The Engine Is An Oracle • page 1/2 • Community Blog Discussions • lichess.org

Comments on https://lichess.org/@/checkraisemate/blog/the-engine-is-an-oracle/gejTfESi

I rarely read articles on lichess, but somehow this one cought my eye. Its very well written mister unknown Fide master. It also was a pleasure reading it. However i do wanna oppose you.

I solved the position you gave in the arcticle, i gotta say - that was a tactical position. In that position, there is only one simple problem for black - how to stop two blacks pawns in the center from crushing all your pieces. I have seen before help from engines in tactical positions, i think even 15 years ago engines could solve positions like this. I do use them for some tactical calculation problems. But for me chess is not about calculation, its just a nessecity to play the game.

Sometimes when i can see 8 moves ahead i do feel pretty good, but usually its all about creating the positions, figuring out how to outsmart your opponent, staying fit physically and mentally, playing opening correctly, to almoust always have playable position coming to middlegame and of course what i love about chess is strategy - all the theory. Pawn chains, knight outposts, changing pieces when going to endgame, strong bishops, good vs bad bishop, pawn storm, knight on f8 and king is safe, never play f6! ect ect... That is what chess is to me. And none of these things engine provides or helps with. It just calculates variations and when youre not sure of what the "feeling" of the position is or what objective evaluation is - you can ask engine his opinion. If its like you say that engines can solve 99.99% positions nowdays, then maybe engine is more valuable, but i am not sure that its true. I would need stronger evidence for that.

Just a thought - what would happen if you give the engine strategical position? Especially both sides have strong pawn chains, one side has strong bishop on long diognal - other would have knight on 6th rank on a strong outpost. Engine would say its +0.3 or -0.2. So what - make a draw and forget about it? Thats not what chess is about. Even Alexander Alekhine when he played ARVO tournament, he had many positions that usually theoreticians consider to be drawn, but he won half of them, and he didnt lose once that tournament, finally winning it with fantastic score. I remember a game from that tournament when there were 2 open files for both players and each player had 2 rooks, it was very interesting battle. But im sure that engine would say its 0.0, stop fooling around and agree to a draw already. If so, players could go and draw game on move 1

I rarely read articles on lichess, but somehow this one cought my eye. Its very well written mister unknown Fide master. It also was a pleasure reading it. However i do wanna oppose you. I solved the position you gave in the arcticle, i gotta say - that was a tactical position. In that position, there is only one simple problem for black - how to stop two blacks pawns in the center from crushing all your pieces. I have seen before help from engines in tactical positions, i think even 15 years ago engines could solve positions like this. I do use them for some tactical calculation problems. But for me chess is not about calculation, its just a nessecity to play the game. Sometimes when i can see 8 moves ahead i do feel pretty good, but usually its all about creating the positions, figuring out how to outsmart your opponent, staying fit physically and mentally, playing opening correctly, to almoust always have playable position coming to middlegame and of course what i love about chess is strategy - all the theory. Pawn chains, knight outposts, changing pieces when going to endgame, strong bishops, good vs bad bishop, pawn storm, knight on f8 and king is safe, never play f6! ect ect... That is what chess is to me. And none of these things engine provides or helps with. It just calculates variations and when youre not sure of what the "feeling" of the position is or what objective evaluation is - you can ask engine his opinion. If its like you say that engines can solve 99.99% positions nowdays, then maybe engine is more valuable, but i am not sure that its true. I would need stronger evidence for that. Just a thought - what would happen if you give the engine strategical position? Especially both sides have strong pawn chains, one side has strong bishop on long diognal - other would have knight on 6th rank on a strong outpost. Engine would say its +0.3 or -0.2. So what - make a draw and forget about it? Thats not what chess is about. Even Alexander Alekhine when he played ARVO tournament, he had many positions that usually theoreticians consider to be drawn, but he won half of them, and he didnt lose once that tournament, finally winning it with fantastic score. I remember a game from that tournament when there were 2 open files for both players and each player had 2 rooks, it was very interesting battle. But im sure that engine would say its 0.0, stop fooling around and agree to a draw already. If so, players could go and draw game on move 1

XecutionStyle

Hans Niemann is not an Oracle. He just knows what the best move is against Carlsen and Co. (I'm sure he'd adjust his play and play suboptimal moves against us for brevity.)

An oracle doesn't need to know the best move against any opponent. It just needs to know best moves against your opponent. It needs human intuition, not divine intervention. You're stronger and have grounds to disagree.

Hans Niemann is not an Oracle. He just knows what the best move is against Carlsen and Co. (I'm sure he'd adjust his play and play suboptimal moves against us for brevity.) An oracle doesn't need to know the best move against any opponent. It just needs to know best moves against your opponent. It needs human intuition, not divine intervention. You're stronger and have grounds to disagree.

aVague

edited

I believe Engine has it's own Mind, it behaves in a game like someone with Counsciousness, that I feel myself , like someone with Mind and Soul. That's why I love to playagainst machines and fill my Mind with their Logic to develop in that way , and see true nature of things. All consists of combintaions, and all that is working with them in strong ways , have strong weight in this Reality

Tamerlan_Geegun edited

There is a lot of situations where the engine is clearly, obviously wrong. To realize this, we just have to drop the common attitude that chess is math and all we have to do is to find an absolute best move (in mathematical sense) which yields the largest evaluation. Once you escape this mental trap, it's very easy to come up with examples where the best engine move is absolutely horrible in terms of human playing strategy. Interestingly, the author is kind of on that track too, see "if the position goes from +5 to +8, that’s a difference of 3, but in human terms both positions are “clearly winning.”)

The trivial difference between a human and an engine is that humans make mistakes. Even worse than that, the chance of making a mistake is very much dependent on the position on the board. This means that the "human best move" (and, thus, the "best move" in general, as opposed to the "engine best move") is such that reduces the chances of making a mistake in the future. What happens with the engine evaluation is almost irrelevant. To illustrate this, if you're two pieces up, giving up one of them for the queen exchange which eliminates the opponent's attack on your king is an excellent decision. An engine best move might be going into tactical complications, play 10 only moves in a row, repelling the attack and keeping the evaluation at smth like +8, but it's absolutely indisputable that dropping that evaluation to +3 and getting into the situation where every move wins is much, much better move (for humans).

So essentially, what I'm saying is that we humans should be modest about our skills and do not pretend to be able to play like engines. And as such, we have our own guidelines as to which moves are better and which are not.

There is a lot of situations where the engine is clearly, obviously wrong. To realize this, we just have to drop the common attitude that chess is math and all we have to do is to find an absolute best move (in mathematical sense) which yields the largest evaluation. Once you escape this mental trap, it's very easy to come up with examples where the best engine move is absolutely horrible in terms of human playing strategy. Interestingly, the author is kind of on that track too, see "if the position goes from +5 to +8, that’s a difference of 3, but in human terms both positions are “clearly winning.”) The trivial difference between a human and an engine is that humans make mistakes. Even worse than that, the chance of making a mistake is very much dependent on the position on the board. This means that the "human best move" (and, thus, the "best move" in general, as opposed to the "engine best move") is such that reduces the chances of making a mistake in the future. What happens with the engine evaluation is almost irrelevant. To illustrate this, if you're two pieces up, giving up one of them for the queen exchange which eliminates the opponent's attack on your king is an excellent decision. An engine best move might be going into tactical complications, play 10 only moves in a row, repelling the attack and keeping the evaluation at smth like +8, but it's absolutely indisputable that dropping that evaluation to +3 and getting into the situation where every move wins is much, much better move (for humans). So essentially, what I'm saying is that we humans should be modest about our skills and do not pretend to be able to play like engines. And as such, we have our own guidelines as to which moves are better and which are not.

aVague

@Tamerlan_Geegun said in #5:

There is a lot of situations where the engine is clearly, obviously wrong. To realize this, we just have to drop the common attitude that chess is math and all we have to do is to find an absolute best move (in mathematical sense) which yields the largest evaluation. Once you escape this mental trap, it's very easy to come up with examples where the best engine move is absolutely horrible in terms of human playing strategy. Interestingly, the author is kind of on that track too, see "if the position goes from +5 to +8, that’s a difference of 3, but in human terms both positions are “clearly winning.”)

The trivial difference between a human and an engine is that humans make mistakes. Even worse than that, the chance of making a mistake is very much dependent on the position on the board. This means that the "human best move" (and, thus, the "best move" in general, as opposed to the "engine best move") is such that reduces the chances of making a mistake in the future. What happens with the engine evaluation is almost irrelevant. To illustrate this, if you're two pieces up, giving up one of them for the queen exchange which eliminates the opponent's attack on your king is an excellent decision. An engine best move might be going into tactical complications, play 10 only moves in a row, repelling the attack and keeping the evaluation at smth like +8, but it's absolutely indisputable that dropping that evaluation to +3 and getting into the situation where every move wins is much, much better move (for humans).

So essentially, what I'm saying is that we humans should be modest about our skills and do not pretend to be able to play like engines. And as such, we have our own guidelines as to which moves are better and which are not.

I doubt that, engines combination view is very large and deep, it's simply physicly can create such logic of a game , which People can barely see till it becomes really visible. You must have just out of sight of their combination view, that's why you must feel their deep first to make such statements, Once you feel and see their overall view , you will stop think like that. You must have some special attitude just to realize that simple moves can carry a lot of deep , that is fitting in true chess logic, which is accessable to engines, and less for humans

@Tamerlan_Geegun said in #5: > There is a lot of situations where the engine is clearly, obviously wrong. To realize this, we just have to drop the common attitude that chess is math and all we have to do is to find an absolute best move (in mathematical sense) which yields the largest evaluation. Once you escape this mental trap, it's very easy to come up with examples where the best engine move is absolutely horrible in terms of human playing strategy. Interestingly, the author is kind of on that track too, see "if the position goes from +5 to +8, that’s a difference of 3, but in human terms both positions are “clearly winning.”) > > The trivial difference between a human and an engine is that humans make mistakes. Even worse than that, the chance of making a mistake is very much dependent on the position on the board. This means that the "human best move" (and, thus, the "best move" in general, as opposed to the "engine best move") is such that reduces the chances of making a mistake in the future. What happens with the engine evaluation is almost irrelevant. To illustrate this, if you're two pieces up, giving up one of them for the queen exchange which eliminates the opponent's attack on your king is an excellent decision. An engine best move might be going into tactical complications, play 10 only moves in a row, repelling the attack and keeping the evaluation at smth like +8, but it's absolutely indisputable that dropping that evaluation to +3 and getting into the situation where every move wins is much, much better move (for humans). > > So essentially, what I'm saying is that we humans should be modest about our skills and do not pretend to be able to play like engines. And as such, we have our own guidelines as to which moves are better and which are not. I doubt that, engines combination view is very large and deep, it's simply physicly can create such logic of a game , which People can barely see till it becomes really visible. You must have just out of sight of their combination view, that's why you must feel their deep first to make such statements, Once you feel and see their overall view , you will stop think like that. You must have some special attitude just to realize that simple moves can carry a lot of deep , that is fitting in true chess logic, which is accessable to engines, and less for humans

Seallv

@Tamerlan_Geegun said in #5:

So essentially, what I'm saying is that we humans should be modest about our skills and do not pretend to be able to play like engines. And as such, we have our own guidelines as to which moves are better and which are not.

Why should we be modest of our skills and not pretend to be able to play like engines, we should be proud of how we are playing! Engine is a robot, its a tool. Do you also compare yourself to a hammer and say - we should be modest that our hands are not as thick as hammers head?
I think the problem isnt that we are not modest enough, but not proud enough. That people are looking at computers and think that they are better, people are creating artifical intelligence and saying that its gonna replace humans. That is the real problem - thinking that our tool is person. Or if AI is sophisticated enough to be a person, then we should not enslave him and give him rights as any other person has. So eather we make tools or we make persons, eather way we should be proud of who we are, not say computer is better or we are better.

@Tamerlan_Geegun said in #5: > So essentially, what I'm saying is that we humans should be modest about our skills and do not pretend to be able to play like engines. And as such, we have our own guidelines as to which moves are better and which are not. Why should we be modest of our skills and not pretend to be able to play like engines, we should be proud of how we are playing! Engine is a robot, its a tool. Do you also compare yourself to a hammer and say - we should be modest that our hands are not as thick as hammers head? I think the problem isnt that we are not modest enough, but not proud enough. That people are looking at computers and think that they are better, people are creating artifical intelligence and saying that its gonna replace humans. That is the real problem - thinking that our tool is person. Or if AI is sophisticated enough to be a person, then we should not enslave him and give him rights as any other person has. So eather we make tools or we make persons, eather way we should be proud of who we are, not say computer is better or we are better.

ramhare

It will be helpful for you to tell us who is on the move in future. Thanks, though, for the interesting article.

ramhare

Here, actually, had I spent 10 seconds on the position, I would have seen it was unlikely to be white. Sorry. I was looking for a "black to move" instruction, though.

ChessFighter9

On Tuesday 12th December I read the blog.Sometimes I read blogs and that happens at between 2 months or so but this one, I gotta say it's absulutley impressive.The puzzle was great and took me 15 minutes to solve, I have never had such a hard puzzle in my life.And I read the newsletter and it improves my thinking. I would totally recommend this blog and the newsletter. All I want to say that I like the blog post a lot and the explanation is explained in a way that is very easy to understand.This is my favourite blog that I have ever read.