bh6 49.Qe8+ Kh7 50.Bd4 Bd2 51.Qc6 Bb4 52.f3 Kh8 53.Qe8+ Kh7 54.g4 Qc2+ 55.Kh3 Qc4 56.Qe4+ Kh8 57.g5 Qc8+ 58.Kg2 Qg8 59.Qd5 Bd2 60.Qe5 Qg6 61.Kh3 Qh5 62.Kg3 Qg6 63.Qe4 Kh7 64.Kg4 Kg8 65.Qa8+ Kh7 66.Qd5 Qg8 67.Qe4+ Qg6
+- (6.73) Depth: 59/92 02:43:05 379919MN
31.Qf5 c4 32.Rd4 a2 33.Rh4 Qg8 34.Rxh7+ Qxh7 35.Qxc8+ Qg8 36.Qxc4 Bd6 37.Bc3 Qg6 38.Qxa2 Qd3 39.Qa4 Kh7 40.Qh4+ Kg8 41.Qg4+ Qg6 42.Qc8+ Kh7 43.Qe8 Bc5 44.Qe5 Bf8 45.Qd5 Bh6 46.h4 Kh8 47.Qd4 Bc1 48.Qe5 Bh6 49.Qe8+ Kh7 50.Bd4 Bd2 51.Qc6 Bb4 52.f3 Kh8 53.Qe8+ Kh7 54.g4 Qc2+ 55.Kh3 Qc4 56.Qe4+ Kh8 57.g5 Qc8+ 58.Kg2 Qg8 59.Qd5 Bd2 60.Qe5 Qg6 61.Kh3 Qh5 62.Kg3 Qg6 63.Kg2 Qh7 64.Qe4 Qxe4 65.fxe4 Bb4 66.Kh3 Ba5 67.Kg4 Bd2
+- (6.73) Depth: 60/74 02:44:23 383032MN
31.Qf5
+- (6.81 ++) Depth: 61/79 03:03:15 428313MN
31.Qf5 Qg8 32.Bf4 Ra8 33.Bh6 Qg6 34.Bg7+ Kg8 35.Qf3 a2 36.Rd1 a1Q 37.Rxa1 Rxa1 38.Qd5 Qxg7 39.fxg7 Re1 40.h4 h6 41.g4 Bc3 42.Qxc5 Bxg7 43.f4 Bf8 44.Qc4 Re6 45.h5 Bd6 46.g5 Bxf4 47.Qxf4 hxg5 48.Qxg5+ Kh7 49.Qf5+ Kg7 50.Qg4+ Kh7 51.Kh3 Re3+ 52.Kh4 Re6 53.Qg5 Rh6 54.Kg4 Re6 55.Qf5+ Kg7 56.Qc5 Rh6 57.Qe5+ Kg8 58.Qg5+ Kh7 59.Qf5+ Kg8 60.Qd5 Re6 61.Kf3 Kh7 62.Kf2 Kg8 63.Qg5+ Kh7 64.Qf5+ Kg7 65.Kg2 Kg8 66.Kf3 Kg7 67.Qg5+ Kh7 68.Qd8 Kg7 69.Kg4 Rh6 70.Qd4+ Kg8 71.Qc4 Re6 72.Kf5 Rh6 73.Qh4 Kh7 74.Qe4 Re6
+- (6.85) Depth: 61/98 03:24:01 479663MN
31.Qf5 Qg8 32.Bf4 Ra8 33.Bh6 Qg6 34.Bg7+ Kg8 35.Qf3 a2 36.Rd1 a1Q 37.Rxa1 Rxa1 38.Qd5 Qxg7 39.fxg7 Re1 40.h4 h6 41.g4 Bc3 42.Qxc5 Bxg7 43.f4 Bf8 44.Qc4 Re6 45.h5 Bd6 46.g5 Bxf4 47.Qxf4 hxg5 48.Qxg5+ Kh7 49.Qf5+ Kg7 50.Qg4+ Kh7 51.Kh3 Re3+ 52.Kh4 Re6 53.Qg5 Rh6 54.Kg4 Re6 55.Qf5+ Kg7 56.Qc5 Rh6 57.Qe5+ Kg8 58.Qg5+ Kh7 59.Qf5+ Kg8 60.Kg5 Kg7 61.Qd7 Re6 62.Kf5 Kg8 63.Qd8+ Kg7 64.Qd4+ Kh7 65.Qd5 Kg7 66.Kg5 Rh6 67.Qf5 Re6 68.Kg4 Rh6 69.Qc5 Kg8 70.Kg5 Kg7 71.Qc8 Re6 72.Kf4 Rf6+ 73.Kg4 Rh6 74.Qc3+ Kh7 75.Qd3+ Kg8 76.Qd7 Kg7 77.Qd4+ Kh7 78.Qf4 Kg7
+- (6.85) Depth: 62/97 03:27:23 487856MN
31.Qf5 Qg8 32.Bf4 Ra8 33.Bh6 Qg6 34.Bg7+ Kg8 35.Qf3 a2 36.Rd1 a1Q 37.Rxa1 Rxa1 38.Qd5 Qxg7 39.fxg7 Re1 40.h4 h6 41.g4 Bc3 42.Qxc5 Bxg7 43.f4 Bf8 44.Qc4 Re6 45.h5 Bd6 46.g5 Bxf4 47.Qxf4 hxg5 48.Qxg5+ Kh7 49.Qf4 Kg7 50.Qc7 Re2+ 51.Kf3 Re6 52.Qg3+ Kh7 53.Qf4 Kg7 54.Qg5+ Kh7 55.Qf5+ Kg7 56.Kg4 Kg8 57.Qc5 Kg7 58.Qd4+ Kg8 59.Kg3 Kh7 60.Kf4 Kg8 61.Kg4 Rh6 62.Kg5 Kh7 63.Qd5 Kg7 64.Qd7 Re6 65.Kf5 Rf6+ 66.Kg4 Re6 67.Qd3 Kh6 68.Qd2+ Kh7 69.Kg5 Kg7 70.Qc3+ Kh7 71.Qc5 Kg7 72.Kg4 Rh6 73.Qe5+ Kg8 74.Qg5+ Kh7 75.Qf5+ Kg8 76.Qc8+ Kg7 77.Qc3+ Kg8 78.Qe5 Re6 79.Qd5 Rh6 80.Kh4 Kg7 81.Qd4+ Kg8
+- (6.85) Depth: 63/112 03:28:17 489998MN
31.Qf5 Qg8 32.Bf4 Ra8 33.Bh6 Qg6 34.Bg7+ Kg8 35.Qf3 a2 36.Rd1 a1Q 37.Rxa1 Rxa1 38.Qd5 Qxg7 39.fxg7 Re1 40.h4 h6 41.g4 Bc3 42.Qxc5 Bxg7 43.f4 Bf8 44.Qc4 Re6 45.h5 Bd6 46.g5 Bxf4 47.Qxf4 hxg5 48.Qxg5+ Kh7 49.Qf4 Kg7 50.Qc7 Re2+ 51.Kf3 Re6 52.Qg3+ Kh7 53.Qf4 Kg7 54.Qg5+ Kh7 55.Kf4 Rh6 56.Ke4 Re6+ 57.Kd5 Rh6 58.Qg3 Kh8 59.Qg4 Kh7 60.Qg5 Re6 61.Qf4 Kg7 62.Qd4+ Kg8 63.Qd2 Kg7 64.Qg2+ Kh7 65.Qc2+ Kh8 66.Qb2+ Kh7 67.Qb8 Rh6 68.Qb1+ Kg7 69.Qg1+ Kh7 70.Qg4 Re6 71.Qf3 Kg7 72.Qc3+ Kg8 73.Qc1 Kh7 74.Qc7 Kg7
+- (6.85) Depth: 64/105 03:43:52 528547MN
31.Qf5 Qg8 32.Bf4 Ra8 33.Bh6 Qg6 34.Bg7+ Kg8 35.Qf3 a2 36.Rd1 a1Q 37.Rxa1 Rxa1 38.Qd5 Qxg7 39.fxg7 Re1 40.h4 h6 41.g4 Bc3 42.Qxc5 Bxg7 43.f4 Bf8 44.Qc4 Re6 45.h5 Bd6 46.g5 Bxf4 47.Qxf4 hxg5 48.Qxg5+ Kh7 49.Qf4 Kg7 50.Qd4+ Kg8 51.Qd8+ Kg7 52.Qd7 Kg8 53.Qd4 Kh7 54.Kf3 Kg8 55.Qd8+ Kg7 56.Qc7 Rf6+ 57.Ke4 Rh6 58.Qe5+ Kh7 59.Qg5 Re6+ 60.Kd5 Rh6 61.Qg3 Kh8 62.Qg4 Kh7 63.Qg5 Re6 64.Qf4 Kg7 65.Qd4+ Kg8 66.Qd2 Kg7 67.Qg2+ Kh7 68.Qc2+ Kh8 69.Qb1 Rh6 70.Qf5 Kg8 71.Qc8+ Kg7 72.Qc3+ Kh7 73.Qd3+ Kg8 74.Qe2 Kf8 75.Qg4 Re6 76.Qb4+ Kg7 77.Qc3+ Kg8 78.Qc8+ Kg7
+- (6.85) Depth: 65/96 03:47:58 538227MN
31.Qf5 Qg8 32.Bf4 Ra8 33.Bh6 Qg6 34.Bg7+ Kg8 35.Qf3 a2 36.Rd1 a1Q 37.Rxa1 Rxa1 38.Qd5 Qxg7 39.fxg7 Re1 40.h4 h6 41.g4 Bc3 42.Qxc5 Bxg7 43.f4 Bf8 44.Qc4 Re6 45.h5 Bd6 46.g5 Bxf4 47.Qxf4 hxg5 48.Qxg5+ Kh7 49.Qf4 Kg7 50.Qd4+ Kg8 51.Qd8+ Kg7 52.Kf3 Rf6+ 53.Kg4 Rh6 54.Qd4+ Kg8 55.Qe5 Kh7 56.Qe8 Kg7 57.Qe7 Kg8 58.Qc5 Re6 59.Qc3 Rh6 60.Qd4 Re6 61.Kg5 Kh7 62.Qd3+ Kg7 63.Qc3+ Kh7 64.Qh3 Rh6 65.Qg4 Re6 66.Qf5+ Kg7 67.Qd5 Rh6 68.Qd4+ Kh7 69.Qe4+ Kg7 70.Qg4 Kh7 71.Kf4 Re6 72.Qf5+ Kg7 73.Kg5 Rh6 74.Qd5 Re6 75.Qc5 Rh6 76.Qc8 Re6 77.Qc7 Rh6 78.Qc3+ Kh7 79.Qd3+ Kg7 80.Qd7
+- (6.85) Depth: 66/100 03:48:34 539697MN
31.Qf5 Qg8 32.Bf4 Ra8 33.Bh6 Qg6 34.Bg7+ Kg8 35.Qf3 a2 36.Rd1 a1Q 37.Rxa1 Rxa1 38.Qd5 Qxg7 39.fxg7 Re1 40.h4 h6 41.g4 Bc3 42.Qxc5 Bxg7 43.f4 Bf8 44.Qc4 Re6 45.h5 Bd6 46.g5 Bxf4 47.Qxf4 hxg5 48.Qxg5+ Kh7 49.Qf4 Kg7 50.Qd4+ Kg8 51.Qd8+ Kg7 52.Kf3 Rf6+ 53.Kg4 Rh6 54.Qd4+ Kg8 55.Qe5 Kh7 56.Qe8 Kg7 57.Qe7 Kg8 58.Qc5 Re6 59.Qc3 Rh6 60.Qd4 Re6 61.Kg5 Kh7 62.Qd3+ Kg7 63.Qc3+ Kh7 64.Qh3 Rh6 65.Qg4 Re6 66.Qf5+ Kg7 67.Qd5 Rh6 68.Kf4 Re6 69.Qd4+ Kg8 70.Qd8+ Kg7 71.Qd7 Kg8 72.Qc8+ Kg7 73.Qc3+ Kh7 74.Qd3+ Kh6 75.Kg4 Kg7 76.Kg5 Re5+ 77.Kh4 Re6 78.Qd4+ Kh7 79.Kg5
+- (6.85) Depth: 67/98 03:49:16 541379MN
31.Qf5 Qg8 32.Bf4 Ra8 33.Bh6 Qg6 34.Bg7+ Kg8 35.Qf3 a2 36.Rd1 a1Q 37.Rxa1 Rxa1 38.Qd5 Qxg7 39.fxg7 Re1 40.h4 h6 41.g4 Bc3 42.Qxc5 Bxg7 43.f4 Bf8 44.Qc4 Re6 45.h5 Bd6 46.g5 Bxf4 47.Qxf4 hxg5 48.Qxg5+ Kh7 49.Qf4 Kg7 50.Qd4+ Kg8 51.Qd8+ Kg7 52.Kf3 Rf6+ 53.Kg4 Rh6 54.Qd4+ Kg8 55.Qe5 Kh7 56.Qe8 Kg7 57.Qe7 Kg8 58.Qc5 Re6 59.Qc3 Rh6 60.Qd4 Re6 61.Kg5 Kh7 62.Qd3+ Kg7 63.Qc3+ Kh7 64.Qh3 Rh6 65.Qg4 Re6 66.Qf5+ Kg7 67.Qd5 Rh6 68.Kf4 Re6 69.Qb7 Rf6+ 70.Kg5 Rh6 71.Qe4 Rd6 72.Qe8 Re6 73.Qa4 Re5+ 74.Kh4 Re6 75.Qd4+ Kh7 76.Kg5 Rh6 77.Qe4+ Kg7 78.Qg4 Kh7 79.Qc8 Kg7 80.Qc3+ Kh7
+- (6.85) Depth: 68/98 04:01:00 569767MN
31.Qf5
+- (6.92 ++) Depth: 69/100 04:59:26 715532MN
31.Qf5
+- (7.00 ++) Depth: 69/100 06:05:50 876127MN
31.Qf5
+- (7.11 ++) Depth: 69/100 08:01:27 1152327MN
31.Qf5
+- (7.27 ++) Depth: 69/105 09:39:44 1393741MN
31.Qf5
+- (7.49 ++) Depth: 69/116 11:26:52 1658111MN



