View previous topic :: View next topic 
Author 

herO Member
Joined: 12 Nov 2014 Posts: 85

Posted: Tue Sep 05, 2017 12:01 pm Post subject: Hiarcs vs 400 years old chess puzzle 


Hiarcs and Shredder engines are able to solve many chess puzzles, even if tablebases are turned off.
Look at this position:
It is very interesting position. Many engines like Komodo 11.2.1 and Stockfish 8 cannot solve this position at all. They shows massive advantage for white.
video:
https://www.youtube.com/watch?v=tnHKUtUXaCM 

Back to top 


donkeylane Member
Joined: 29 Aug 2016 Posts: 77

Posted: Tue Sep 05, 2017 7:32 pm Post subject: 


A very interesting problem , although I do have Stockfish 6 on my mobile phone analyze this app ,and it did solve the problem and get the draw ,which was the solution , after 1 minute 10 seconds , the key being Ra1+ exchanging rooks and then Bh3!! ,which, as the narrator exclaimed was very reminiscent of a move by the GM Alexei Shirov, which I have seen often said to be the most brilliant chess move ever.Then after the pawn takes, and there are doubled pawns on the h file ,the black king can easily hold the draw by finding refuge on the black h 8 square, and white cannot dislodge it .Simple when you say it like that. Steve C. 

Back to top 


Peter Grayson Member
Joined: 04 Aug 2007 Posts: 203
Full Name: Peter Grayson
Location: South Wales, UK

Posted: Tue Sep 05, 2017 9:05 pm Post subject: Re: Hiarcs vs 400 years old chess puzzle 


herO wrote:  Hiarcs and Shredder engines are able to solve many chess puzzles, even if tablebases are turned off.
Look at this position:
r7/8/8/5bk1/8/5B2/5RPP/6K1 b   0 1
It is very interesting position. Many engines like Komodo 11.2.1 and Stockfish 8 cannot solve this position at all. They shows massive advantage for white.

Who runs engines without TB's for serious analysis? Komodo and Stockfish 8 find the solution instantaneously using Syzygy bases. They are designed to run with Syzygy bases to run at their strongest. Removing features provided by the engine authors will potentially weaken engnes, for example preventing HIARCS from using the supplied engine book will likely cause it to play weaker.
I cannot see merit in referring to an engine's failings when key elements are disabled to make it play weaker.
PeterG 

Back to top 


kgvetter Member
Joined: 12 May 2012 Posts: 90

Posted: Wed Sep 06, 2017 12:16 am Post subject: 


Komodo 11.1 64 bit solves immediately without tablebases...
Gerhard 

Back to top 


Cubeman Member
Joined: 31 Jul 2007 Posts: 442

Posted: Wed Sep 06, 2017 4:33 am Post subject: 


Komodo and Stockfish solve it and I don't have TB or Snynergy bases. Don't know why the first post says they can't? 

Back to top 


Cubeman Member
Joined: 31 Jul 2007 Posts: 442

Posted: Wed Sep 06, 2017 7:27 am Post subject: 


I just see that it mentions Stockfish 8, I was just using the Stockfish app on iphone which is a few years old so probably Stockfish 6. They must have changed something to make it miss the immediate draw possibility. 

Back to top 


herO Member
Joined: 12 Nov 2014 Posts: 85

Posted: Wed Sep 06, 2017 10:18 am Post subject: Re: Hiarcs vs 400 years old chess puzzle 


Peter Grayson wrote:  herO wrote:  Hiarcs and Shredder engines are able to solve many chess puzzles, even if tablebases are turned off.
Look at this position:
r7/8/8/5bk1/8/5B2/5RPP/6K1 b   0 1
It is very interesting position. Many engines like Komodo 11.2.1 and Stockfish 8 cannot solve this position at all. They shows massive advantage for white.

Who runs engines without TB's for serious analysis? Komodo and Stockfish 8 find the solution instantaneously using Syzygy bases. They are designed to run with Syzygy bases to run at their strongest. Removing features provided by the engine authors will potentially weaken engnes, for example preventing HIARCS from using the supplied engine book will likely cause it to play weaker.
I cannot see merit in referring to an engine's failings when key elements are disabled to make it play weaker.
PeterG 
Endgame tablebases will not help you in this position. There are 8 pieces on the board. 8man tableses does not exist. Do you have Lomonosov 7man tablebases on your computer? I guess not, because Lomonosov tablebases requires at least 100TB free storage.
There are tons of positions that engines are not able to solve it. Endgame tablebases does not help engine in these positions at all. 

