Which engine sees a draw in this position ?

[Matthias Gemuh]

Which engine sees a draw in this position ?

[D]7R/8/1p6/7k/8/8/1K5p/7r b - -

[michiguel]

Which engine sees a draw in this position ?

[D]7R/8/1p6/7k/8/8/1K5p/7r b - -

They have to have the right knowledge, which I doubt most will care to have. "a" and "b" pawns are draw, the rest black wins. A human player that studied endgames spots this immediately.

The other possibility is to have 6pc TBs, of course.

Miguel

[AdminX]

[d]7R/8/1p6/7k/8/8/1K5p/7r b - - 0 1

Analysis by Smaug 2.2.1 64-bit:

1. = (0.00): 1...Kg6
2. = (0.00): 1...Kg4
3. = (0.00): 1...Kg5

So I believe Glaurung and Stockfish will see it as well.

[royb]

[d]7R/8/1p6/7k/8/8/1K5p/7r b - - 0 1

Analysis by Smaug 2.2.1 64-bit:

1. = (0.00): 1...Kg6
2. = (0.00): 1...Kg4
3. = (0.00): 1...Kg5

So I believe Glaurung and Stockfish will see it as well.

I do NOT get this result with Smaug 2.2.1 (32-bit on this particular system):

25 -4.62 1.... Kg6 2.Rh4 b5 3.Rh8 Kf5 4.Rh4 Kg5 5.Rh8 Kg4 6.Rg8+ Kf3 7.Rf8+ Kg2 8.Rg8+ Kf2 9.Rf8+ Kg3 10.Rg8+ Kf4 11.Rf8+ Ke3 12.Rh8 Ke4 13.Re8+ Kd3 14.Rd8+ Kc4 15.Rh8 Kd5 16.Rd8+ Kc5 17.Rc8+ Kd6 18.Rh8 Ke6 (21.73)
26 -4.64 1.... Kg6 2.Rh4 b5 3.Rh8 Kf5 4.Rh4 Kg5 5.Rh8 Kg4 6.Rg8+ Kf3 7.Rf8+ Kg2 8.Rg8+ Kf2 9.Rf8+ Kg3 10.Rg8+ Kf4 11.Rf8+ Ke3 12.Rh8 Ke4 13.Re8+ Kd3 14.Rd8+ Kc4 15.Rh8 Kd5 16.Rd8+ Kc5 17.Rc8+ Kd6 18.Rd8+ Ke5 19.Re8+ Kd4 20.Kc2 (28.50)
27 -4.50 1.... Kg6 2.Rh4 b5 3.Rh8 Kf5 4.Rh4 Kg5 5.Rh8 Kg4 6.Rg8+ Kf3 7.Rf8+ Kg2 8.Rg8+ Kf2 9.Rf8+ Kg3 10.Rg8+ Kf4 11.Rf8+ Ke3 12.Rh8 Ke4 13.Re8+ Kd3 14.Rd8+ Kc4 15.Rh8 Kd5 16.Rd8+ Kc5 17.Rc8+ Kd6 18.Rd8+ Ke6 19.Re8+ Kd7 20.Rh8 Kc6 (228.53)
28 -4.62 1.... Kg6 2.Rh4 b5 3.Rh8 Kf5 4.Rh4 Kg5 5.Rh8 Kg4 6.Rg8+ Kf3 7.Rf8+ Kg2 8.Rg8+ Kf2 9.Rf8+ Kg3 10.Rg8+ Kf4 11.Rf8+ Ke3 12.Re8+ Kd4 13.Rd8+ Ke5 14.Re8+ Kf6 15.Rh8 b4 16.Rh5 Ke6 17.Rh8 Kf5 18.Rh4 Ke5 19.Rh5+ Kd4 20.Rh4+ Kc5 21.Kc2 (244.71)

Is this a bug of some sort?

Roy

[Eelco de Groot]

One possibility is that the Chessbase GUI takes over using tablebases, because it is already a 6 men position to start with. But that would mean that Ted has the 6 men tablebases installed for this position and he would have figured that out. I would be amazed if Glaurung or Smaug can see a draw here, without even 5 men tablebases used by the engine!