As you can see the eval is up to 7.49 for white now which would signify an easy win for Stockfish I do concede that it might take several more moves before mate, but its just a matter of technique for Stockfish, and by all means please take your time replying especially if you are busy playing a game against Louis's SF. Good luck in that game. I didn't realize you were playing a game, so I'll definitely have to check that out.


I don't generally run matches, especially when analysis mode shows me what I need to know. I would wager that from that position after 30...Rc8 that Stockfish would win 100% of the time against any program or human or any combination of human + comp.

Have a nice evening.
we will have to go through the moves one by one then, as I simply don't believe this is winning.

it might be, please don't understand me wrong, but I have to be convinced, as I am seeing to many draw variations on my computer.

I don't quite have the time right now for a full analysis, and I see you are also busy, but when possible, we might continue a bit with it.

for example, in your posted analysis, after Qf5, I would like Qg8 more.

I don't know what to trust: your LTC analysis, or my 20/20 drawn stats in shootouts.

sometimes, direct games featuring live move sequences are more meaningful that having an engine analyse from a distant root position.

it will not be the first time that a LTC engine analysis proves wrong, skipping certain moves here and there.

when possible, I will post a more substantial continuation, but I don't know when this will be.

btw., what would other sources quote on this?(you certainly know more about its history)
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Old position I posted a while back

Post by Lyudmil Tsvetkov »

oops, I missed your comments SF would win 100% from this position against any human or engine.

come on bro :), (I respect you very much, Andrew), I showed 20/20 drawn stats with self-playing SF and Komodo, and you still insist on that...

run some matches starting with this position, and then please feel free to post back.

if anything, the exercise would not be 'white starts and wins', rather than 'white starts and possibly wins'.

we could not be certain, it takes some 100 moves or so until the conclusion, so this test position is anything else but a straightforward one.

evaluation is simply almost equal, how could I believe white wins easily?
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Old position I posted a while back

Post by Lyudmil Tsvetkov »

as said, there are too many variations available, for any human or computer to succeed in fully encompassing them, and for the position to be a straightforward test one.

for example, I think I have found an improvement for black earlier on.

after Rd5 ed5 Rd1, black plays Qg8, to try early challenge the white queen, and neither a4 or d4, then following up, after the forced Bd5 Rd5(maybe there are alternatives too, keeping both rooks) Rd5 Re8:

[d]4r1qk/5p1p/5P2/pppRB2Q/1b6/1P4P1/5P1P/6K1 w - - 0 5

we get here

what does engine analysis suggest?

I see straightforward draw, but this is only a quick check, really no time for more.( and I have to rely on SF scores, as everyone else)
acase
Posts: 981
Joined: Wed Mar 15, 2006 1:14 am
Location: Columbus, Ohio USA
Full name: Andrew R Case

Re: Old position I posted a while back

Post by acase »

Lyudmil Tsvetkov wrote:oops, I missed your comments SF would win 100% from this position against any human or engine.

come on bro :), (I respect you very much, Andrew), I showed 20/20 drawn stats with self-playing SF and Komodo, and you still insist on that...

run some matches starting with this position, and then please feel free to post back.

if anything, the exercise would not be 'white starts and wins', rather than 'white starts and possibly wins'.

we could not be certain, it takes some 100 moves or so until the conclusion, so this test position is anything else but a straightforward one.

evaluation is simply almost equal, how could I believe white wins easily?


Lyudmil my good friend,

I have the utmost respect for you also (even though I disagree with you most of the time, but not always) The eval for the position is now over +8 for white, if that isn't a win I don't know what is. I certainly agree with you that Stockfish (or any other program for that matter) doesn't always play the best move for every position and that its eval is wrong sometimes (rarely), but this isn't one of those positions, and I still maintain that from this position Stockfish would win 100% of the time without any draws (or loses).