Back to top 


herO Member
Joined: 12 Nov 2014 Posts: 85

Posted: Wed Sep 06, 2017 10:27 am Post subject: 


Cubeman wrote:  Komodo and Stockfish solve it and I don't have TB or Snynergy bases. Don't know why the first post says they can't? 
If I start Stockfish 8 on my computer, Stockfish will not find the right combination for white. It shows a HUGE advantage for white and this is wrong.
This position is a dead draw.
Stockfish 8 analysis
1...Ra1+ { +2.35/34 } 2.Rf1 Ra4 3.Kf2 Kf4 4.Re1 Rb4 5.h4 Bd7 6.Re7 Rb2+ 7.Re2 Rb4 8.Rd2 Be6 9.Be2 Ra4 10.Rb2 Ke5 11.Rb5+ Kf6 12.Kg3 Ra3+ 13.Bf3 Bc4 14.Rc5 Ba6 15.Rc2 Bb7 16.Rf2 Kg7 17.Kh2 Bc8 18.g4 Rb3 19.Kg2 Bb7 20.Bxb7 Rxb7 21.Rf5 Rb3 22.Rf3 Rb2+ 23.Kg3 Kg6 24.h5+ Kg5 25.Re3 Rb4 *
Komodo 11.2.1 analysis
1...Ra1+ { +2.45/33 } 2.Rf1 Ra4 3.Kf2 Kf4 4.Re1 Rb4 5.h4 Bd7 6.Re7 Rb2+ 7.Re2 Rb4 8.g3+ Kf5 9.Rc2 Kf6 10.Rc5 Be6 11.Ke3 Bf5 12.Be2 Re4+ 13.Kf2 Rb4 14.Rc6+ Kg7 15.Bf3 Be4 16.Bxe4 Rxe4 17.Kf3 Re7 18.Rd6 Rf7+ 19.Kg2 Rb7 20.Kh3 Rb3 21.h5 Re3 *
So it seems those engines are not able to solve this position. 

Back to top 


herO Member
Joined: 12 Nov 2014 Posts: 85

Posted: Wed Sep 06, 2017 10:47 am Post subject: 


Chess engine are not able to solve these positions. These positions are draw.
8/6pp/1k2p3/p1p1P1P1/P1R2r1P/3P4/1P2P1K1/8 b   0 1
4k3/5p2/1p1p1P1p/1P1P1P1P/3P4/8/4K3/8 w   0 1
6kr/5b1p/2p3pP/rpPp1pP1/pP1PpP2/P3P3/1K6/8 w   0 1
k7/8/6p1/1p1p1pP1/1P1P1P2/1P1K4/3B4/8 w   0 1
4k3/2p5/8/1p1pPp1p/1P1P1P1P/8/2P2P2/4K3 b   0 1
Here Stockfish and Komodo need to calculate at least to depth 50, then they finally "see" that this position is a draw.
8/p7/kpP5/qrp1b3/rpP2b2/pP4b1/P3K3/8 w   0 1
No engine is able to find the right solution here.
3B4/1r2p3/r2p1p2/bkp1P1p1/1p1P1PPp/p1P4P/PPB1K3/8 w   0 1
This position is extremely difficult for chess engines! Even the newest version of Komodo cannot see the winning moves for white! If you do the right moves by yourself, Komodo will show you that this position is always draw. It is not true. After few moves, Komodo FINALLY change evaluation from 0.00 to mate in 4 Even Stockfish or Houdini cannot solve this position. Look at the position:
4r3/p1p1pPp1/P1P1P1P1/5K2/3p2P1/7p/3P1ppr/3R1nkq w   0 1
You can get to this position with perfect moves.
8/p1P5/P5K1/3p4/8/3p1ppp/3P1ppr/3R1nkq w   0 1
In this position, human players know that white is winning, but Komodo, Stockfish and Dumbini still "think" that this position is a draw. They always recommend a move c8=Q. I am glad that Hiarcs 14 WCSC is able to find the right solution here c8=N after 1 minute of calculation (depth 53). But tactical engines still recommend nonsence move c8=Q even at depth 130. If you play 1.c8=N d4 2.Nb6 axb6 3.a7 b5 4.a8=N by yourself, then those tactical engines find a "miracle"! They change drastically their evaluation from 0.00 to mate in 4 

Back to top 


Harvey Williamson Site Admin
Joined: 29 Jul 2007 Posts: 5925
Full Name: Harvey Williamson
Location: Media City, UK

Posted: Wed Sep 06, 2017 11:34 am Post subject: Re: Hiarcs vs 400 years old chess puzzle 


herO wrote: 
Endgame tablebases will not help you in this position. 
Tablebases will help as they are used in the search and as soon as the engine finds a position with 6 or less pieces it will get the result from a tb look up. _________________ www.Hiarcs.com 

Back to top 


herO Member
Joined: 12 Nov 2014 Posts: 85

Posted: Wed Sep 06, 2017 12:07 pm Post subject: Re: Hiarcs vs 400 years old chess puzzle 


Harvey Williamson wrote:  herO wrote: 
Endgame tablebases will not help you in this position. 
Tablebases will help as they are used in the search and as soon as the engine finds a position with 6 or less pieces it will get the result from a tb look up. 
Thank you for your information Harvey. You are right. If there are many pieces on the board, most of the time engine do not get into a position with 6 or less pieces, so then tablebases will not help them.
For example
There are 22 pieces on the board. 