Regards, Eelco


7R/8/1p6/7k/8/8/1K5p/7r b - -

Engine: Blueberry Beta 4 DM70 build 393 (128 MB)
by F. Letouzey, T. Gaksch, E. de Groot


5/42 0:45 +3.18 1...Kg5 2.Rh7 b5 3.Rh8 b4 4.Rg8+ Kf4
5.Rf8+ Ke4 6.Re8+ Kd5 7.Rd8+ Kc4
8.Rc8+ Kd4 (26.707.128) 592

5/44 0:47 +3.19 1...Kg4 2.Rg8+ Kf5 3.Rh8 b5 4.Rh3 Kg5
5.Rh8 Kf4 6.Rh3 b4 7.Rh8 Kf5 8.Rf8+ Ke4
9.Re8+ Kd5 (27.473.802) 580

5/44 0:47 +3.19 1...Kg4 2.Rg8+ Kf4 3.Rf8+ Ke5 4.Re8+ Kf6
5.Rh8 Kg6 (27.475.120) 580

5/44 0:47 +3.19 1...Kg4 2.Rg8+ Kf4 3.Rf8+ Ke5 4.Re8+ Kf6
5.Rh8 Kg6 (27.499.562) 580

6/44 0:47 +3.19 1...Kg4 2.Rg8+ Kh4 3.Rh8+ Kg3 4.Rg8+ Kf4
5.Rf8+ Kg5 6.Rg8+ Kf5 7.Rh8 (27.577.990) 575

6/44 0:52 +3.22 1...Kg6 2.Rh3 b5 3.Rh8 b4 4.Ka2 Kf5
5.Kb2 b3 6.Rh3 Kg4 7.Rh8 Kg3 8.Rg8+ Kf4
9.Rf8+ Ke3 10.Re8+ Kd3 11.Rd8+ Kc4
12.Rc8+ Kb4 (29.421.856) 552

6/46 0:54 +3.22 1...Kg6 2.Rh3 b5 3.Rh8 b4 4.Ka2 Kf5
5.Kb2 b3 6.Rh3 Ke4 7.Rh7 Kf3 8.Rh8 Kg3
9.Rg8+ Kf2 (30.626.298) 563

6/46 0:57 +3.26 1...Kg6 2.Rh3 b5 3.Rh8 b4 4.Ka2 b3+
5.Kb2 Kg5 6.Rh3 Kf4 7.Rh8 Kg4 8.Rh7 Kg3
9.Rg7+ Kf4 10.Rf7+ Ke4 11.Re7+ Kd5
12.Rd7+ Ke6 (33.271.671) 579

7/46 0:57 +3.18 1...Kg6 2.Rh3 b5 3.Rh8 b4 4.Rh4 Kg5
5.Rh7 Kf5 6.Rf7+ Ke4 7.Re7+ Kd5
8.Rd7+ Kc4 9.Rc7+ Kd4 (33.284.542) 579

7/46 1:00 +3.19 1...Kg4 2.Rg8+ Kf4 3.Rh8 Ke5 4.Rh7 Kf5
5.Rh3 b5 6.Rh5+ Kg4 7.Rh8 b4 8.Rg8+ Kh4
9.Rh8+ Kg5 10.Rg8+ Kf6 11.Rh8 Kf5
12.Rf8+ Ke4 13.Re8+ Kd5 (35.313.846) 580

7/46 1:00 +3.19 1...Kg4 2.Rg8+ Kf4 3.Rf8+ Ke5 4.Re8+ Kf6
5.Rh8 b5 (35.370.981) 580

7/48 2:05 +3.26 1...Kg4 2.Rg8+ Kf4 3.Rf8+ Kg5 4.Rh8 b5
5.Rg8+ Kf6 6.Rh8 b4 7.Rh5 Kg6 8.Rh3 b3
9.Rh8 Kg5 10.Rh7 Kg4 11.Rg7+ Kf4
12.Rf7+ Ke4 13.Re7+ Kd5 14.Rd7+ Ke6 (97.567.845) 778

8/48 2:05 +3.22 1...Kg4 2.Rg8+ Kf4 3.Rf8+ Kg5 4.Rh8 b5
5.Rg8+ Kf6 6.Rh8 b4 7.Rh3 Kf5 8.Rf3+ Ke4 (97.629.149) 778

8/48 5:11 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 b3
5.Rh8 Kg5 6.Rh3 Kf4 7.Rh8 Kg3 8.Rg8+ Kh4
9.Rh8+ Kg4 10.Rh7 Kg3 11.Rg7+ Kf4
12.Rf7+ Ke5 13.Re7+ Kd6 (290.699.368) 934

8/48 5:28 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 b3
5.Rh8 Kg5 6.Rh3 Kf4 7.Rh8 Kf3 8.Rf8+ Ke3
9.Rh8 Kf2 10.Rf8+ Kg2 11.Rg8+ (308.028.126) 936

8/52 11:35 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 b3
5.Rh8 Kg5 6.Rh3 Kf4 7.Rh8 Kf3 8.Rh7 Kg3
9.Rg7+ Kf3 10.Rf7+ Kg4 (678.881.636) 976

9/52 11:36 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 Kg5
5.Rh7 Kg4 6.Rh8 b3 7.Rh7 Kg3 8.Rg7+ Kf4
9.Rf7+ Ke5 10.Re7+ Kd6 (679.428.600) 975

10/52 11:41 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 Kg5
5.Rh7 Kg4 6.Rh8 b3 7.Rh7 Kg3 8.Rg7+ Kf4
9.Rf7+ Ke5 10.Re7+ Kd6 (683.093.640) 974

11/52 11:47 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 b3
5.Rh8 Kg5 6.Rh3 Kf6 7.Rh5 Kf7 8.Rh7+ Kg6 (688.681.189) 972

12/52 11:59 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 b3
5.Rh8 Kg5 6.Rh3 Kf4 7.Rh8 Kf3 8.Rh7 Kg3
9.Rg7+ Kf4 10.Rf7+ Ke5 11.Re7+ Kd6 (698.919.379) 970

13/52 12:20 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 b3
5.Rh8 Kg5 6.Rh3 Kf4 7.Rh8 Kf3 8.Rh7 Kg3
9.Rg7+ Kf3 10.Rf7+ Ke4 11.Re7+ Kd5
12.Rd7+ Ke6 (715.350.975) 966

14/52 12:52 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 b3
5.Rh8 Kg5 6.Rh3 Kf4 7.Rh8 Kf3 8.Rh7 Kg3
9.Rg7+ Kh4 10.Rh7+ Kg5 11.Rg7+ Kf4
12.Rf7+ Ke5 13.Re7+ Kd6 (741.775.613) 960

15/58 12:57 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 b3
5.Rh8 Kg5 6.Rh3 Kf4 7.Rh8 Kf3 8.Rh7 Kg3
9.Rg7+ Kh4 10.Rh7+ Kg4 11.Rh8 Kg3
12.Rg8+ Kf4 13.Rf8+ Ke5 14.Re8+ Kd6 (746.762.070) 960

16/64 14:04 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 b3
5.Rh8 Kg5 6.Rh3 Kf4 7.Rh8 Kf3 8.Rh7 Kg3
9.Rg7+ Kf4 10.Rh7 Ke4 11.Re7+ Kf4
12.Rf7+ Kg4 13.Rg7+ Kf5 14.Rf7+ Ke6 (805.385.261) 953

17/64 15:43 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 b3
5.Rh8 Kg5 6.Rh3 Kf4 7.Rh8 Kf3 8.Rh7 Ke4
9.Re7+ Kf4 10.Rf7+ Kg5 11.Rh7 Kg4
12.Rg7+ Kf5 13.Rf7+ Ke5 14.Re7+ Kd5 (896.365.376) 950

18/78 18:25 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 b3
5.Rh8 Kg5 6.Rh3 Kf4 7.Rh8 Kf3 8.Rh7 Kg3
9.Rg7+ Kf4 10.Rh7 Ke4 11.Re7+ Kf5
12.Rh7 Kg4 13.Rg7+ Kf5 14.Rf7+ Ke5 (1.045.905.253) 946

19/78 23:23 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 b3
5.Rh8 Kg5 6.Rh3 Kf4 7.Rh8 Kf3 8.Rh7 Kg3
9.Rg7+ Kf4 10.Rh7 Ke4 11.Re7+ Kd5
12.Rd7+ Ke5 13.Rh7 Kf4 14.Rf7+ Kg4 (1.315.115.955) 937

20/78 32:36 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 b3
5.Rh8 Kg5 6.Rh3 Kf4 7.Rh8 Kf3 8.Rh7 Kg3
9.Rg7+ Kf4 10.Rh7 Kg4 11.Rh8 Kg3
12.Rg8+ Kh4 13.Rh8+ Kg4 14.Rh7 Kf5 (1.823.074.324) 931