Here is the rest of the analysis:


New game
2r1q2k/5p1p/5P2/2pRB3/1b6/p4QP1/5PKP/8 w - - 0 1

Analysis by Stockfish 190817 64 BMI2:


+- (7.78 ++) Depth: 69/116 14:29:55 2106834MN
31.Qf5 Ra8 32.Rd7 Qg8 33.Qh5 Qg6 34.Qf3 Rf8 35.Qd5 Qg8 36.Bd6 Rc8 37.Rxf7 Bc3 38.Qe6 h6 39.Bf4 Qe8 40.Qc4 Qg8 41.Bxh6 Bxf6 42.Qd5 a2 43.Qxa2 Re8 44.Qd5 Qg6 45.Be3 Bg7 46.Rf4 Qe6 47.Qxc5 Rd8 48.Bd4 Qd5+ 49.Qxd5 Rxd5 50.Bxg7+ Kxg7 51.Ra4 Kf6 52.g4 Rb5 53.h4 Kg7 54.Ra6 Rb7 55.Ra5 Rb4 56.Rg5+ Kh6 57.f3 Rb3 58.Rh5+ Kg6 59.Ra5 Rb2+ 60.Kg3 Kh7 61.Rg5 Rb1 62.Rh5+ Kg6 63.Ra5 Kh6 64.Ra6+ Kg7 65.Ra7+ Kg8 66.Ra5 Rg1+ 67.Kf2 Rc1 68.Kg2 Kh7 69.Rg5 Rc3 70.Rf5 Rc2+ 71.Kh3 Rc1 72.Rh5+ Kg6 73.Rg5+ Kf6 74.Rf5+ Kg6
+- (7.78) Depth: 69/128 18:23:38 2692856MN
31.Qf5
+- (7.85 ++) Depth: 70/102 20:53:05 3084531MN
31.Qf5
+- (7.93 ++) Depth: 70/103 21:13:17 3137275MN
31.Qf5
+- (8.04 ++) Depth: 70/103 21:21:56 3160529MN



I get the feeling that you might not be satisfied unless a mate is announced

:wink:
Last edited by acase on Tue Aug 22, 2017 11:12 am, edited 1 time in total.
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Old position I posted a while back

Post by Lyudmil Tsvetkov »

Qe8 (Rd5 ed5 Rd1 Qe8) for some reason, not entirely clear to me, probably because the queen attacks the bishop on e5 and is able to deliver check to the white king via e4 when on g2, seems even stronger, but too many tactics for a human to apprehend everything.

[d]r2rq2k/5p1p/5P2/ppppB2Q/1b6/1P4P1/5PBP/3R2K1 w - - 0 3

what do engines say on this?

there are some small details in the different positions, but they obviously do matter tactically.

I see mostly draws(again relying on buggy SF), and also some lines, where no one knows if it is a win or draw, like this one:

sorry, can not find my saved line, went somewhere, but basically it is an ending like KQB KQB, with white having 3 unadvanced pawns on the king side vs 1 black pawn there, and 1 very advanced black passer on c3 on the queen side, which no one knows what it is, SF scores change on every move from +80 to +150 and back, to finally settle for the lower score, even if one, you need some 100+ moves to win it, not to mention different ramifications before.
acase
Posts: 981
Joined: Wed Mar 15, 2006 1:14 am
Location: Columbus, Ohio USA
Full name: Andrew R Case

Re: Old position I posted a while back

Post by acase »

Lyudmil Tsvetkov wrote:as said, there are too many variations available, for any human or computer to succeed in fully encompassing them, and for the position to be a straightforward test one.

for example, I think I have found an improvement for black earlier on.

after Rd5 ed5 Rd1, black plays Qg8, to try early challenge the white queen, and neither a4 or d4, then following up, after the forced Bd5 Rd5(maybe there are alternatives too, keeping both rooks) Rd5 Re8:

[d]4r1qk/5p1p/5P2/pppRB2Q/1b6/1P4P1/5P1P/6K1 w - - 0 5

we get here

what does engine analysis suggest?

I see straightforward draw, but this is only a quick check, really no time for more.( and I have to rely on SF scores, as everyone else)

Lyudmil,


In this position Stockfish varies from 26.Rd5 that you played for white white and instead improves upon that by playing 26.Bxd5! which was in the original analysis posted by Eelco.
acase
Posts: 981
Joined: Wed Mar 15, 2006 1:14 am
Location: Columbus, Ohio USA
Full name: Andrew R Case

Re: Old position I posted a while back

Post by acase »

Lyudmil Tsvetkov wrote:Qe8 (Rd5 ed5 Rd1 Qe8) for some reason, not entirely clear to me, probably because the queen attacks the bishop on e5 and is able to deliver check to the white king via e4 when on g2, seems even stronger, but too many tactics for a human to apprehend everything.

[d]r2rq2k/5p1p/5P2/ppppB2Q/1b6/1P4P1/5PBP/3R2K1 w - - 0 3

what do engines say on this?

there are some small details in the different positions, but they obviously do matter tactically.

I see mostly draws(again relying on buggy SF), and also some lines, where no one knows if it is a win or draw, like this one:

sorry, can not find my saved line, went somewhere, but basically it is an ending like KQB KQB, with white having 3 unadvanced pawns on the king side vs 1 black pawn there, and 1 very advanced black passer on c3 on the queen side, which no one knows what it is, SF scores change on every move from +80 to +150 and back, to finally settle for the lower score, even if one, you need some 100+ moves to win it, not to mention different ramifications before.


Lyudmil,

Here is what Stockfish says so far in the position after 24.Rxd5 exd5 25Rd1 Qe8:


New game
[d]r2rq2k/5p1p/5P2/ppppB2Q/1b6/1P4P1/5PBP/3R2K1 w - - 0 1

Analysis by Stockfish 190817 64 BMI2:

