chesswrit

Just another WordPress.com site

Archive for the ‘Chess programs’ Category

The Non-Human Factor (Part 3 )

leave a comment »

Q.7.- I have always thought that chess-playing engines do cheat. One may accept it or not, but they not only cheat, I suppose that to be able to play Chess, they have to cheat: when you play against one of them (OTB games) , you cannot use written notes, books, magazines etc. But they do use the opening book they have. This prevents them from forgetting opening lines, changing the order of moves, etc. And the same is valid for the analysis they carry out later in the game. As a human, you may end up mixing lines, playing wrong moves in advance, overlooking some move when having to play. They will never make such mistakes. Engines do not overlook  moves once they have decided on a certain line of play. And they do not change moves by mistake. So, if you want to beat them ,you will have to be more cunning than them ,but do not put your hopes in human-like type of mistakes. Then the question is still there: Can they be beaten? How?. Well, I do not have a straightforward answer. I do not think there is such an answer.  I suppose you should survive in the opening by achieving a balanced position, play positionally active during the middlegame without entering no-way-out alleys, choose the right balance of forces, assess the matter of the Bishop pair or the B vs. N as accurately as possible and try to keep the middlegame as closed as possible BUT as full of possibilities as possible. Today’s programs can play positional Chess with a high degree of proficiency. A different matter is when the game reaches positions with less and less tactics and more and more strategy. If the engine plays to win, it may give you some fresh targets to shoot at. If not, at least you will have the draw at hand… Playing against them as if they were humans, will not pay off. Forget about passive opening set-ups even if they lead to closed or semi-closed positions. With the initiative, you will be smashed. Forget about open positions or tactical melées with interchange of blows by taking pieces on each turn while it also takes yours: if it goes into such lines, it is because it has seen all the possible intermediate moves. The normal outcome would be ending up with one or several pieces down…So, the best policy for you would be to keep the piece balance as tight as possible . On the other hand, one may think that keeping it simple will be useful. Forget about it too. The main problem is to understand how they are able to “see”” all the drawbacks every move contains. And this is not easy to explain.

Well, you may be wonder why writing about this topic since everybody knows they are (nearly) unbeatable. Perhaps because the whole thing is our last frontier today and perhaps because it can help us understand the way we think too. 

Be that as it may, it has not deterred me from keeping on playing Chess. So, do not think about engines, plies and other time consuming , mind-boggling  concepts and … play the Eternal Game. The rest may only be an odd mental state. 

(Note: This is a personal view susceptible of change as time passes by. Please remember that in Chess the best opinion is your opinion, the best ideas are your ideas and that everything is relative. )

pridaux 1

Prideaux.  Mate in 3 moves

reinmann

Reinmann. Mate in 3 moves.

QChess.

Written by QChess

January 5, 2016 at 8:22 am

The Non-Human Factor . (Part 2)

leave a comment »

(Note: You may notice sometimes I employ some terms which would be  more appropriate for human beings. It is for the sake of clarity. I would like to make it clear that a machine is a machine  and a human being is a human being. Bear in mind though that  any Chess program has a human team working in the shadow.)

Q.3.- This is evident: they see every tactical trick, they see the strongest points in their openings and the weakest of moves in yours. They are able to exploit the slightest of inaccuracies and make you commit more mistakes. In zeitnot positions, you are ,simply, lost.

They always see any attacking possibility, as well as the way of carrying the attack out without hesitation by using direct attacks, weird regrouping of pieces, intermediate moves, etc.

They also perceive all the drawbacks in any move immediately pinpointing it and taking advantage of it. They never tire, so they are always looking for the strongest attacking move in every position. When calculating, they never change the move order by mistake, forget moves and they continually put you under a stressing pressure .

Q.4/Q.5: When you play against one of this programs, you feel a terrible pressure.(They play on your nervous resistance too). This is because they always manage to turn the position into a tactical battle. Even in closed positions they are always  building up pressure by the continual creation of threats. The more they put pressure on your position, the quicker it will be torn apart (with your pieces lacking coordination until the position is impossible to be hold together and breaks down). Their amazing opening book and their skills at assessing positions (the more tactical the better but I think they are no longer afraid of strategical positions either) turn them into formidable opponents. When they clinch something, it is forever, no matter if it is an advantage or material. This is why gambits for the sake of them are disappearing from CC (remember: you cannot speculate against a computer. Your way of seeing things is not “their” way of the same.) Another feature is that they abandon their own opening book when they find something better while analysing. A human chessplayer only leaves the book when his opponent does it first, when s/he has a TN of his/her own , when s/he forgets the moves or when the book line ends.