21/78 48:02 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 b3
5.Rh8 Kg5 6.Rh3 Kf4 7.Rh8 Kf3 8.Rh7 Kg3
9.Rg7+ Kf4 10.Rh7 Ke5 11.Rh8 Ke4
12.Rh3 Kf5 13.Rh7 Kg4 14.Rg7+ Kf5 (2.679.321.606) 929

22/78 80:35 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 b3
5.Rh8 Kg5 6.Rh3 Kf4 7.Rh8 Kf3 8.Rh7 Kg3
9.Rg7+ Kf4 10.Rh7 Kf5 11.Rh4 Ke5
12.Rh8 Kf6 13.Rh5 Kf7 14.Rh7+ Kg6 (4.540.064.557) 938

23/78 145:50 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 b3
5.Rh8 Kg5 6.Rh3 Kf4 7.Rh8 Kf3 8.Rh7 Kg3
9.Rg7+ Kf4 10.Rh7 Kf5 11.Rh4 Ke5
12.Rh8 Ke4 13.Rh3 Kd4 14.Rh7 Ke5 (8.286.144.751) 946

24/90 333:04 +3.26 1...Kg6 2.Ka2 b5 3.Kb2 b4 4.Rh3 b3
5.Rh8 Kg5 6.Rh3 Kf4 7.Rh8 Kf3 8.Rh7 Kg3
9.Rg7+ Kf4 10.Rh7 Kf5 11.Rh4 Kf6
12.Rh5 Ke6 13.Rh3 Ke5 14.Rh7 Ke4 (20.290.407.846)1015

best move: Kh5-g6 time: 381:45.047 min n/s: 1.020.707 CPU 100.0% n/s(1CPU): 1.020.707 nodes: 23.358.490.000

not found...

[ernest]

Analysis by Smaug 2.2.1 64-bit:

1. = (0.00): 1...Kg6
2. = (0.00): 1...Kg4
3. = (0.00): 1...Kg5

So I believe Glaurung and Stockfish will see it as well.

I do NOT get this result with Smaug 2.2.1

I think this output is kind of a (bad/stupid?) joke: maybe that's what you get (with GUI detection) if you have the relevant 6-piece Nalimov...

Anyway, I pushed the analysis a bit longer, and got no further than you did:
Analysis by Smaug 2.2.1:
1...Kg4 2.Rg8+ Kh5 3.Rh8+ Kg5 4.Rg8+ Kh4 5.Rh8+ Kg3 6.Rg8+ Kf2 7.Rf8+ Kg2 8.Rg8+ Kf1 9.Rf8+ Ke2 10.Re8+ Kd3 11.Rd8+ Ke3 12.Re8+ Kd4 13.Rh8 Ke5 14.Rh6 b5 15.Rh8 Kf5 16.Rf8+ Ke4 17.Re8+ Kf3 18.Rh8 Kf4 19.Rf8+ Ke3 20.Re8+ Kd4 21.Rd8+ Kc4 22.Rh8 b4 23.Rh4+ Kc5 24.Kc2
-+ (-4.62) Depth: 28 00:02:42 545mN
1...Kg4 2.Rh6 b5 3.Rh7 b4 4.Rg7+ Kf4 5.Rf7+ Kg3 6.Rg7+ Kf2 7.Rf7+ Ke3 8.Re7+ Kd3 9.Rh7 Ke4 10.Rh4+ Kf3 11.Rh7 b3 12.Rf7+ Ke4 13.Re7+ Kf5 14.Rh7 Kf4 15.Rh6 Ke4 16.Rh8 Kf3 17.Rf8+ Kg2 18.Rg8+ Kf2 19.Rf8+ Ke2 20.Re8+ Kd3 21.Rd8+ Ke3 22.Rh8 Ke4 23.Re8+ Kf3 24.Kxb3
-+ (-3.94) Depth: 29 00:03:25 710mN
1...Kg4 2.Rh6 b5 3.Rh7 Kg3 4.Rg7+ Kh3 5.Rh7+ Kg2 6.Rg7+ Kf2 7.Rf7+ Ke3 8.Rh7 Kf3 9.Rh8 Kg3 10.Rg8+ Kh4 11.Rh8+ Kg5 12.Ka2 Kf5 13.Rh5+ Ke6 14.Rh6+ Kd5 15.Rh5+ Kc4 16.Kb2 b4 17.Rh8 Kb5 18.Rh3 Kc6 19.Rh4 Kc5 20.Kc2 b3+ 21.Kb2 Kd5 22.Kxb3
-+ (-3.74) Depth: 30 01:03:50 15976mN

[royb]

Analysis by Smaug 2.2.1 64-bit:

1. = (0.00): 1...Kg6
2. = (0.00): 1...Kg4
3. = (0.00): 1...Kg5

So I believe Glaurung and Stockfish will see it as well.