26.Bxd5 a4 27.Bxa8 Rxa8 28.bxa4 bxa4 29.Qf5
+/= (0.69) Depth: 7/8 00:00:00 27kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.Bf4 Bc3 29.bxa4 bxa4
+/= (0.67) Depth: 8/8 00:00:00 35kN
26.Bxd5 a4 27.bxa4 bxa4 28.Bxa8 Rxa8 29.Qf5 Rd8 30.Rxd8
+/= (0.43) Depth: 9/12 00:00:00 80kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.bxa4 bxa4 29.Bf4 Bc3 30.Qg5 Qg8 31.Qxc5 Bxf6 32.Be5
+/= (0.49) Depth: 10/13 00:00:00 115kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.Bf4 Bc3 29.Qg5 Qf8 30.Bd6 Qg8 31.Qxc5 Bxf6 32.Qxb5
+/= (0.55) Depth: 11/13 00:00:00 142kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.bxa4 bxa4 29.Bf4 Bc3 30.Qg5 Qf8 31.Bd6 Qg8 32.Qxc5 Qg4 33.Rd3 Bxf6 34.Qd5
+/= (0.50) Depth: 12/19 00:00:00 187kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.bxa4 bxa4 29.Bf4 Bc3 30.Qg5 Qg8 31.Qxc5 Bxf6 32.Be5 Bg7 33.Bxg7+ Qxg7 34.Qd5 Qf8
+/= (0.44) Depth: 13/19 00:00:00 270kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.Bf4 Bc3 29.bxa4 bxa4 30.Qg5 Qf8 31.Bd6 Qg8 32.Qxc5 Qg4 33.Rd3 Bxf6 34.Be5 Bxe5
+/= (0.49) Depth: 14/19 00:00:00 323kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.Bf4 Bc3 29.bxa4 bxa4 30.Qg5 Qf8 31.Bd6 Qg8 32.Qxc5 Bxf6 33.Be5 Bg7 34.Bxg7+ Qxg7 35.Qc6 Qf8 36.Rd7 Kg8 37.Rd4
+/= (0.45) Depth: 15/23 00:00:00 387kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.Bf4 Bc3 29.bxa4 bxa4 30.Qg5 Qg8 31.Qxc5 Bxf6 32.Be5 Bg7 33.Bxg7+ Qxg7 34.Qc6 Qf8 35.Rd7 Kg8 36.Qd5 a3
+/- (0.71) Depth: 16/25 00:00:00 502kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.Bf4 Bc3 29.Qg5 Qf8 30.Bd6 Qg8 31.Qxc5 Bxf6 32.Be5 Bg7 33.Bxg7+ Qxg7 34.bxa4 bxa4 35.Rd7 Qf6 36.Qd5 Rf8 37.Kg2 a3 38.Ra7 Rd8 39.Qf3 Qxf3+ 40.Kxf3 Kg7 41.Rxa3
+/- (0.80) Depth: 17/38 00:00:00 1266kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.Bf4 Bc3 29.Qg5 Qf8 30.Bd6 Qg8 31.Qxc5 Bxf6 32.bxa4 bxa4 33.Be5 Bg7 34.Bxg7+ Qxg7 35.Rd7 a3 36.Qd5 Rf8 37.Kg2
+/= (0.45) Depth: 18/38 00:00:00 2516kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.Bf4 Bc3 29.Qg5 Qg8 30.Qxc5 Bxf6 31.Be5 Bg7 32.Bxg7+ Qxg7 33.bxa4 bxa4 34.Rd7 Qf6 35.Qd5 Rf8 36.Kg2 Kg8 37.Ra7 Rd8 38.Qe4 Qd4 39.Rxa4 Qxe4+ 40.Rxe4 Kg7 41.Kf3 Rd2 42.Re7 h5 43.Rc7
+/- (0.84) Depth: 19/38 00:00:00 2934kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.Bf4 Bc3 29.bxa4 bxa4 30.Qg5 Qg8 31.Qxc5 Bxf6 32.Be5 Bg7 33.Bxg7+ Qxg7 34.Rd7 Kg8 35.Qd5 Rc8 36.Ra7 Qa1+ 37.Kg2 Qf6 38.Rxa4 Qc6 39.Qxc6 Rxc6 40.g4 Kg7 41.h4
+/- (0.94) Depth: 20/44 00:00:00 5325kN
26.Bxd5 c4 27.bxc4 bxc4 28.Kg2 Rxd5 29.Rxd5 Bf8 30.Rd4 Qb5 31.Rh4 Qd5+ 32.Kg1 h6 33.Rd4 Qe6 34.Bd6 Kh7 35.Rg4 Qxf6 36.Qd5 Re8 37.Bxf8 Rxf8 38.Qxc4 Qe6 39.Qd4 f6
+/- (0.80) Depth: 21/44 00:00:00 7686kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.Bf4 Bc3 29.Qg5 Qf8 30.Bd6 Qg8 31.Qxc5 Bxf6 32.Be5 Bg7 33.Bxg7+ Qxg7 34.bxa4 bxa4 35.Rd7 Kg8 36.Qd5 Rb8 37.Ra7 Qg6 38.Kg2 Rc8 39.Rxa4 Qc6 40.Qxc6 Rxc6 41.g4 Rc3 42.Ra7 Kg7 43.h4 Kf6 44.g5+ Ke6 45.Rb7 Rc2 46.Kg3
+/- (0.90) Depth: 22/44 00:00:00 9038kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.Bf4 Bc3 29.Qg5 Qf8 30.Bd6 Qg8 31.Qxc5 Bxf6 32.bxa4 bxa4 33.Be5 Bg7 34.Bxg7+ Qxg7 35.Rd7 Kg8 36.Qd5 Rb8 37.Ra7 Qg6 38.Rxa4 Qb1+ 39.Kg2 Qb7 40.Qxb7 Rxb7 41.g4
+/- (0.98) Depth: 23/44 00:00:00 9213kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.Bf4 Bc3 29.Qg5 Qf8 30.Bd6 Qg8 31.Qxc5 Bxf6 32.bxa4 bxa4 33.Be5 Bg7 34.Bxg7+ Qxg7 35.Rd7 Kg8 36.Qd5 Rb8 37.Ra7 Qg6 38.Rxa4 Qb1+ 39.Kg2 Qb7 40.Qxb7 Rxb7 41.g4 Kg7 42.h4 Kf6 43.Kg3
+/- (0.98) Depth: 24/44 00:00:00 10198kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.Bf4 Bc3 29.Qg5 Qf8 30.Bd6 Qg8 31.Qxc5 Bxf6 32.bxa4 bxa4 33.Be5 Bg7 34.Bxg7+ Qxg7 35.Rd7 Kg8 36.Qd5 Rb8 37.Kg2 Qg6 38.Qf3 Rc8 39.Rxf7 Qxf7 40.Qg4+ Kh8 41.Qxc8+ Qg8 42.Qc6
+/- (0.92) Depth: 25/44 00:00:00 11090kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.Bf4 Bc3 29.bxa4 bxa4 30.Qg5 Qf8 31.Bd6 Qg8 32.Qxc5 Bxf6 33.Be5 Bxe5 34.Qxe5+ Qg7 35.Qe4 Qf8 36.Rd7 Kg7 37.Qd4+ Kg8 38.Qg4+ Kh8 39.Qf3 Kg7 40.Rd4 h6 41.Rg4+ Kh7 42.Qd3+ Kh8 43.Qd4+ Kh7 44.Qe4+ Kh8 45.Rh4 Kg7 46.Kg2 a3
+/- (0.82) Depth: 26/44 00:00:00 16783kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.bxa4 bxa4 29.Bf4 Bc3 30.Qg5 Qg8 31.Qxc5 Bxf6 32.Be5 Bxe5 33.Qxe5+ Qg7 34.Qe4 Qf8 35.Rd7 Kg8 36.Qg4+ Kh8 37.Qf3 Kg7 38.Rd4 h6 39.Rg4+ Kh7 40.Rf4 Kg7 41.Qc3+ Kh7 42.Qc2+ Kh8 43.Rxa4 Rxa4 44.Qxa4 Qc5 45.Qe4 Kg7 46.Kg2
+/- (0.77) Depth: 27/46 00:00:01 25449kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.bxa4 bxa4 29.Bf4 Bc3 30.Qg5 Qg8 31.Qxc5 Bxf6 32.Be5 Bxe5 33.Qxe5+ Qg7 34.Qe4 Qf8 35.Rd7 Kg8 36.Qg4+ Kh8 37.Qf3 Kg7 38.Rd4 h6 39.Rg4+ Kh7 40.Rf4 Kg7 41.Qc3+ Kg8 42.Rg4+ Kh7 43.Qc2+ Kh8 44.Rxa4 Rxa4 45.Qxa4 Qc5 46.Qf4 Kg7 47.Qe4 Qc1+ 48.Kg2 Qc5 49.g4 h5 50.gxh5 Qxh5