Back to top 


Martin Hertz Member
Joined: 07 Oct 2015 Posts: 41

Posted: Wed Sep 06, 2017 5:31 pm Post subject: 


MChess 5.0 from 1995 needs only 1 second to see the draw with a 0.00 score in ply 8:
Code:  06 Bh3 3.36
07 Rf1+ 0.45
08 Rf1+ 0.00 


Back to top 


donkeylane Member
Joined: 29 Aug 2016 Posts: 77

Posted: Wed Sep 06, 2017 10:39 pm Post subject: 


Yep you can't beat the oldies, 1995 was a good year, the 1995 R30 got it instantly too. Steve C. 

Back to top 


Peter Grayson Member
Joined: 04 Aug 2007 Posts: 203
Full Name: Peter Grayson
Location: South Wales, UK

Posted: Thu Sep 07, 2017 5:40 pm Post subject: 


herO wrote:  Chess engine are not able to solve these positions. These positions are draw.
8/6pp/1k2p3/p1p1P1P1/P1R2r1P/3P4/1P2P1K1/8 b   0 1
4k3/5p2/1p1p1P1p/1P1P1P1P/3P4/8/4K3/8 w   0 1
6kr/5b1p/2p3pP/rpPp1pP1/pP1PpP2/P3P3/1K6/8 w   0 1
k7/8/6p1/1p1p1pP1/1P1P1P2/1P1K4/3B4/8 w   0 1
4k3/2p5/8/1p1pPp1p/1P1P1P1P/8/2P2P2/4K3 b   0 1
Here Stockfish and Komodo need to calculate at least to depth 50, then they finally "see" that this position is a draw.
8/p7/kpP5/qrp1b3/rpP2b2/pP4b1/P3K3/8 w   0 1
No engine is able to find the right solution here.
3B4/1r2p3/r2p1p2/bkp1P1p1/1p1P1PPp/p1P4P/PPB1K3/8 w   0 1
This position is extremely difficult for chess engines! Even the newest version of Komodo cannot see the winning moves for white! If you do the right moves by yourself, Komodo will show you that this position is always draw. It is not true. After few moves, Komodo FINALLY change evaluation from 0.00 to mate in 4 :D Even Stockfish or Houdini cannot solve this position. Look at the position:
4r3/p1p1pPp1/P1P1P1P1/5K2/3p2P1/7p/3P1ppr/3R1nkq w   0 1
You can get to this position with perfect moves.
8/p1P5/P5K1/3p4/8/3p1ppp/3P1ppr/3R1nkq w   0 1
In this position, human players know that white is winning, but Komodo, Stockfish and Dumbini still "think" that this position is a draw. They always recommend a move c8=Q. I am glad that Hiarcs 14 WCSC is able to find the right solution here c8=N after 1 minute of calculation (depth 53). But tactical engines still recommend nonsence move c8=Q even at depth 130. If you play 1.c8=N d4 2.Nb6 axb6 3.a7 b5 4.a8=N by yourself, then those tactical engines find a "miracle"! :D They change drastically their evaluation from 0.00 to mate in 4 :D 
You are being very selective in your choice of positions. In the original position Stockfish fails to solve because it does not consider a Knight under promotion for move 4.a8=N.
Try the following position with Stockfish and Komodo and then try with HIARCS. The reason HIARCS fails is for the same reason as Stockfish and Komodo missed your position. It ignores an underpromotion.
3n4/8/n6p/2pB1K1k/1b6/6P1/2p5/4n3 w   0 1
Analysis by Stockfish 8:
1.Bc4 c1N 2.Bb5 Nc6 3.Bxc6 Nc7 4.Ba4 Nc2 5.Bxc2 Ne2 6.Bd1 Na6 7.Bxe2#
Analysis by Komodo 11.2.2 32bit:
1.Bc4 c1N 2.Bb5 Nc6 3.Bxc6 Nc7 4.Ba4 Nb3 5.Bxb3 Nd5 6.Bd1+ Nf3 7.Bxf3#
Analysis by Deep HIARCS 14 WCSC:
1.Bc4 Nd3 2.Bxd3 c1N 3.Be4 Nb8 4.Bf3#
The Earliest HIARCS engine that I have has the same problem
Analysis by Hiarcs8 Bareev:
1.Bc4 c1Q 2.Bxa6
+ (19.50) Depth: 2/6 00:00:00
1.Bc4 Nf3 2.Be2
+ (14.98) Depth: 2/10 00:00:00
1.Bc4 c1R
+ (14.05) Depth: 3/10 00:00:00
1.Bc4 Nd3 2.Bxd3 c1N 3.Bxa6
+ (8.18) Depth: 3/10 00:00:00
1.Bc4 Nd3 2.Bxd3 c1N 3.Bxa6
+ (8.18) Depth: 4/10 00:00:00
1.Bc4 Nd3 2.Bxd3 c1N 3.Bxa6
+ (8.18) Depth: 4/10 00:00:00
1.Bc4 Nd3 2.Bxd3 c1N 3.Bxa6
+ (8.18) Depth: 6/10 00:00:00 1kN
1.Bc4 Nd3 2.Bxd3 c1N 3.Bxa6 Nc6
+ (8.27) Depth: 6/11 00:00:00 2kN
1.Bc4 Nd3 2.Bxd3 c1N 3.Be4 Nd3 4.Bf3#
+ (#4) Depth: 7/13 00:00:00 3kN
1.Bc4 Nd3 2.Bxd3 c1N 3.Be4 Nd3 4.Bf3#
+ (#4) Depth: 7/13 00:00:00 3kN
1.Bc4 Nd3 2.Bxd3 c1N 3.Be4
+ (#4) Depth: 7/13 00:00:00 4kN
So for HIARCS this is a hereditary issue going back to at least HIARCS 8.
I would also comment that with PERFECT PLAY your last position
8/p1P5/P5K1/3p4/8/3p1ppp/3P1ppr/3R1nkq w   0 1
could never be achieved! The position may be legal that is probably what you meant however, I would like you to show the sequence of moves to get there, particularly the sequence with "perfect play".
In blockade positions, most engines tend to score based on material difference. Shredder is one engine that is capable of scoring some blockade positions correctly but perhaps the more valid question is "would engines play the positions correctly despite their scoring?" That is a far better measure and more telling than just the score of a fixed position.
PeterG 

