World Computer Chess Championships Underway

azaris writes "While the FIDE World Championships for human players in Tripoli, Libya are down to the last two contestants, the computers are playing their own 12th World Computer Chess Championship in Ramat-Gan, Israel. How will the open source chess engine Crafty do against the proprietary closed engines? Will the computers play more interesting chess than their human counterparts?"

Comments

Well...

[Anonymous Coward]

As long as you don't give them exclusive control to the pod bay door, I think computers should be allowed to play chess as they please.

Re:Well...

[Anonymous Coward]

Ah.. but if the computers "please" to play chess, they must have intelligence.. artificial intelligence. And given the extreme intelligence of these chess playing computers, they could give themself exclusive access to the pod bay doors!

Re:Well...

[rd4tech]

as an extra measure for blackmail if they are about to lose their chess game

Re:Well...

[Edward W.]

But a computer needs control of the pod bay door to play chess to the best of its ability. Recall the scene in 2001 where the humanoid spaceship Discovery opens one of its three mouths (pod bay doors), sticks out its tongue (pod launching ramp), blows a bubble (spherical space pod), and watches the bubble rise over his head. This tactic could be used to great advantage in a chess game. The opponent would naturally be surprised by this playful display of anthropomorphism and would watch the bubble rise. The computer would then quickly open a second mouth, lick up a nearby bishop (or rook, whatever), and close its mouth with the enemy piece inside. Returning her attention to the board, the enemy would get all rattled when she (Judit Polgar?) saw her position was weaker than she recalled. This would be the beginning of the end.

Re:Well...

[hunterx11]

The chess problem in 2001 was really well set up, but unfortunately they used descriptive notation which had been overwhelming replaced with algebraic notation well before 2001!

Re:Well...

[Fred_A]

I'm so disapointed, I read the story title as a "computer cheese championship" and it's just a boring board game. What a let down.

Re:Well...

[Wyrmw00d]

Did you set the resolution to 1600x1200 on a 15" monitor again?

Interesting computer Chess?

[gambit3]

Will the computers play more interesting chess than their human counterparts?"

I don't think so (replying to the question posed by the original poster), because I believe a well-programmed algorithm would care only about winning, and not necessarily taking chances or exploring possibilities that a human player would...

Re:Interesting computer Chess?

[kyle_b_gorman]

you think kasparov is interested in any move that won't (at least indirectly) help him win?

Re:Interesting computer Chess?

[Kenja]

Also, the computer isn't going to throw a hissyfit when they lose. Or am I the only one who watches chess just for the losing nerd thorwing a complete wobbler and whining like a little girl?

Re:Interesting computer Chess?

[hunterx11]

True...someone like Mikhail Tal would probably fare poorly against computers if he played the kind of spectacular, speculative, and psychologically devastating sacrifices that made him so popular.

Actually, I think the real reason the computer drew in the latest matches against Kasparov and Kramnik was psychology more than anything else. The computer does not get stressed or fatigued when it is under pressure, nor does it lose morale after a blunder (like Kramnik) or have any fears of losing (like Kasparov).

Re:Interesting computer Chess?

[ookabooka]

True, except human players can make simple mistakes. A computer never makes a mistake, it just only looks so far. Deep Fritz 8 has beatten Kasporov. I mean, if you can evaluate 2.4 million positions a second (i can on my dually) there does come a point where a human player cant keep up, what if that were 24 million, 240 million? Computers dont have finite limits like a person. It used to take a super computer to even compete with a GM (grandmaster) now the average home computer can give him a run for his money.

Re:Interesting computer Chess?

[Short Circuit]

Except a human has this neat thing called intuition.

Your intuition can tell you things that will take you hours and hours to prove on paper. Or even in your head, following logic.

Re:Interesting computer Chess?

[ChuckleBug]

A computer never makes a mistake, it just only looks so far.

This is equivalent to when Bill Gates said Microsoft's software has no bugs.