+/- (0.77) Depth: 28/51 00:00:01 29120kN
26.Bxd5 a4 27.Bxa8 Rxa8 28.Bf4 Bc3 29.Qg5 Qg8 30.Qxc5 Bxf6 31.bxa4 bxa4 32.Be5 Bg7 33.Bxg7+ Qxg7 34.Rd7 Qf8 35.Qd4+ Kg8 36.Qg4+ Kh8 37.Qf3 Kg7 38.Rd4 h6 39.Qc3 Kg8 40.Rg4+ Kh7 41.Qc2+ f5 42.Rxa4 Rxa4 43.Qxa4 Qe7 44.Qc6 Qe4 45.Qxe4 fxe4 46.g4 Kg6 47.h4 Kf6 48.Kg2
+/- (0.85) Depth: 29/51 00:00:01 39318kN
26.Bxd5 c4 27.bxc4 bxc4 28.Kg2 Rxd5 29.Rxd5 c3 30.Rd4 c2 31.Rc4 Qe6 32.Rxc2 Qd5+ 33.Qf3 Qxf3+ 34.Kxf3 a4 35.g4 a3 36.Ra2 Rc8 37.Ke3 Kg8 38.f4 Kf8 39.g5 Ke8 40.Kd3 Kd7 41.Re2 Bf8 42.h4 Ra8
+/- (0.92) Depth: 30/51 00:00:02 71344kN
26.Bxd5 a4
+/- (0.85 --) Depth: 31/49 00:00:03 102MN
26.Bxd5
+/- (0.92 ++) Depth: 31/59 00:00:03 113MN
26.Bxd5 a4 27.bxa4 bxa4 28.Bxa8 Rxa8 29.Bf4 Bc3 30.Qg5 Qf8 31.Bd6 Qg8 32.Qxc5 Bxf6 33.Be5 Bxe5 34.Qxe5+ Qg7 35.Qe4 Qf8 36.Rd7 Kg8 37.Qg4+ Kh8 38.Qh5 Kg8 39.Qg5+ Kh8 40.Qe5+ Kg8 41.Rd4 h6 42.Rg4+ Kh7 43.Qe4+ f5 44.Qb7+ Kh8 45.Rg6 Re8 46.Ra6 a3 47.Qf3
+/- (0.85) Depth: 31/59 00:00:03 115MN
26.Bxd5
+/- (0.93 ++) Depth: 32/62 00:00:04 134MN
26.Bxd5 c4
+/- (0.78 --) Depth: 32/62 00:00:04 149MN
26.Bxd5 c4
+/= (0.67 --) Depth: 32/62 00:00:06 192MN
26.Bxd5 c4 27.bxc4 bxc4 28.Kg2 Rxd5 29.Rxd5 c3 30.Rd4 c2 31.Rh4 Qg8 32.Rc4 Qg6 33.Qf3 Rd8 34.Qe2 Re8 35.f4 a4 36.Qxc2 Qxc2+ 37.Rxc2 Ra8 38.g4 h5 39.g5 a3 40.Ra2 Kh7 41.Kf3 Kg6 42.Ke4 Ra4 43.Kd5 Bf8 44.Kc6 Kf5 45.Kb5 Rb4+ 46.Kc6 Rc4+ 47.Kd5 Rc5+ 48.Kd4 Rc8
+/= (0.59) Depth: 32/62 00:00:07 235MN
26.Bxd5 c4
+/= (0.52 --) Depth: 33/49 00:00:10 322MN
26.Bxd5
+/= (0.59 ++) Depth: 33/55 00:00:11 361MN
26.Bxd5
+/= (0.70 ++) Depth: 33/55 00:00:11 377MN
26.Bxd5 c4 27.bxc4 bxc4 28.Kg2 Rxd5 29.Rxd5 c3 30.Rd4 c2 31.Rc4 Qe6 32.Rxc2 Qd5+ 33.f3 Bd6 34.Qg5 Bf8 35.Rd2 Qb5 36.Rd4 Rc8 37.Bf4 Qe2+ 38.Kh3 Qxf3 39.Qxa5 h6 40.Qd5 Qxd5 41.Rxd5 Kh7 42.Kg4 Kg6 43.h4 Kxf6 44.Kh5 Ke7 45.Bxh6 Bxh6 46.Kxh6 Kf6 47.Kh5 Rg8 48.Rd3 Ke6 49.Rf3 f6 50.Rb3 Ke5
+/- (0.81) Depth: 33/55 00:00:12 393MN
26.Bxd5 c4
+/- (0.73 --) Depth: 34/60 00:00:13 427MN
26.Bxd5 c4
+/= (0.66 --) Depth: 34/63 00:00:14 459MN
26.Bxd5
+/- (0.73 ++) Depth: 34/63 00:00:14 465MN
26.Bxd5 c4 27.bxc4 bxc4 28.Kg2 Rxd5 29.Rxd5 c3 30.Rd4 c2 31.Rc4 Qe6 32.Rxc2 Qd5+ 33.f3 Bd6 34.Qg5 Bf8 35.Rd2 Qb5 36.Rd4 Rc8 37.Bf4 Qe2+ 38.Kh3 Qxf3 39.Qxa5 h5 40.Qe5 Rc2 41.Rd2 Qg4+ 42.Kg2 Rxd2+ 43.Bxd2 h4 44.Qf4 Qe2+ 45.Kh3 hxg3 46.Kxg3 Qh5 47.h4 Qd5 48.Qg5 Bd6+ 49.Kf2 Bc5+ 50.Be3 Qd2+ 51.Kf3 Qd1+ 52.Kf4 Bxe3+ 53.Kxe3
+/- (0.87) Depth: 34/63 00:00:15 504MN
26.Bxd5 c4
+/- (0.79 --) Depth: 35/36 00:00:16 530MN
26.Bxd5
+/- (0.87 ++) Depth: 35/46 00:00:17 566MN
26.Bxd5
+/- (0.98 ++) Depth: 35/48 00:00:18 612MN
26.Bxd5
+/- (1.14 ++) Depth: 35/53 00:00:22 718MN
26.Rxd5
+/- (1.35 ++) Depth: 35/67 00:00:30 975MN
26.Rxd5
+- (1.65 ++) Depth: 35/67 00:00:32 1069MN
26.Rxd5
+- (2.03 ++) Depth: 35/67 00:00:39 1278MN
26.Rxd5 Qg8 27.Bd6 Qg6 28.Qe2 Bc3 29.Be4 Qg8 30.Rh5 Bxf6 31.Bxc5 Re8 32.Qd3 Rxe4 33.Qxe4 Ra6 34.Bd4 Bxd4 35.Qxd4+ Qg7 36.Qc5 Qa1+ 37.Kg2 Kg7 38.Qxb5 Rg6 39.Qxa5 Qxa5 40.Rxa5 Rb6 41.Ra3 Kf6 42.Kf3 Ke6 43.Ke4 f5+ 44.Kd3 Rd6+ 45.Ke3 Rb6 46.h4 h5 47.Kd4 Rd6+ 48.Kc5 Rd2 49.f4 Rc2+ 50.Kb4 Rg2 51.Ra6+ Kf7
+- (2.02) Depth: 35/67 00:00:44 1466MN
26.Rxd5
+- (2.10 ++) Depth: 36/59 00:00:49 1629MN
26.Rxd5
+- (2.17 ++) Depth: 36/59 00:00:51 1678MN
26.Rxd5 Qg8 27.Bd6 Qg6 28.Qe2 Bc3 29.Be4 Qg8 30.Qe3 Bxf6 31.Qf4 Bg7 32.Rh5 Rxd6 33.Qxd6 Qf8 34.Qd7 Ra6 35.Rxh7+ Kg8 36.Rh4 Rh6 37.Rg4 Rd6 38.Qxb5 Rd4 39.Qc6 Qe7 40.Kg2 Qd7 41.Qa8+ Qd8 42.Qa6 Qe8 43.Kf3 Kf8 44.Qxa5 Qc8 45.Rf4 Bh6 46.Rf6 Kg7 47.Qa6 Qxa6 48.Rxa6 Bd2 49.h4 c4 50.bxc4 Rxc4 51.Rd6 Bb4 52.Rd7 Kf6 53.Bd5
+- (2.21) Depth: 36/60 00:00:55 1806MN
26.Rxd5 Qg8
+- (2.14 --) Depth: 37/63 00:00:56 1851MN
26.Rxd5 Qg8 27.Bd6 Qg6 28.Qe2 Bc3 29.Be4 Qg8 30.Qe3 Bxf6 31.Qf4 Bg7 32.Rh5 Rxd6 33.Qxd6 Qf8 34.Qd7 Ra6 35.Rxh7+ Kg8 36.Rh4 Rh6 37.Rg4 Rd6 38.Qxb5 Rd4 39.Qc6 Qe7 40.Kg2 Qd7 41.Qa8+ Qd8 42.Qa6 Qe8 43.Kf3 Kf8 44.Qxa5 Qc8 45.Rf4 Rd6 46.b4 cxb4 47.Bd5 Rd7 48.Qxb4+ Kg8 49.Qb5 Qd8
+- (2.12) Depth: 37/63 00:00:57 1899MN
26.Rxd5 Qg8 27.Bd6 Qg6 28.Qe2 Bc3 29.Be4 Qg8 30.Qe3 Bxf6 31.Qf4 Bg7 32.Rh5 Rxd6 33.Qxd6 Qf8 34.Qd7 Ra6 35.Rxh7+ Kg8 36.Rh4 Rh6 37.Rg4 Rd6 38.Qxb5 Rd4 39.Qc6 Qe7 40.Qc8+ Rd8 41.Qf5 Kf8 42.Bd5 Qd7 43.Qg5 f6 44.Qg6 Re8 45.Re4 Rxe4 46.Qxe4 f5 47.Qf3 Ke7 48.Bc4 Bd4 49.h4 Qd6 50.Kg2 Qd7 51.h5 Kf6 52.Qf4 Ke7 53.Qg5+ Kd6 54.h6 Kc7 55.Qf4+ Kb6 56.Bf1
+- (2.18) Depth: 38/68 00:01:11 2340MN
26.Rxd5 Qg8
+- (2.11 --) Depth: 39/36 00:01:12 2393MN
26.Rxd5
+- (2.18 ++) Depth: 39/45 00:01:19 2610MN
26.Rxd5
+- (2.30 ++) Depth: 39/45 00:01:22 2719MN
26.Rxd5
+- (2.45 ++) Depth: 39/47 00:01:27 2891MN
26.Rxd5 Qg8 27.Bd6 Qg6 28.Qe2 Bc3 29.Be4 Qg8 30.Qe3 Bxf6 31.Qf4 Bg7 32.Rh5 Rxd6 33.Qxd6 Qf8 34.Qd7 Ra6 35.Rxh7+ Kg8 36.Rh4 Rh6 37.Rg4 Qd6 38.Qe8+ Qf8 39.Qxb5 Qd8 40.Bf3 Rf6 41.Re4 Re6 42.Rxe6 fxe6 43.Qxc5 Bd4 44.Qh5 Kf8 45.Kg2 Qb6 46.Qg5 Qc5 47.Qd2 Ke7 48.h4 Kf7 49.Qf4+ Ke7 50.Qg4 Bf6 51.h5 Qg5 52.Qc4 Bg7 53.Qc7+ Kf8 54.Qd6+ Kf7 55.Qd7+ Qe7
+- (2.47) Depth: 39/63 00:01:40 3324MN
26.Rxd5
+- (2.55 ++) Depth: 40/54 00:01:44 3492MN
26.Rxd5
+- (2.62 ++) Depth: 40/54 00:01:54 3801MN
26.Rxd5 Qg8
+- (2.40 --) Depth: 40/60 00:02:00 4008MN
26.Rxd5
+- (2.56 ++) Depth: 40/83 00:02:03 4118MN
26.Rxd5 Qg8 27.Bd6 Qg6 28.Qe2 Bc3 29.Be4 Qg8 30.Qe3 Bxf6 31.Qf4 Bg7 32.Rh5 Re8 33.Bxh7 Qxh7 34.Rxh7+ Kxh7 35.Qxf7 c4 36.bxc4 b4 37.c5 a4 38.c6 a3 39.Bxb4 a2 40.Bc3 a1Q+ 41.Bxa1 Rxa1+ 42.Kg2 Rf8 43.Qe7 Rc1 44.c7 Rf6 45.Qe4+ Kh8 46.Qe8+ Rf8 47.Qd7 Rc3 48.f4 Rc2+ 49.Kh3 Bf6 50.Kg4 Bg7 51.Qe7 Rc3 52.Qh4+ Kg8 53.Qd8 Bf6 54.Qd7 Bg7 55.Kh4 Rc2 56.g4 Rxh2+ 57.Kg3 Rc2 58.g5 Kh8 59.Kg4
+- (2.50) Depth: 40/83 00:02:10 4360MN
26.Rxd5
+- (2.57 ++) Depth: 41/40 00:02:15 4517MN
26.Rxd5
+- (2.65 ++) Depth: 41/55 00:02:23 4790MN
26.Rxd5 Qg8 27.Bd6 Qg6 28.Qe2 Bc3 29.Be4 Qg8 30.Qe3 Bxf6 31.Qf4 Bg7 32.Rh5 Rxd6 33.Qxd6 Qf8 34.Qd7 Ra6 35.Rxh7+ Kg8 36.Rh4 Rh6 37.Rg4 Qd6 38.Qe8+ Qf8 39.Qxb5 Qd8 40.Bf3 Rf6 41.Re4 Re6 42.Rxe6 fxe6 43.Qxc5 Bd4 44.Qh5 Kg7 45.Kg2 Bb6 46.Qe5+ Qf6 47.Qd6 e5 48.Qd7+ Qf7 49.Qc6 Bd4 50.h4 Qg6 51.Qd7+ Qf7 52.Qd8 Qf6 53.Qc8 Qf7 54.h5 Qf6 55.Qc7+ Kf8 56.Qxa5
+- (2.72) Depth: 41/66 00:02:36 5218MN
26.Rxd5 Qg8 27.Bd6 Qg6 28.Qe2 Bc3 29.Be4 Qg8 30.Qe3 Bxf6 31.Qf4 Bg7 32.Rh5 Rxd6 33.Qxd6 Qf8 34.Qd7 Ra6 35.Rxh7+ Kg8 36.Rh4 Rh6 37.Rg4 Qd6 38.Qe8+ Qf8 39.Qxb5 Qd8 40.Bf3 Rf6 41.Re4 Re6 42.Rxe6 fxe6 43.Qxc5 Bd4 44.Qh5 Kg7 45.Kg2 Bb6 46.Qe5+ Kf7 47.Bh5+ Ke7 48.Qg7+ Kd6 49.h4 Qe7 50.Qh8 Qd8 51.Qh6 Qc8 52.Bg6 Qa8+ 53.f3 Qd8 54.Qd2+ Ke7 55.Qf4 Qc7 56.Qg5+ Kd7 57.Qb5+ Kd6 58.Be4 Ke7 59.Qg5+ Kd7 60.Qg7+ Kd6 61.Qh6 Qc5
+- (2.79) Depth: 42/72 00:02:51 5763MN
26.Rxd5 Qg8
+- (2.71 --) Depth: 43/82 00:03:23 6828MN
26.Rxd5
+- (2.79 ++) Depth: 43/82 00:03:25 6901MN
26.Rxd5 Qg8 27.Bd6 Qg6 28.Qe2 Qb1+ 29.Rd1 Qg6 30.Be4 Qg8 31.Rd5 Bc3 32.Qe3 Bxf6 33.Qf4 Bg7 34.Rh5 Rxd6 35.Qxd6 Qf8 36.Qd7 Ra6 37.Rxh7+ Kg8 38.Rh4 Rh6 39.Rg4 Qd6 40.Qe8+ Qf8 41.Qxb5 Qd8 42.Bf3 Rg6 43.Re4 Re6 44.Rxe6 fxe6 45.Qxc5 Bd4 46.Qh5 Kf8 47.Qh6+ Ke7 48.Qh4+ Bf6 49.Qh7+ Kd6 50.h4 Qb6 51.Qd3+ Bd4 52.Kg2 Ke7 53.Qh7+ Kd6 54.Qc2 Qc5 55.Qd2 Ke7 56.h5 Bf6 57.Qf4 Kf7 58.Qb8 Qg5 59.Bc6 Bd8 60.Qc8 Qf6 61.Be4
+- (2.77) Depth: 43/86 00:03:43 7512MN
26.Rxd5 Qg8 27.Bd6 Qg6 28.Qe2 Qb1+ 29.Rd1 Qg6 30.Be4 Qg8 31.Rd5 Bc3 32.Qe3 Bxf6 33.Qf4 Bg7 34.Rh5 Rxd6 35.Qxd6 Qf8 36.Qd7 Ra6 37.Rxh7+ Kg8 38.Rh4 Rh6 39.Rg4 Qd6 40.Qe8+ Qf8 41.Qxb5 Qd8 42.Bf3 Rg6 43.Re4 Re6 44.Rxe6 fxe6 45.Qxc5 Bd4 46.Qh5 Kf8 47.Qh6+ Ke7 48.Qh4+ Bf6 49.Qh7+ Kd6 50.h4 Qb6 51.Qd3+ Bd4 52.Kg2 Ke7 53.Qh7+ Kd6 54.Qc2 Qc5 55.Qd2 Ke7 56.h5 Bf6 57.Qf4 Kf7 58.Qb8 Qg5 59.Bc6 Bd8 60.Qc8 Qf6 61.Be4 Be7 62.Qc7 Qd4 63.Qf4+ Qf6 64.Bg6+ Kg8 65.Qc7
+- (2.77) Depth: 44/79 00:04:00 8077MN
26.Rxd5
+- (2.84 ++) Depth: 45/81 00:05:22 10730MN
26.Rxd5 Qg8 27.Bd6 Qg6 28.Qe2 Bc3 29.Be4 Qg8 30.Qe3 Bxf6 31.Qf4 Bg7 32.Rh5 Rxd6 33.Qxd6 Qf8 34.Qd7 Ra6 35.Rxh7+ Kg8 36.Rh4 Rh6 37.Rg4 Qd6 38.Qe8+ Qf8 39.Qxb5 Qd8 40.Bf3 Rg6 41.Re4 Re6 42.Rxe6 fxe6 43.Qxc5 Bd4 44.Qh5 Kf8 45.Qh6+ Ke7 46.Qh4+ Bf6 47.Qh7+ Kd6 48.h4 Bd4 49.Kg2 Qb6 50.Qc2 Qc5 51.Qd2 Ke7 52.h5 Bf6 53.Qd3 Qf5 54.Qa6 Kf7 55.h6 Be7 56.Qe2 Qf6 57.h7
+- (2.78) Depth: 45/81 00:06:11 12284MN
26.Rxd5
+- (2.85 ++) Depth: 46/65 00:06:43 13296MN
26.Rxd5 Qg8
+- (2.70 --) Depth: 46/77 00:06:49 13494MN
26.Rxd5 Qg8 27.Bd6 Qg6 28.Qe2 Bc3 29.Be4 Qg8 30.Qe3 Bxf6 31.Qf4 Bg7 32.Rh5 Rxd6 33.Qxd6 Qf8 34.Qd7 Ra6 35.Rxh7+ Kg8 36.Rh4 Rh6 37.Rg4 Qd6 38.Qe8+ Qf8 39.Qxb5 Qd8 40.Bf3 Rg6 41.Re4 Re6 42.Rxe6 fxe6 43.Qxc5 Bd4 44.Qh5 Kg7 45.Kg2 Bb6 46.Qe5+ Kf7 47.Bh5+ Ke7 48.Qg7+ Kd6 49.h4 Qe7 50.Qh8 Kc7 51.Bg4 Qd6 52.Qg7+ Qd7 53.Qb2 Qd5+ 54.Bf3 Qd4 55.Qc2+ Qc5 56.Qd2 Qd4 57.Qe2 Kd6 58.h5 Qe5 59.Qc4 Qd4 60.Qc6+ Ke7 61.Qb7+ Kf8 62.Kh3 Bd8 63.Qc8 Kf7 64.Kg2 Bb6
+- (2.70) Depth: 46/85 00:07:01 13862MN
26.Rxd5
+- (2.77 ++) Depth: 47/82 00:07:33 14877MN
26.Rxd5
+- (2.85 ++) Depth: 47/82 00:08:09 16026MN
26.Rxd5
+- (2.96 ++) Depth: 47/82 00:11:48 23164MN
26.Rxd5 Qg8 27.Bd6 Qg6 28.Qe2 Bc3 29.Be4 Qg8 30.Qe3 Bxf6 31.Qf4 Bg7 32.Rh5 Rxd6 33.Qxd6 Qf8 34.Qd7 Ra6 35.Rxh7+ Kg8 36.Rh4 Rh6 37.Rg4 Qd6 38.Qe8+ Qf8 39.Qxb5 Qd8 40.Bf3 Rf6 41.Re4 Re6 42.Rxe6 fxe6 43.Qxc5 Bd4 44.Qh5 Kf8 45.Qh6+ Kf7 46.Kg2 Qb6 47.h4 Ke7 48.Qh7+ Kd6 49.Qc2 Qc5 50.Qd2 Ke7 51.h5 Bf6 52.Qd3 Qg5 53.Qh7+ Bg7 54.Qc2 Bf8 55.Qc8 Kf7 56.Bc6 Qf5 57.Qd7+ Kg8 58.Bf3 Bb4 59.Qc7 Qg5 60.Qc8+ Kf7 61.Bc6 Kg7 62.Qd7+ Qe7
+- (2.93) Depth: 47/84 00:12:21 24214MN