Back to top 


Harvey Williamson Site Admin
Joined: 29 Jul 2007 Posts: 5925
Full Name: Harvey Williamson
Location: Media City, UK

Posted: Thu Sep 07, 2017 10:59 pm Post subject: Re: Hiarcs vs 400 years old chess puzzle 


herO wrote:  Harvey Williamson wrote:  herO wrote: 
Endgame tablebases will not help you in this position. 
Tablebases will help as they are used in the search and as soon as the engine finds a position with 6 or less pieces it will get the result from a tb look up. 
Thank you for your information Harvey. You are right. If there are many pieces on the board, most of the time engine do not get into a position with 6 or less pieces, so then tablebases will not help them.
For example
There are 22 pieces on the board. 
with only 5 man tbs enabled there are tb hits after about 14 seconds so some positions are being searched with 5 or less pieces on the board.
0.14 1.fxe8=Q (depth 66, 1/19), 47M nodes (3.3M nps), 29 tbhits
0.00: 7.65/3 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/4 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/5 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/6 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/7 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/8 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/9 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/10 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/11 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/12 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/13 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/14 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/15 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/16 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/17 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/18 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/19 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/20 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/21 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/22 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/23 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/24 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/25 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/26 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/27 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/28 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/29 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/30 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/31 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/32 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/33 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/34 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/35 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/36 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/37 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/38 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/39 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/40 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/41 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/42 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/43 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/44 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/45 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/46 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/47 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/48 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/49 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/50 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/51 1.fxe8=Q d3 2.Qxe7
0.00: 0.00/52 1.fxe8=Q d3 2.Qxe7
0.01: 0.00/53 1.fxe8=Q d3 2.Qxe7
0.01: 0.00/54 1.fxe8=Q d3 2.Qxe7
0.01: 0.00/55 1.fxe8=Q d3 2.Qxe7
0.01: 0.00/56 1.fxe8=Q d3 2.Qxe7
0.02: 0.00/57 1.fxe8=Q d3 2.Qxe7
0.02: 0.00/58 1.fxe8=Q d3 2.Qxe7
0.03: 0.00/59 1.fxe8=Q d3 2.Qxe7
0.04: 0.00/60 1.fxe8=Q d3 2.Qxe7
0.04: 0.00/61 1.fxe8=Q d3 2.Qxe7
0.06: 0.00/62 1.fxe8=Q d3 2.Qxe7
0.07: 0.00/63 1.fxe8=Q d3 2.Qxe7
0.09: 0.00/64 1.fxe8=Q d3 2.Qxe7
0.13: 0.00/65 1.fxe8=Q d3 2.Qxe7 _________________ www.Hiarcs.com 

Back to top 