Re:Interesting computer Chess?

[TastyWords]

Tal may not have had as many problems as suggested. One of the issues for the computerized chess is when there are no time limits - grind, grind, grind, ...

When they're playing on the clock, it makes a world of difference.

Re:Interesting computer Chess?

[jeblucas]

If the computer considers the library of Emmanuel Lasker, then it could be "more interesting". He was (in)famous for making -ahem- startling moves for someone of his caliber. Traditional chess theory would call them blunders, but they would serve to complicate the board to a degree that his opponents could not as easily determine the best moves before he could. He would thus confuse, recover, and gain advantage before they could adequately respond. A chess program designed to confuse more rigid chess programs could serve to benefit in the same way Lasker did.

Re:Interesting computer Chess?

[TastyWords]

Kasparov *did* admit to screwing the pooch on his first match vs. the computer: he attempted to determine (or obtain research of) what the "weaknesses" of the software was and attempted to play to that and put too much weight on the analysis instead of playing his own game. Right or wrong, this is what he said in the press.

Re:Interesting computer Chess?

[pnot]

Except a human has this neat thing called intuition.

Your intuition can tell you things that will take you hours and hours to prove on paper.

Indeed. It seems intuitively obvious to me that the earth is flat and the sun moves round it, and I *still* haven't managed to prove it on paper.

Re:Interesting computer Chess?

[nacturation]

Except a human has this neat thing called intuition.

Your intuition can tell you things that will take you hours and hours to prove on paper. Or even in your head, following logic.

Intuition is not much more than having a large sample set from which to draw and using that sample set to infer generalities. Those generalities allow you to recognize certain patterns and also reject other patterns outright so that you don't consciously consider them. When you think about it, intuition really means "I have a hunch", and those hunches are formed on the basis of past experience. There's nothing that prevents a computer from building up a sufficiently large body of samples and, with the proper programming of course, inferring patterns from it as well.

Re:Interesting computer Chess?

[sokoban]

Well, computers have played some pretty damn interesting chess in the past. Let's take the Deep Blue-Kasparov matches as a much overused and overhyped example:
1996 game 1: 23. d5 was a brilliant move on deep blue's part. I would hope that I would find such a good move there. Yes, the computer found that move through calculation, but the move itself shows a great understanding of spatial and pawn structure elements of the position.

1997 game 2: this game was riddled with awesome moves, but 23. Rec1 33. Nf5 and 24. Ra3 are the cream of the crop.
23 Rec1 is annoying and almost a human move. The computer is playing almost perfectly here. I think Kasparov has very little counterplay here. Kasparov's queenside is UGLY and he gets little to nothing in compensation.
24. Ra3 just rocks here. Deep blue is playing the Ruy Loppez like he means it. That move made me really wonder about who was behind the computer. The Ruy Lopez is a rich opening with lots of crazy details regarding strategies in each variation, but deep blue nailed them like any world level player who playes the lopez should. Basically, deep blue couldn't have forseen a Lopez variation, but found the correct strategies all the same. Also in this game, Deep Blue psyched Kasparov out of a draw. In his top form, Kasparov WOULD HAVE SEEN the draw. 45... Qe3 does it. At the least, it is a draw by perpetual check, and if Deep Blue tried to stop it he gets crushed.
33. Nf5 is a very "computer" move, but really blows away Kaspy. It doesn't make sense muc really, but brings the bishop into play fast and kind of psyched out Kasparov.

So you see, computers can play interesting chess. These are only 2 great games I have around of a computer-human match. there are others, but theses are the most dramatic.

Re:Interesting computer Chess?

[AshtangiMan]

yes he is . . . but not from the economic only standpoint that the computer algorithms demand. Looking at even recent Kasparov games shows that he does not play the economy equation the way computers do. This is why he (and the other human) chess players are infinitely interesting to watch, while the computer, while nearly unbeatable, is also very boring.

Re:Interesting computer Chess?

[raytracer]

Except a human has this neat thing called intuition.

Your intuition can tell you things that will take you hours and hours to prove on paper. Or even in your head, following logic.

Indeed. Once in a great, great while, your intuition may even tell you something which turns out to actually be true.

Re:Interesting computer Chess?

[thedillybar]

>Computers dont have finite limits like a person.

Although the limits are very large, they are finite in the example of the computer. This can be shown trivially.

What makes you think the human mind is so limited? I am not convinced that a human can't evaluate nearly the same number of possibilities as a computer can in a set amount of time. It's much easier for the human to rule out possibilities, one thing it is difficult for the computer to do. While humans are still playing quite well against computers, I wouldn't count us out now, or ten years from now.

Re:Interesting computer Chess?

[Phrogger]

Such a move, as we in science term it, is a bad move. :-)

Re:Interesting computer Chess?

[bobhagopian]

Computers dont have finite limits like a person. That's true. I can only count to 32768.

Re:Interesting computer Chess?

[cynical kane]

32768? I can only count to 32767!

Re:Interesting computer Chess?

[BiggerIsBetter]

I noticed you used the word "designed". Any idea if anyone has tried evolving chess playing algorithms?

Re:Interesting computer Chess?

[medelliadegray]

intuition can be great when you can go back and say "well that hunch didnt pan out, lets try soething else"

intuition is not always correct.

Re:Interesting computer Chess?

[michael_cain]

Those generalities allow you to recognize certain patterns and also reject other patterns

Don't the top-notch computer programs already do this in one form or another? I mean, it seems like such an obvious line of research for pattern matching and pattern recognition people to explore. Or perhaps not -- that may be one of the reasons that computers aren't as good at Go as they are at chess, Go appearing (to a rank beginner) to depend more on pattern recognition and less on straight-forward deductive analysis.

Re:Interesting computer Chess?

[Anonymous Coward]

Actually not. See Godel for details.

Re:Interesting computer Chess?

[Anonymous Coward]

computers have finite and well knowns limits unlike a person. A good player can still win a game with more possibilities (like the chinese go ) against any computer.

Re:Interesting computer Chess?

[boardumb]

you didn't see the under 20 move draw between the two finalists last night did you? zzzzzz...

at least computers will fight to the death, or until the last few bits can't mate anymore.

Re:Interesting computer Chess?

[TygerFish]

Intuition is not much more than having a large sample set from which to draw and using that

sample set to infer generalities. Those generalities allow you to recognize certain patterns and also reject other patterns outright so that you don't consciously consider them. When you think about it, intuition really means "I have a hunch", and those hunches are formed on the basis of past experience. There's nothing that prevents a computer from building up a sufficiently large body of samples and, with the proper programming of course, inferring patterns from it as well.

Intuition plays a strong role in the play of human players great and small and it is the basis by which one can understand the differences not only between human and computer players but between interesting and uninteresting chess games.

Having a hunch about the nature of a position and the posibilities therein have allowed some of the greatest tactical games ever played and this is the identifying characteristic of the matter in understanding the nature of the game. Were humans different, the game before computers would have been different: humans who saw every possibility in a continuation leading to a 'decided' position at the end of each line would have simply announced the result or range of results and the nature of chess itself would be unrecognizable to us ('Mate in at least 37 or at most 103!').

Human intuition allows the 'miracles' of chess--the elegance of chess--in those games that make the game breathtaking and that inspire players to play in the hope of generating them (think of classics like Morphy-v-Consultants or Lasker-v-thomas or many of Mikhail Tal's best games). The intuition or, indeed, inspiration, of games like those are more than instances of the inference from generalities; they are instances of a grandiose specific arising from a game's sea of possibilities. It is the elegance of a queen sacrifice leading not to ineluctable mate (a combination like the end of Morphy-v-consultants), but to a powerful attack with a favorable conclusion (say, the end of Lasker-v-Thomas or of Reti-v-Capablanca) which, as an act, is as difficult to quantify as is the word, 'beauty.'

In a broad sense, a machine's ability to process advantage takes the wonder out of the thing because you know that there is nothing going on but the examination of a great number of positions but it is hard to imagine to imagine programmers 'weighting' their programs for positions conducive to the types of continuations that made of chess-players bother with chess in the first place.

The short form of all of the above is: 'computer programs either are or will soon be the strongest players on earth, but their games tend to be dry.'

Re:Interesting computer Chess?

[mbrother]

I used to play high school chess (hey, I was the cool one on the team...really!). There was this kid we called "The Zapper." He had cerbreal palsy and most people, the first time they saw him, respected him for working against his physical limitations and competing like anyone else. And he wasn't that bad for his age. But...he was a terrible loser! He would literally throw hissy fits and knock the pieces all over the table, the floor, where ever, when he lost, and yell out obscure threats. He was one of these people who used his condition as an excuse to be a big baby whiner, and I found it quite shocking. It helped me see past handicaps to the people beyond, good and bad. Good people come in all shapes and sizes, and so do bad people.

As a teenage chess player, I had long hair and listened to loud, hard rock and metal on my walkman, but I would play really boring, solid moves. I got a draw off Boris Spassky in an exhibition once playing the Caro-Kann. My friend played a double King Pawn and lost in 5 hours in a wild King's Gambit game, the last game going. I kind of wish I'd played more aggresively now, although I cherished the draw for many years and had a calculus test to study for.

Re:Interesting computer Chess?

[Edward W.]

A human player of Kasparov's stature may well be able to "keep up" without being able to match the computer's ability to evaluate 2.4 million positions a second. Two things explain the human's ability to equal or outperform the computer.

First, only a tiny fraction of those positions need to be evaluated. All but a small fraction of them result from the human's making one of a large number of foolish next moves -- for example, a pointless king move when the king is under no immediate threat and in a safe position. And at each move up the line all but a small fraction of the moves are again not worth considering. So the computer is wasting time evaluating millions of positions that don't require evaluation.

Second, the quality of the evaluations is at least as important as the quantity of evaluations. Kasparov and his peers (and near peers) can apply prodigious amounts of theoretical knowledge and experience to evaluate the subtleties of a position. Winning evaluation algorithms are far more complex than adding up the standard point values (queen = 10 or 9 1/2, rook = 5, bishop = 3, etc.) for the pieces of each side at a certain position eighteen moves down the line. Pieces must be evaluated according to complex circumstances. Doubled rooks are worth more, and how much more depends on the precise situation. A queen doubled behind a rook is more valuable than a rook behind a queen (the rook can be exchanged, the queen generally can't). When both sides have just one bishop, the bishop's value to each side may depend on whether the bishops are of the same or opposite colors when one side or the other has a material advantage. In late middle game and end game situations, a pair of bishops is worth more than a bishop and a knight or two knights. Knights at the edge of the board or -- worse -- in or beside a corner -- are worth less. Blocked bishops and rooks are worth less. Isolated pawns and doubled pawns are worth less. A piece vulnerable to exchange is worth less to the player with a material disadvantage. Numerous other considerations affect the value of the pieces. The adjustments in value are subtle and extremely hard to make mechanically; they require the judgement that the grandmaster can bring to bear. A computer must put specific values on each piece in each situation, probably using increments of 0.1 point (where pawn = 1.0 at start). This is no easy task.

Beyond revaluation of the pieces according to circumstances, values must be places on positional factors. Does the position yield the initiative to the opponent, who can (say) check your king or attack a piece? How many pieces are aimed at the enemy king? How well defended are the kings? Has the ability to castle been lost? How many king moves, and how many enemy pawn moves, will it take to reach a promotion square that the king must defend? Many other far more subtle positional factors must be evaluated. The possible situations are so numerous that I doubt the ability of any computer algorithm to account for all of the point adjustments that must be made, and to assign realistic values to each variation or complication. But humans know how to do this without resorting to a mechanical algorithm of the sort a computer must use.

Re:Interesting computer Chess?

[hunterx11]

Computers are getting better at chess, but so are humans. In fact some people think that computers are getting better no faster than humans.

Re:Interesting computer Chess?

[nacturation]

The short form of all of the above is: 'computer programs either are or will soon be the strongest players on earth, but their games tend to be dry.'

And what causes that? Computers play always at 100% of their maximum possible strength. That's why you'd think they're dry -- it's the consistency. Just like great boxing fights aren't ones where the opponent comes out and always knocks the other guy out cold in the first 10 seconds. A great fight is where it's undecided up until the very last round and then someone lands a stunning blow and wins the victory. Yet the guy who can knock his opponent out in 10 seconds every time is by far the superior boxer (judging by results alone).

Humans play chess, on the other hand, probably at some fraction of what they're capable of and, every once in a while, hit that 100% mark. Hitting 100% looks like inspiration compared to normal, more mediocre play but if every person hit 100% all the time, their games would be similarly boring -- though boring at a much higher level of play.

Re:Interesting computer Chess?

[helfen]

But look at Kasparov vs Deep Junior matches (and click on the "Follow the Match in 2D"). Have you noticed a 10th move of Deep Junior in game 5? Isn't this sacrifice intersting, especially when computer plays like that.

Re:Interesting computer Chess?

[TygerFish]

And what causes that? Computers play always at 100% of their maximum possible strength. That's why you'd think they're dry -- it's the consistency. Just like great boxing fights aren't ones where the opponent comes out and always knocks the other guy out cold in the first 10 seconds. A great fight is where it's undecided up until the very last round and then someone lands a stunning blow and wins the victory. Yet the guy who can knock his opponent out in 10 seconds every time is by far the superior boxer (judging by results alone).

Humans play chess, on the other hand, probably at some fraction of what they're capable of and, every once in a while, hit that 100% mark. Hitting 100% looks like inspiration compared to normal, more mediocre play but if every person hit 100% all the time, their games would be similarly boring -- though boring at a much higher level of play.

I should probably not answer your assertions, not because they are undeserving, but because they are *too* interesting. I sense that you are contentious by nature and I don't have the time to do what you write real justice and still give my own projects the time they need. Be that as it may, I can write that I disagree with you on several levels and tell you why.

Yes, it's true that computer programs operate consistently--right down to their flaws--but that is not the thing that makes computer chess games sterile and often devoid of beauty. When talking about humans versus humans and the role of intuition in chess you eventually come to look at the thing inside out as you come to the point where in order to discuss a game between two people, you have to look at it using methods and language ordinarily used only to talk about the mechanics of the thing. You have to talk about the game as if it were a program.

A chess game is a symmetrical set of operations forming a perfect illustration of the verbal description of von Neumann's Theory of Games. It is a two-person, zero-sum game with complete information. It is only asymmetrical in that the second player has the disadvantage of the second move and must make some response to white's initiative (the tens- if not hundreds of thousands of pages devoted to opening theory are the direct result of this fact). With this in mind, we must recognize that playing a chess game involves the creation of one or more imbalances in the factors of a chess game: time, space or force and that the sharpest and most radical imbalances are the key characteristic of the most interesting games.

Returning to Morphy-v-Consultants for a moment you see one of chess history's greatest examples of this thesis: in the final position, white has exchanged or sacrificed literally every piece but the king, one bishop and one rook which together deliver mate.

Now it need not be that crude: the winning imbalance in a game need be nothing more than good knight versus bad bishop in an endgame with material parity and that in itself is exciting in its demonstrating that the margin for winning can be that slim and strategy that deep.

Computers, which as you point out play at 100% of their strength at all times, will, when well-matched to their opponents, not find, or use the pathways that lead to wildly imbalanced positions.

The thing analogous to the knock-out punch you mention is nearly always absent. The denouement of games that live in your head forever isn't there; the problem-like mating combinations aren't there like the time Alekhine finished a game with Qg6!! Where it could be captured by either of two pawns leading to mate in the next move is never there, but that isn't the full measure of what is there in computers, nor is it the sum of what is missing.

Using your example, we return to mechanics and see that the knock-out punch in ten seconds is a possibility and one that is infinitely desirable: it is the ultimate elegance of which a boxing match is capable. For it to happen, one boxer has to create an opening for a hand

Re:Interesting computer Chess?

[nacturation]

The same thing is true of chess games. The creation of a (sudden) change in the balance and the exploitation of that change is an expression of the real elegance within chess which makes it more interesting than watching videotapes of accountants.

I agree with you, though I believe the fact that computers don't currently do this is due to a lack of sufficient "foresight". Humans create an imbalance not to merely spice up the game, but because they see an eventual advantage from it. Given that a chess program can only look so many levels ahead, it is typically programmed to only consider moves which lead to an advantage in order to cut down the nodes it must search. When programs become sufficiently fast, powerful, and flexible, I think we'll start seeing the imbalances of which you speak in grandmaster level games... but only when a program has sufficient confidence that the imbalance created will eventually lead to a positive outcome.

Re:Interesting computer Chess?

[timts]

computer can do all those you mentioned, as long as you "program" it to do so.
a computer which can really think might come out too

Re:Interesting computer Chess?

[fish_in_the_c]

Oh really, how do you know what is the cause of human intuition? Last time I checked it was an unexplained phenomena. Maybe we will understand it better when we have some idea how we store memories, but even that is almost unknown and very poorly understood. Intuition could be caused by all kinds of things we don't understand perhaps our brains have a relation to quantum physics that allows us to actually have some feeling as to what will happen in the future. My point is no one really knows ... or even comes close to having an idea. You are stating a specific theory as if it is a fact and that is always dangerous. Or at least somewhat silly as many people who believed the earth was flat and the humors of the body had to be in balance for health could attest to if they were still alive.

Re:Interesting computer Chess?

[nacturation]

Oh really, how do you know what is the cause of human intuition?

I'm stating what I think is how intuition works. Obviously, you may disagree with it. Feel free to argue why you think computers may never achieve what we perceive as intuition, but don't take the approach of ridiculing my attempt at an explanation.

Intuition could be caused by all kinds of things we don't understand perhaps our brains have a relation to quantum physics that allows us to actually have some feeling as to what will happen in the future.

The method of how our brains work is fairly well understood for the most part. Neurons firing, synapses, exchanges of chemicals, etc. We don't yet know all the facts, but the basics are understood enough for form theories. But given that there's something like 100 billion neurons in the brain and we don't know how it is all wired up, it's a bit difficult to speculate. Perhaps you're right. Maybe there's some kind of quantum entanglement going on and intuition is the mechanism by which we perceive things. However, given that the brain is just a bunch of cells which started out as sperm and egg and have divided, then it all but rules out this possibility. It's not as if the uterus is an appropriate environment for quantum entanglement to occur. Or even if it is the case, we could still produce computers which are biological in nature and share the same properties.

My point is no one really knows ... or even comes close to having an idea.

I have an idea yet you dismiss it without stating reasons other than waving your hands and saying "well nobody knows". Are you against the exploration of new ideas?

You are stating a specific theory as if it is a fact and that is always dangerous. Or at least somewhat silly as many people who believed the earth was flat and the humors of the body had to be in balance for health could attest to if they were still alive.

I'm stating my theory. How is that dangerous? If you wish to prevent all scientists from stating theories unless they are first proven, then I guess we might as well scrap all scientific publications while we're at it. And really... is my theory so far-fetched? The human brain is a vastly complex parallel biological computer. Regular computers today are mostly linear and comparitively simple. If we manage to get computers to process massively parallel information and an order of magnitude more complex, I'm extrapolating that there's no reason why a computer, under those circumstances, couldn't achieve the processing power of the brain. What is perceived as intuition to us is, in all likelihood, simply an emergent result of highly complex parallel computations. Sure, we don't understand the mechanism. But let's not quell the effort of trying, shall we?