I do NOT get this result with Smaug 2.2.1

I think this output is kind of a (bad/stupid?) joke: maybe that's what you get (with GUI detection) if you have the relevant 6-piece Nalimov...

Anyway, I pushed the analysis a bit longer, and got no further than you did:
Analysis by Smaug 2.2.1:
1...Kg4 2.Rg8+ Kh5 3.Rh8+ Kg5 4.Rg8+ Kh4 5.Rh8+ Kg3 6.Rg8+ Kf2 7.Rf8+ Kg2 8.Rg8+ Kf1 9.Rf8+ Ke2 10.Re8+ Kd3 11.Rd8+ Ke3 12.Re8+ Kd4 13.Rh8 Ke5 14.Rh6 b5 15.Rh8 Kf5 16.Rf8+ Ke4 17.Re8+ Kf3 18.Rh8 Kf4 19.Rf8+ Ke3 20.Re8+ Kd4 21.Rd8+ Kc4 22.Rh8 b4 23.Rh4+ Kc5 24.Kc2
-+ (-4.62) Depth: 28 00:02:42 545mN
1...Kg4 2.Rh6 b5 3.Rh7 b4 4.Rg7+ Kf4 5.Rf7+ Kg3 6.Rg7+ Kf2 7.Rf7+ Ke3 8.Re7+ Kd3 9.Rh7 Ke4 10.Rh4+ Kf3 11.Rh7 b3 12.Rf7+ Ke4 13.Re7+ Kf5 14.Rh7 Kf4 15.Rh6 Ke4 16.Rh8 Kf3 17.Rf8+ Kg2 18.Rg8+ Kf2 19.Rf8+ Ke2 20.Re8+ Kd3 21.Rd8+ Ke3 22.Rh8 Ke4 23.Re8+ Kf3 24.Kxb3
-+ (-3.94) Depth: 29 00:03:25 710mN
1...Kg4 2.Rh6 b5 3.Rh7 Kg3 4.Rg7+ Kh3 5.Rh7+ Kg2 6.Rg7+ Kf2 7.Rf7+ Ke3 8.Rh7 Kf3 9.Rh8 Kg3 10.Rg8+ Kh4 11.Rh8+ Kg5 12.Ka2 Kf5 13.Rh5+ Ke6 14.Rh6+ Kd5 15.Rh5+ Kc4 16.Kb2 b4 17.Rh8 Kb5 18.Rh3 Kc6 19.Rh4 Kc5 20.Kc2 b3+ 21.Kb2 Kd5 22.Kxb3
-+ (-3.74) Depth: 30 01:03:50 15976mN

It turns out that Smaug (and other Glaurung derivatives) do NOT use TBs at all, so that's not a factor here. Your output is much closer to mine in that Smaug does not see a draw by any means. I'm still baffled as to how Ted got those values from Smaug.

[ernest]

It turns out that Smaug (and other Glaurung derivatives) do NOT use TBs at all, so that's not a factor here.

Yes, but as I said "with GUI detection", the GUI will give that output, whether or not the engine uses Nalimovs.
Ted was probably abused by that, and if he apologizes very deeply, maybe we'll forgive him.

[royb]

It turns out that Smaug (and other Glaurung derivatives) do NOT use TBs at all, so that's not a factor here.

Yes, but as I said "with GUI detection", the GUI will give that output, whether or not the engine uses Nalimovs.
Ted was probably abused by that, and if he apologizes very deeply, maybe we'll forgive him.

Ah, I see. I use SCID which does not allow the GUI to read TBs, but I believe that is not the case with Ted.

[AdminX]

One possibility is that the Chessbase GUI takes over using tablebases, because it is already a 6 men position to start with. But that would mean that Ted has the 6 men tablebases installed for this position and he would have figured that out. I would be amazed if Glaurung or Smaug can see a draw here, without even 5 men tablebases used by the engine!


Regards, Eelco

I don't believe the engines in my post use Tablebases. Maybe the GUI is the cause for the output. I will check. I was using the Chessbase 10 GUI.