Threats, threats, threats: they are always creating threats and forcing you to take defensive measures, having to change your best-placed pieces and keep the clumsiest ones.

Another amazing skill is the apparent easiness with which they provoke play with heterogeneous forces. They are very cunning in this field. While you may be contemplating lines with the same balance of forces they are always exploring changes ( Queen vs. 3 pieces or 2 pieces and 2 Pawns, or vs. 2 Rooks, etc.) They seem to possess an extreme ability at assessing these positions and go for them. So the human opponent , only thinking in the same correlation of forces may be abruptly woken up by a sudden change in that correlation of forces.  (Boris Spassky has always said Bobby Fischer was the first in showing a computer-like style of playing, and Bobby was very good at this method too (see my post “A Contribution by Bobby Fischer”, published on Feb 1st, 2014 in this same blog). Another feature opposite to humans in many cases is that they do not particularly feel any leaning towards keeping the Bishops/ exchanging BxN, but when they decide to keep the Bishop pair it can be lethal . They have reached a deep insight into what to change / what to keep in the matter of B’s and N’s.

Q.6.- Of course an absolutely strong GM could beat a program any time or at least one time. The problem is how many times in a row and after how many previous (human) defeats in a row too, etc. I think nobody would try to do it in short-time scheduled games like blitz. Perhaps a good test would be a 40-move-in-two-and-a-half-hour game  + a classical adjournment with home analysis using a computer to analyse (so as to level the chances. After all, the program has a program to analyse too! But the matter of openings should have to be somewhat rebalanced. Or perhaps not, to leave the human being to their own devices  ). Moreover, considering that players use programs in CC and victories keep taking place, I tend to think that a hole , leading to some sort of mistakes , does exist…Where is it if any?

(So as not to tire the reader, I will leave the matter of the possible weaknesses of programs and the answer to the 7th question for the next post.)

 williams

Mate in 3 moves. P.H. Williams 1904

wieck

F.G. Wieck 1859. Mate in 3.

Happy New Year 2016 to all my readers!.

QChess.

Written by QChess

January 1, 2016 at 9:05 am

The Non-Human Factor. (Part 1)

leave a comment »

“Nor even top professionals play serious games against today’s Chess programs: they are too strong”.  I have read this  statement a lot of times. Well, once established the main idea, I was on a train some days ago, a bit lost in my own thoughts and yes, daydreaming, when I began to think about it. Because one thing is to say something, accept it without hesitation because it was said by someone “important” and a different think is to put the idea before us an try to explain it clearly. So, in my opinion, the main questions could be: 

1.Why are today´s programs so strong /much stronger than top chessplayers?.

2.-In which ways “they” play different to us? Conversely: in which way do we play different to “them”?

3.-What do “they” see we do not see?

4.-In what way do “they” see/assess the elements of a game that we are unable to match them?

5.-In what way do “they” apply the elements of Chess that we fail to match their prowess?

6.- Can we really make any attempt at beating them?

7.- And the Big One:  Do “they” cheat??

I have devoted a lot of time to think about all this. I suppose the main fields to investigate are the openings and the middlegames (and apply the above-mentioned questions to them. Why?.- Because if you survive to reach an endgame ,the game will be lost/won/or drawn, and everything will depend on accurate calculation of variations (if it is drawn or won. If it is lost for you, there will be no problem: you will be scientifically killed and dissected by the monster…)

Possible answers that may change as time passes by…:

Q.1/ Q.2 – They play much stronger due to a variety of factors: they never forget the openings and play them without mistakes. They are extremely strong at tactical calculation. They never speculate or feel fear. They have no interest in their opponents’ ideas, emotions, body language, noise, etc. They are not pressed by rating points, timetables, previous results, statistics, final standings, money they can earn or lose, they are never tired or bored, they are never hungry or sleepy, they never have “a bad day” and have no prejudices, there are no ugly moves/positions for them. They are only there TO PLAY CHESS AND ONLY CHESS. During the past few years,apart from refining the tactical side of the program as far as calculation of movements is concerned, programmers have learnt to include strategical ideas in their way of choosing a move, making them even more deadly: for instance the always complex matters of B vs. N , the change B for N, the different value of B’s and N’s according to the position -and more if the position is tactical, etc. 

(To be continued)

Bruski

Mate in 3 moves. Bruski 1906

Ernst

Mate in 3 moves. G. Ernst. 

Happy New Year 2016 to all my readers. 2015 has been very bad for me but let me tell you I wish all your dreams come true this next year. Thank you very much for being there.

QChess.

Written by QChess

December 30, 2015 at 7:19 am

The WordPress.com Blog

The latest news on WordPress.com and the WordPress community.

%d bloggers like this: