Solving chess

Maybe, I do not know.  What is this?

The author of this thread (and some of the followers for that matter) are a bit too obsessed with statistics. All it takes is one little bit of randomness to totally mess up your whole day. Chess isn't "solve-able" like a mathematical formula. If it were we would have already soved it (or a computer that can calculate about 10,000 times faster than a person can woud have). Some of the greatest minds in the world have been chess players. Many people have devoted their entire LIVES to its study. As long as human beings play eachother you have human elements that factor in, like emotion. White may be proved to have a slight advantage statistically, but tell that to the Grandmaster who just played white and lost. Human game.

@SavageLotus:

(1) computers are currently the strongest chess players, and their lead is growing

(2) chess is definitely solvable in theory, but the computational demands are prohibitive. A lot more than 10,000 x the speed of a human (add 50 zeros and then some more)

(3) the complexity and size of the game of chess allows plenty of room for humans to have fun playing against each other

________________________________________________________________________________

"In theory, theory and practice are the same. In practice, they are not" - Lawrence Peter Berra

Oh this is too precious!

It's pretty pathetic to try to act like the chess.com police here. Seriously. You're a general member, and not even a paid one. Also, you need to look up the definition of an ad hominem attack. I'm not saying because you're wrong about 10 things, you're wrong about everything else (after all, we do agree that chess is probably a draw, right?) - I'm saying that you're wrong, and listing the ways. In fact, even if I did say that, while it's still technically an ad hominem attack, saying "you've been wrong 10 times in this thread" is a pretty good indicator that you'll continue to be wrong (and you just keep proving that right), but it's not a guarantee, which is why I didn't say it. Chess.com also does not have an ad hominem attack detector, and ban people for for making a logical fallacy. It bans people for abuse, which I have not engaged in. Complain to whoever you like.

You think because I list on my profile that I'm a musician, that that is all I am? I have studied maths. Admittedly only an undergrad degree, so it's likely you are correct about being beyond that, but I have studied it, and have an understanding of it well beyond my formal education, because I continued learning it on my own since. Game theory is not an area of it I have payed much attention to, because I feel it is trivial (my pet field is number theory), and I already have too many other trivial things in my life in which I indulge. But knowing that with perfect play, the game is win/draw/loss and only mistakes change that? Come on, man, you don't need to have studied at all for that, it's just basic common sense. I slipped on terminology, you assumed ruby7 meant advantage in a game theory sense when they didn't. Lighten up!

And I'm not arguing because I feel snubbed. I'm arguing because I enjoy challenging fragile egos like yours (there are so many of them in chess at all levels). It's pretty hilarious to see how much you will dance around saying that you're wrong without actually saying it.

Since you seem to think it worthy of comment, my name comes from Franz Schubert, my favourite composer, and from my past when I'd play minesweeper (another trivial thing I've removed from my life except when something takes forever to print), where you are dealing with bombs. I found it amusing (humour! have you heard of it?) that combining the two had connotations connected to the pathetically botched attempt of someone or other to ignite a shoe bomb which had happened just before I started using this handle.

@Schubomb, I am sorry that this discussion became so unpleasant and unproductive. As Dale Carnegie wisely said "you can't win an argument" (his "you" meant anyone, of course).

Assuming you have the maturity to draw hostilities to a close, it might be productive to point out that we appear to agree completely on the simple game theoretic result that chess has a definite value. You fully understand the part of game theory that is relevant to this discussion about chess (many others do not). While the enjoyable time I have studying game theory (and doing some original research) on and off since I graduated may give me a greater knowledge of the broader subject than you (and makes me realise it is deep, powerful and fascinating rather than being "trivial") this is irrelevant to this discussion and hardly relevant to chess.

[I've often pondered on why people with my original qualification are called wranglers .]

Hasn't been unpleasant for me :) You're constantly good value - I mean, look at how quickly you backed off your you're-going-to-get-banned! stupidity - you can turn on a dime! As for unproductive, if I was that worried about productivity, I'd cancel my chess.com account.

I'd love to let you get to know my more mature side, but what you call "draw hostilities to a close" and "maturity" really means "agree with everything I say, or at the very least don't point out when I'm wrong, and bow down to my superior age, experience and rating" and "please don't tease me when I say something silly". Admittedly the second part has something to do with maturity, but the first part is the opposite. I'd be happy to engage in serious discussion on any topic that I have some interest in with someone who is open-minded and has a sense of humour and humility about themself, though.

You aren't, and never will be that person. (edit: hence, the only way I can enjoy interaction with you is to mock you)

Here's some news that might mean that we're closer to solving chess:

http://www.physorg.com/news/2011-06-d-wave-commercial-quantum.html

Fascinating news! Perhaps quite a few steps to solving chess though.

[deleted duplicate post]

Oh no, this ad hominem, now everybody speaking latin, I told you chessplayers are so smart!!

Is this STILL going on?!?!

Native Latin speakers think the same about people who speak English.

That chess is a theoretical draw is an educated guess, not a fact.

Chess won't be solved soon with conventional computers. There is a faint chance of it being solved using some sort of quantum computer (this may or may not be physically possible).

88-100 moves might be how many is needed. Or maybe less. Or it might be 200 moves or more! Remember the perfect computer has to make correct moves in all positions, so it's the most deep and difficult position that is the significant one.

Perfect play may be able to beat Houdini, or a computer that crushed Houdini almost all the time. We can't be certain about this yet. But if it is true (as I believe) perfect play is further above Kasparov in strength than Kasparov is above me. Which is hilarious!

I am pretty sure some of my above statements sound like me questioning someone who was agreeing with my statement earlier ...

I don't think that's accurate. The advantage of .15 would show that white has an easier time winning than black.

Repeating here what I wrote in my blog about a perfect game.

First, what is a perfect game of chess?  I think a perfect game is one in which each player, (White and Black), make the best move possible from the first move onward.  The best move possible would be the one move, given the current position, that applies the most force by the most economical unit(s) toward the enemy forces while doing the least amount of damage to one's own forces or position.

Second, at the very beginning of the game, both sides, White and Black, are exactly equal.  They control the same amount of squares on the board and have the same levels of power.  Everything is balanced.  (A possible exception is that White has his King on the right and Black has his King on the left.) 

To start the game, White has the first move.  Black could copy White's moves in the game  to keep the balance of power and position equal; but, there does come a point in the game where White can, and often does make a move that Black can not copy.  Such as a check or capture the leaves Black with no opportunity to bring the board back into symmetry.

Chess, in its current form, has been played for over 300+ years.  One would think that if there was a perfect opening move, that was proven to win for White everytime (barring no subsequent errors on White's part), that everyone from novice to the highest grandmaster would be playing it.    Instead, there are numerous openings being played at all levels.  In all the games that have been recorded, what opening move(s) have the highest number of wins and draws.  (And, has a significant winning/drawing percentage over 50%, say around 75% and above?  Two sides, White and Black, one wins/draws, one loses/draws.)

Because of the existence of all the various openings I wonder if there is such a thing as a perfect game in chess.  I tend to think not.  I suppose that there could be a perfect game for each specific opening; but, no perfect game for chess as a whole.  It leads to the question, "Has anyone ever played, or will play,  a perfect game using a specific opening"?  That means that will be a perfect game possible for the Ruy Lopez, the Silician, the English, the Bird, etc.  Or, it is possible that even the openings do not have perfect game?  But, back to the topic at hand:  If there is no such thing as a perfect game as a whole for chess, or even the openings, how can it be said that chess has been solved even though chess does have a limited and finite number of positions?

Dee

@defrancis7: I think a better definition of a perfect game is a game where neither side ever has the chance to play a move that forces their win. I.e., if it were an endgame tablebase, the only possibility for descriptions of all possible moves in any position, for either side, is "Draw in X moves" or "Lose in Y moves"; NEVER "Win in Z moves". This means there are several, probably many, "perfect" games. There is, I believe, one ideal game - but this is more difficult to assess. One idea is that the ideal game is the perfect game which gives each opponent the most chances of passing an opportunity to "Win in Z moves", on each move. But for human chess, this would need to be assessed in terms of how difficult it is for the opponent to find the correct defensive moves. For that reason the ideal game might be a sicilian, rather than a Caro-Kann (to use the stereotypes for these openings) since the sicilian might offer the most chances for each opponent to lose on each move.

Your ideas on making a perfect game more perfect seem rather subjective choices. A simpler one is to say that the best line is the one that reaches the theoretical result quickest (against most obstructive opponent play). People often take this for granted in a calculable ending (slower wins are slightly inferior) but, in principle, the same criterion could be applied to drawn positions, where one player tries to prove a draw (so that they can go and do something more worthwhile), while the other one tries to delay the end of the game as long as possible (because they have the typical chess player mentality).

What it "seems" to you may or may not be the truth. The jury is out on quantum computation. What you may be missing is that if it becomes possible to build quantum computers with large number of bits, the rate of increase of speed of computation will explode at a stupendous rate compared with the snail's pace of doubling less than once a year. If fairly large quantum computers prove not to be possible, they will have no effect at all on the speed of computation. Very different possible futures.

If you read my post, I think you'll see I was expressing my uncertainty about how strong a perfect player is. Glad we agree on that. I did estimate this quantitatively, but I hope I made it quite clear that my results were not to be treated as being a reliable indication - the question is intrinsically very difficult.

A perfect game (in the simplest sense) is where neither player ever makes a mistake that changes the result his opponent can achive, with the (unrealistic) assumption that we have the 32-piece tablebase of chess available to determine this. This is a bit more precise than your statement.

Technically, yes -- but practically black has to neutralize the advantage; the fact that white does not have to go through this process is an advantage in itself.

Yes.

The only reason I didn't want to get into that is because you could also argue that what we call "objectively worse" can have practical advantages as well, like being simpler to play or something. So this .15 is like some sort of "objective practical advantage." But not really :) I guess it's just another type of practical advantage.