I'll leave it running again on this position while I go to work, be back in about 11-12 hours.
User avatar
Eelco de Groot
Posts: 4561
Joined: Sun Mar 12, 2006 2:40 am
Full name:   

Re: Old position I posted a while back

Post by Eelco de Groot »

Lyudmil Tsvetkov wrote:Qe8 (Rd5 ed5 Rd1 Qe8) for some reason, not entirely clear to me, probably because the queen attacks the bishop on e5 and is able to deliver check to the white king via e4 when on g2, seems even stronger, but too many tactics for a human to apprehend everything.

[d]r2rq2k/5p1p/5P2/ppppB2Q/1b6/1P4P1/5PBP/3R2K1 w - - 0 3

what do engines say on this?

there are some small details in the different positions, but they obviously do matter tactically.

I see mostly draws(again relying on buggy SF), and also some lines, where no one knows if it is a win or draw, like this one:

sorry, can not find my saved line, went somewhere, but basically it is an ending like KQB KQB, with white having 3 unadvanced pawns on the king side vs 1 black pawn there, and 1 very advanced black passer on c3 on the queen side, which no one knows what it is, SF scores change on every move from +80 to +150 and back, to finally settle for the lower score, even if one, you need some 100+ moves to win it, not to mention different ramifications before.
Hello Lyudmil, I accidently loaded my Ancalagon here and it was like watching grass grow. I have no idea why. It would have been okay if the analysis was good, but it was not.

This I got after almost 30 minutes:

Engine: Ancalagon HT (512 MB)
by T. Romstad, M. Costalba, J. Kiiski, G. Linscott

36 28:47 +1.08 1.Bxd5 c4 2.bxc4 bxc4 3.Kg2 Rxd5
4.Rxd5 c3 5.Rd4 c2 6.Rc4 Bf8 7.Rxc2 Qe6
8.Rc7 Kg8 9.Bd4 a4 10.Qg5+ Kh8
11.Bc5 Qe4+ 12.f3 Qe2+ 13.Kh3 h6
14.Qd5 (21.661.241.772) 12539

36 28:47 +0.84 1.Bc7 c4 2.Bxd8 Rxd8 3.Rxd5 Rxd5
4.Bxd5 cxb3 5.Bxb3 Qe1+ 6.Kg2 Qe4+
7.f3 Qg6 8.Qe5 a4 9.Qb8+ Qg8 10.Qxb5 Bc3
11.Qxa4 Bxf6 12.Qd7 Kg7 13.h4 Qf8
14.Qg4+ (21.661.241.772) 12539

36 28:47 +0.53 1.Rxd5 Qg8 2.Be4 Rxd5 3.Bxd5 Re8
4.f4 Rf8 5.Bd6 Bc3 6.Bxf8 Qxf8 7.Kf1 Bxf6
8.Be4 Qg8 9.Qxc5 a4 10.bxa4 bxa4
11.Qc6 Qd8 12.Qxa4 Kg7 13.Qc2 h6
14.Kg2 (21.661.241.772) 12539

The best move is actually Rxd5, at least that is what Stockfish is thinking. It destroys Black's set-up (aggressive mode)

Three moves again, in your position watch the difference :o
(I increased hashtable size for my Stock here by 100%)

[D]r2rq2k/5p1p/5P2/ppppB2Q/1b6/1P4P1/5PBP/3R2K1 w - -

Engine: Test Imbalance HT (1024 MB)
by T. Romstad, M. Costalba, J. Kiiski, G. Linscott

38 6:27 +2.11 1.Rxd5 Qg8 2.Bd6 Qg6 3.Qe2 Qb1+
4.Rd1 Qg6 5.Be4 Qg8 6.Bxa8 Rxa8
7.Qxb5 Bc3 8.Bxc5 Bxf6 9.Qc6 Be5
10.Bd4 Bxd4 11.Rxd4 Rc8 12.Qf6+ Qg7
13.Rd8+ Rxd8 14.Qxd8+ (4.045.984.317) 10443

38 6:27 +1.07 1.Bxd5 c4 2.bxc4 bxc4 3.Kg2 Rxd5
4.Rxd5 c3 5.Rd4 c2 6.Rc4 Qe6 7.Rxc2 Bf8
8.Rc7 Kg8 9.Bd4 a4 10.Qg5+ Kh8
11.Bc5 Ra5 12.Qd2 Ra8 13.Bxf8 Rxf8
14.Qb4 (4.045.984.317) 10443

38 6:27 +1.03 1.Bc7 c4 2.Bxd8 Rxd8 3.Bxd5 Bc3
4.bxc4 bxc4 5.Qh6 Qf8 6.Qh4 Rd6
7.Qxc4 Bxf6 8.Rb1 Rd8 9.Rb5 Qe8
10.Bxf7 Rd4 11.Bxe8 Rxc4 12.Rxa5 Bd4
13.Ra2 Kg7 14.Kg2 (4.045.984.317) 10443

and after 1. Rxd5

[D]r2rq2k/5p1p/5P2/pppRB2Q/1b6/1P4P1/5PBP/6K1 b - -

Engine: Test Imbalance HT (1024 MB)
by T. Romstad, M. Costalba, J. Kiiski, G. Linscott

42 8:20 +2.21 1...Qg8 2.Bd6 Qg6 3.Qe2 Qb1+ 4.Rd1 Qg6
5.Be4 Qg8 6.Bxa8 Rxa8 7.Qxb5 Bc3
8.Bxc5 Bxf6 9.Qc6 Bg7 10.Qf3 Rc8
11.Be3 Qe8 12.Rd5 Bc3 13.Qf5 Qe6
14.Qd3 Qg6 (5.553.328.549) 11086

42 8:20 +5.06 1...Rxd5 2.Bxd5 Ra7 3.Be4 Qg8 4.Bf4 Ra6
5.Bh6 Rxf6 6.Qe5 Qd8 7.Bd5 Bc3
8.Qxc3 Kg8 9.Bg5 Qxd5 10.Bxf6 Qf5
11.Bh8 Qb1+ 12.Kg2 Qe4+ 13.f3 Qe2+
14.Kg1 f6 (5.553.328.549) 11086

42 8:20 +5.24 1...Bd2 2.Be4 Qg8 3.Bd6 Rxd6 4.Rxd6 Re8
5.Bc6 Bf4 6.Rd7 Rf8 7.Rd5 Qg6 8.Qxg6 fxg6
9.gxf4 b4 10.Rxc5 Rxf6 11.Kg2 Rxf4
12.Rxa5 Kg7 13.Rb5 Kf6 14.Kg3 Rf5 (5.553.328.549) 11086

Now, maybe that is a bit optimistic. But I have no idea why Ancalagon only gave the same Rxd5 only 0.53
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
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Old position I posted a while back

Post by Lyudmil Tsvetkov »

acase wrote:
Lyudmil Tsvetkov wrote:oops, I missed your comments SF would win 100% from this position against any human or engine.

come on bro :), (I respect you very much, Andrew), I showed 20/20 drawn stats with self-playing SF and Komodo, and you still insist on that...

run some matches starting with this position, and then please feel free to post back.

if anything, the exercise would not be 'white starts and wins', rather than 'white starts and possibly wins'.

we could not be certain, it takes some 100 moves or so until the conclusion, so this test position is anything else but a straightforward one.

evaluation is simply almost equal, how could I believe white wins easily?


Lyudmil my good friend,

I have the utmost respect for you also (even though I disagree with you most of the time, but not always) The eval for the position is now over +8 for white, if that isn't a win I don't know what is. I certainly agree with you that Stockfish (or any other program for that matter) doesn't always play the best move for every position and that its eval is wrong sometimes (rarely), but this isn't one of those positions, and I still maintain that from this position Stockfish would win 100% of the time without any draws (or loses).


