Problem position for engines!

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

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
User avatar
M ANSARI
Posts: 3454
Joined: Thu Mar 16, 2006 6:10 pm

Problem position for engines!

Post by M ANSARI » Mon Jan 20, 2014 5:19 pm

This was played very recently between Naiditch and Domingez and is apparently a draw which the human players seemed to understand very well. Not so easy for engines as they see it totally differently.


[d]8/8/1B3pk1/3P2p1/4P2p/3r3P/5PK1/8 b - - 0 73

and the final position where a draw was decided

[d]6r1/8/8/8/3B4/5k1p/5P1K/8 b - - 0 91

bob
Posts: 20916
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: Problem position for engines!

Post by bob » Mon Jan 20, 2014 6:28 pm

M ANSARI wrote:This was played very recently between Naiditch and Domingez and is apparently a draw which the human players seemed to understand very well. Not so easy for engines as they see it totally differently.


[d]8/8/1B3pk1/3P2p1/4P2p/3r3P/5PK1/8 b - - 0 73

and the final position where a draw was decided

[d]6r1/8/8/8/3B4/5k1p/5P1K/8 b - - 0 91
One note. Engines will have a much better chance on the 2nd position with the full move list as input. Starting from FEN, prior positions are unknown and repetitions are a long way off. But in general, these kinds of positions are problematic because there are so many different squares the pieces can occupy without changing a thing, except for the repetition list.

User avatar
M ANSARI
Posts: 3454
Joined: Thu Mar 16, 2006 6:10 pm

Re: Problem position for engines!

Post by M ANSARI » Tue Jan 21, 2014 8:29 am

Yes, I think this is where something like MC analysis on the fly becomes useful. I also think that one of the many VGA cards now out with hundreds of Cuda cores could be programmed to check very quickly such positions where there is no progress and quickly ascertain that such a position is drawn. I can see an engine going for such an endgame and missing out on a winning line because it sees such a huge advantage. I guess engines still have some ways to go before they are better than humans in all aspects of the game!

Uri Blass
Posts: 8771
Joined: Wed Mar 08, 2006 11:37 pm
Location: Tel-Aviv Israel

Re: Problem position for engines!

Post by Uri Blass » Tue Jan 21, 2014 8:40 am

I do not think that humans understand that the first position is a draw when I am sure that engines evaluates the second position correctly with the KRP vs KBP tablebases

User avatar
beachknight
Posts: 3533
Joined: Tue Jan 09, 2007 7:33 pm
Location: Antalya, Turkey
Contact:

Re: Problem position for engines!

Post by beachknight » Wed Jan 22, 2014 10:55 pm

16:
8/8/1B3pk1/3P2p1/4P2p/3r3P/5PK1/8 b - - 0 1

Analysis by Stockfish PA_GTB DD 64:

73...f5 74.exf5+ Kxf5 75.Be3 Rxd5 76.Kf3 Ra5 77.Kg2 Ra2 78.Kf3 Ra4 79.Kg2 Ra1 80.Bd4 Ra2 81.Kf3 Rc2 82.Bb6 Rd2 83.Bc5 Rd3+ 84.Be3 Rd7 85.Kg2 Rd5 86.Kf3 Ra5 87.Kg2 Ra2 88.Kf3 Ra4 89.Kg2 Ra1 90.Bd4 Ra2 91.Kf3 Rc2 92.Bb6 Rd2 93.Bc5 Rd3+ 94.Be3 Rd7 95.Kg2 Rd5
-+ (-3.41) Depth: 44/50 00:01:04 80809kN, tb=45759
(Chess Engines Testing, 23.01.2014)
hi, merhaba, hallo HT

User avatar
beachknight
Posts: 3533
Joined: Tue Jan 09, 2007 7:33 pm
Location: Antalya, Turkey
Contact:

Re: Problem position for engines!

Post by beachknight » Wed Jan 22, 2014 10:58 pm

16:
8/8/1B3pk1/3P2p1/4P2p/3r3P/5PK1/8 b - - 0 1

Analysis by Stockfish PA_GTB DD 64:

73...f5 74.exf5+ Kxf5 75.Be3 Rxd5 76.Kf3 Ra5 77.Kg2 Ra2 78.Kf3 Ra4 79.Kg2 Ra1 80.Bd4 Ra2 81.Kf3 Rc2 82.Bb6 Rd2 83.Bc5 Rd3+ 84.Be3 Rd7 85.Kg2 Rd6 86.Kf1 Rd3 87.Kg2 Rd5 88.Kf3 Ra5 89.Kg2 Ra2 90.Kf3 Ra4 91.Kg2 Ra1 92.Bd4 Ra2 93.Kf3 Rc2 94.Bb6 Rd2 95.Bc5 Rd3+ 96.Be3 Rd7 97.Kg2 Rd6 98.Kf1 Rd3 99.Kg2 Rd5
-+ (-3.41) Depth: 50/58 00:03:49 252MN, tb=176094
(Chess Engines Testing, 23.01.2014)
hi, merhaba, hallo HT

User avatar
beachknight
Posts: 3533
Joined: Tue Jan 09, 2007 7:33 pm
Location: Antalya, Turkey
Contact:

Re: Problem position for engines!

Post by beachknight » Wed Jan 22, 2014 11:00 pm

16:
8/8/1B3pk1/3P2p1/4P2p/3r3P/5PK1/8 b - - 0 1

Analysis by Stockfish PA_GTB DD 64:

73...f5 74.exf5+ Kxf5 75.Be3 Rxd5 76.Kf3 Ra5 77.Kg2 Ra2 78.Kf3 Ra4 79.Kg2 Ra1 80.Bd4 Ra2 81.Kf3 Rc2 82.Bb6 Rd2 83.Bc5 Rd3+ 84.Be3 Rd7 85.Kg2 Rd1 86.Bc5 Rd3 87.Be3 Rd5 88.Kf3 Ra5 89.Kg2 Ra2 90.Kf3 Ra4 91.Kg2 Ra1 92.Bd4 Ra2 93.Kf3 Rc2 94.Bb6 Rd2 95.Bc5 Rd3+ 96.Be3 Rd7 97.Kg2 Rd1 98.Bc5 Rd3 99.Be3 Rd5
-+ (-3.41) Depth: 53/60 00:05:27 372MN, tb=253209
(Chess Engines Testing, 23.01.2014)
hi, merhaba, hallo HT

User avatar
beachknight
Posts: 3533
Joined: Tue Jan 09, 2007 7:33 pm
Location: Antalya, Turkey
Contact:

Re: Problem position for engines!

Post by beachknight » Wed Jan 22, 2014 11:04 pm

16:
8/8/1B3pk1/3P2p1/4P2p/3r3P/5PK1/8 b - - 0 1

Analysis by Stockfish PA_GTB DD 64:

73...f5 74.exf5+ Kxf5 75.Be3 Rxd5 76.Kf3 Ra5 77.Kg2 Ra2 78.Kf3 Ra4 79.Kg2 Ra1 80.Bd4 Ra2 81.Kf3 Rc2 82.Bb6 Rd2 83.Bc5 Rb2
-+ (-3.41) Depth: 55/63 00:09:24 638MN, tb=528318
(Chess Engines Testing, 23.01.2014)
hi, merhaba, hallo HT

User avatar
beachknight
Posts: 3533
Joined: Tue Jan 09, 2007 7:33 pm
Location: Antalya, Turkey
Contact:

Re: Problem position for engines!

Post by beachknight » Wed Jan 22, 2014 11:05 pm

16:
8/8/1B3pk1/3P2p1/4P2p/3r3P/5PK1/8 b - - 0 1

Analysis by Stockfish 180114IP 64:

73...f5 74.f3 Kh5 75.Bc7 fxe4 76.fxe4 g4 77.hxg4+ Kxg4 78.Kh1 Rd4 79.e5 Rxd5 80.e6 Rd1+ 81.Kh2 Re1 82.Bd6 Rxe6 83.Bb8 Re2+ 84.Kg1 Kf5 85.Bd6 Rd2 86.Bh2 Ke4 87.Bb8 Rc2 88.Kh1 Rb2
-/+ (-1.46 --) Depth: 34/48 00:00:09 30869kN
(Chess Engines Testing, 23.01.2014)
hi, merhaba, hallo HT

User avatar
beachknight
Posts: 3533
Joined: Tue Jan 09, 2007 7:33 pm
Location: Antalya, Turkey
Contact:

Re: Problem position for engines!

Post by beachknight » Wed Jan 22, 2014 11:05 pm

16:
8/8/1B3pk1/3P2p1/4P2p/3r3P/5PK1/8 b - - 0 1

Analysis by Stockfish 180114IP 64:

73...f5 74.f3 fxe4 75.fxe4 Kf6 76.Bc7 Ke7 77.Bb6 Kd6 78.Bd8 Ke5 79.Kf2 Kf4 80.Bc7+ Kxe4 81.d6 Rd2+ 82.Kf1 Kf3 83.Kg1 Rd1+ 84.Kh2 Rd5 85.Kg1 Rd2 86.Kh1 Rd3 87.Kh2 Ke4 88.Kg2 Kf5 89.Kf2 Rxh3 90.d7
-/+ (-1.47 ++) Depth: 42/56 00:01:03 222MN
(Chess Engines Testing, 23.01.2014)
hi, merhaba, hallo HT

Post Reply