Here is the rest of the analysis:


New game
2r1q2k/5p1p/5P2/2pRB3/1b6/p4QP1/5PKP/8 w - - 0 1

Analysis by Stockfish 190817 64 BMI2:


+- (7.78 ++) Depth: 69/116 14:29:55 2106834MN
31.Qf5 Ra8 32.Rd7 Qg8 33.Qh5 Qg6 34.Qf3 Rf8 35.Qd5 Qg8 36.Bd6 Rc8 37.Rxf7 Bc3 38.Qe6 h6 39.Bf4 Qe8 40.Qc4 Qg8 41.Bxh6 Bxf6 42.Qd5 a2 43.Qxa2 Re8 44.Qd5 Qg6 45.Be3 Bg7 46.Rf4 Qe6 47.Qxc5 Rd8 48.Bd4 Qd5+ 49.Qxd5 Rxd5 50.Bxg7+ Kxg7 51.Ra4 Kf6 52.g4 Rb5 53.h4 Kg7 54.Ra6 Rb7 55.Ra5 Rb4 56.Rg5+ Kh6 57.f3 Rb3 58.Rh5+ Kg6 59.Ra5 Rb2+ 60.Kg3 Kh7 61.Rg5 Rb1 62.Rh5+ Kg6 63.Ra5 Kh6 64.Ra6+ Kg7 65.Ra7+ Kg8 66.Ra5 Rg1+ 67.Kf2 Rc1 68.Kg2 Kh7 69.Rg5 Rc3 70.Rf5 Rc2+ 71.Kh3 Rc1 72.Rh5+ Kg6 73.Rg5+ Kf6 74.Rf5+ Kg6
+- (7.78) Depth: 69/128 18:23:38 2692856MN
31.Qf5
+- (7.85 ++) Depth: 70/102 20:53:05 3084531MN
31.Qf5
+- (7.93 ++) Depth: 70/103 21:13:17 3137275MN
31.Qf5
+- (8.04 ++) Depth: 70/103 21:21:56 3160529MN



I get the feeling that you might not be satisfied unless a mate is announced

:wink:
tomorrow, as you return from work, maybe we will discuss further, I also have other to do now, but it is impossible for me to trust a single(ok, let's be 2) SF run at depth 70 or 100, rather than some 20 full games played with this position.

there are so many lines still uninvestigated, if there is a white win in this position, which I don't believe, as eval basically says otherwise, and most of the lines are easily drawn, then certainly you should baptise your test position 'white to play and mate in 193' :) :)

you know that every 6th or so SF LTC analysis is wrong and full of errors, right?

which does not mean, this time you could not be right.
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Old position I posted a while back

Post by Lyudmil Tsvetkov »

Eelco de Groot wrote:
Lyudmil Tsvetkov wrote:Qe8 (Rd5 ed5 Rd1 Qe8) for some reason, not entirely clear to me, probably because the queen attacks the bishop on e5 and is able to deliver check to the white king via e4 when on g2, seems even stronger, but too many tactics for a human to apprehend everything.

[d]r2rq2k/5p1p/5P2/ppppB2Q/1b6/1P4P1/5PBP/3R2K1 w - - 0 3

what do engines say on this?

there are some small details in the different positions, but they obviously do matter tactically.

I see mostly draws(again relying on buggy SF), and also some lines, where no one knows if it is a win or draw, like this one:

sorry, can not find my saved line, went somewhere, but basically it is an ending like KQB KQB, with white having 3 unadvanced pawns on the king side vs 1 black pawn there, and 1 very advanced black passer on c3 on the queen side, which no one knows what it is, SF scores change on every move from +80 to +150 and back, to finally settle for the lower score, even if one, you need some 100+ moves to win it, not to mention different ramifications before.
Hello Lyudmil, I accidently loaded my Ancalagon here and it was like watching grass grow. I have no idea why. It would have been okay if the analysis was good, but it was not.

This I got after almost 30 minutes:

Engine: Ancalagon HT (512 MB)
by T. Romstad, M. Costalba, J. Kiiski, G. Linscott

36 28:47 +1.08 1.Bxd5 c4 2.bxc4 bxc4 3.Kg2 Rxd5
4.Rxd5 c3 5.Rd4 c2 6.Rc4 Bf8 7.Rxc2 Qe6
8.Rc7 Kg8 9.Bd4 a4 10.Qg5+ Kh8
11.Bc5 Qe4+ 12.f3 Qe2+ 13.Kh3 h6
14.Qd5 (21.661.241.772) 12539

36 28:47 +0.84 1.Bc7 c4 2.Bxd8 Rxd8 3.Rxd5 Rxd5
4.Bxd5 cxb3 5.Bxb3 Qe1+ 6.Kg2 Qe4+
7.f3 Qg6 8.Qe5 a4 9.Qb8+ Qg8 10.Qxb5 Bc3
11.Qxa4 Bxf6 12.Qd7 Kg7 13.h4 Qf8
14.Qg4+ (21.661.241.772) 12539

36 28:47 +0.53 1.Rxd5 Qg8 2.Be4 Rxd5 3.Bxd5 Re8
4.f4 Rf8 5.Bd6 Bc3 6.Bxf8 Qxf8 7.Kf1 Bxf6
8.Be4 Qg8 9.Qxc5 a4 10.bxa4 bxa4
11.Qc6 Qd8 12.Qxa4 Kg7 13.Qc2 h6
14.Kg2 (21.661.241.772) 12539

The best move is actually Rxd5, at least that is what Stockfish is thinking. It destroys Black's set-up (aggressive mode)

Three moves again, in your position watch the difference :o
(I increased hashtable size for my Stock here by 100%)

[D]r2rq2k/5p1p/5P2/ppppB2Q/1b6/1P4P1/5PBP/3R2K1 w - -

Engine: Test Imbalance HT (1024 MB)
by T. Romstad, M. Costalba, J. Kiiski, G. Linscott

38 6:27 +2.11 1.Rxd5 Qg8 2.Bd6 Qg6 3.Qe2 Qb1+
4.Rd1 Qg6 5.Be4 Qg8 6.Bxa8 Rxa8
7.Qxb5 Bc3 8.Bxc5 Bxf6 9.Qc6 Be5
10.Bd4 Bxd4 11.Rxd4 Rc8 12.Qf6+ Qg7
13.Rd8+ Rxd8 14.Qxd8+ (4.045.984.317) 10443

38 6:27 +1.07 1.Bxd5 c4 2.bxc4 bxc4 3.Kg2 Rxd5
4.Rxd5 c3 5.Rd4 c2 6.Rc4 Qe6 7.Rxc2 Bf8
8.Rc7 Kg8 9.Bd4 a4 10.Qg5+ Kh8
11.Bc5 Ra5 12.Qd2 Ra8 13.Bxf8 Rxf8
14.Qb4 (4.045.984.317) 10443

38 6:27 +1.03 1.Bc7 c4 2.Bxd8 Rxd8 3.Bxd5 Bc3
4.bxc4 bxc4 5.Qh6 Qf8 6.Qh4 Rd6
7.Qxc4 Bxf6 8.Rb1 Rd8 9.Rb5 Qe8
10.Bxf7 Rd4 11.Bxe8 Rxc4 12.Rxa5 Bd4
13.Ra2 Kg7 14.Kg2 (4.045.984.317) 10443

and after 1. Rxd5

[D]r2rq2k/5p1p/5P2/pppRB2Q/1b6/1P4P1/5PBP/6K1 b - -

Engine: Test Imbalance HT (1024 MB)
by T. Romstad, M. Costalba, J. Kiiski, G. Linscott

42 8:20 +2.21 1...Qg8 2.Bd6 Qg6 3.Qe2 Qb1+ 4.Rd1 Qg6
5.Be4 Qg8 6.Bxa8 Rxa8 7.Qxb5 Bc3
8.Bxc5 Bxf6 9.Qc6 Bg7 10.Qf3 Rc8
11.Be3 Qe8 12.Rd5 Bc3 13.Qf5 Qe6
14.Qd3 Qg6 (5.553.328.549) 11086

42 8:20 +5.06 1...Rxd5 2.Bxd5 Ra7 3.Be4 Qg8 4.Bf4 Ra6
5.Bh6 Rxf6 6.Qe5 Qd8 7.Bd5 Bc3
8.Qxc3 Kg8 9.Bg5 Qxd5 10.Bxf6 Qf5
11.Bh8 Qb1+ 12.Kg2 Qe4+ 13.f3 Qe2+
14.Kg1 f6 (5.553.328.549) 11086

42 8:20 +5.24 1...Bd2 2.Be4 Qg8 3.Bd6 Rxd6 4.Rxd6 Re8
5.Bc6 Bf4 6.Rd7 Rf8 7.Rd5 Qg6 8.Qxg6 fxg6
9.gxf4 b4 10.Rxc5 Rxf6 11.Kg2 Rxf4
12.Rxa5 Kg7 13.Rb5 Kf6 14.Kg3 Rf5 (5.553.328.549) 11086

Now, maybe that is a bit optimistic. But I have no idea why Ancalagon only gave the same Rxd5 only 0.53
will have to look at it later.

your, Andrew's output and my conception all mystify me.

we should not forget there are fortresses which SF assesses as +1000 and even more, and won games, which SF assesses as 0.0 or even negative, so it is necessary for a human to look at all final positions, but even better check all moves along the way.

very difficult position, I don't know how we are going to sort it out, I would happily have tried, but too many ramifications at each and every ply(pointing at about equal character of the position, as otherwise obviously lost games don't have that many ramifications) to reasonably hope to reach a definite conclusion.