On a positive note, you're not alone. You're in the same club as those who initially ridiculed Einstein's theory of relativity... "Oh, he's just putting forth a specific theory as if it is fact. That's dangerous! Might as well call the earth flat!" Sure, some theories turn out to be false; others, true. But what's the point in insulting a man for trying?

Re:Interesting computer Chess?

[dvdeug]

Oh really, how do you know what is the cause of human intuition? Last time I checked it was an unexplained phenomena.

Not really. Coming up with supernatural explanations for human intuition in chess is like the people who are taught how to "bend spoons with their mind", but still need to help a little with thier hands. It's simply unnecessary. Even in real life, coming up with
supernatural answers for human intuition is copletely unneeded in most cases. We could always assume that we reach into another dimension to add numbers, but there are better simpler answers.

By the way: quantum physics is a real thing, it's not gooblygook. Why don't you read up on it a little before making something up and blaming it on quantum physics.

Re:Interesting computer Chess?

[Mishkin]

...where ever, when he lost, and yell out obscure threats

Damn you! If you ever beat me again I stick a Scold's Bridle on you!



(sorry couldn't resist)

Re:Interesting computer Chess?

[Anonymous Coward]

That's true. I can only count to 32768.

Do you have 15 fingers too!?

Re:Interesting computer Chess?

[Anonymous Coward]

Computers dont have finite limits like a person.

Digital computers do have finite limits, that's why they are known as finite machines.

The thing is though, that their limits nowdays tend to be astronomical and they can reach them at incredible speed.

It's kind of ironic...

[Anonymous Coward]

...that the Computer Chess championship is in Libya, while Qaddafi banned Israeli players from the FIDE championship. Actually, Kirsan Ilyumzhinov, president of FIDE and the Russian state of Kalmykia, previously tried to have the FIDE championship in Baghdad before he was forced not to by the first Gulf War.

Re:It's kind of ironic...

[Anonymous Coward]

Ummm... the computer championships are in Israel... The _human_ championships are in Libya...

Re:It's kind of ironic...

[Anonymous Coward]

Who modded this Informative? The computers are playing in Israel.

Re:It's kind of ironic...

[oob]

while Qaddafi banned Israeli players from the FIDE championship

It's even more ironic that you bitch about this while 4 million Palestinians are imprisoned by the brutal, unconscionable and illegal Zionist occupation of their country.

I don't invite war criminals to my home, neither does Qaddafi. Good for him.

Re:It's kind of ironic...

[Anonymous Coward]

No Qaddafi just kills innocent civilians because of his racial prejudice.

Re:It's kind of ironic...

[Kiryat Malachi]

Qaddafi practically is a war criminal. Let's not forget the Libyan hit squads sent against Libyan dissidents in the 80s, the Black September Massacre he is believed to have financed, his actions in the Lockerbie case, the 86 Berlin bombing he is reported to have controlled, and the Fletcher murder.

As to your other claims, about Palestinians... it's an argument I won't get into, but you're wrong. Sorry.

Shades of WOPR

[erucsbo]

Computers play their own championship?
Ultimately this will have to result in stalemate after stalemate won't it?
Kinda like WOPR in 'Wargames' playing tic-tac-toe with itself.

Re:Shades of WOPR

[ookabooka]

no, definately not, having identical engines on the same computer can result in completely different games. It comes down to processing power, hash table size, and the actual structure of the engine. In this tournament they put all the engines on identical comptuers, so its the architecture of the engine that is tested.

Re:Shades of WOPR

[Short Circuit]

If they're able to compile for the target architecture, neat. Imagine the escrow for the source code to that kind of gaming logic.

Re:Shades of WOPR

[terber]

> In this tournament they put all
> the engines on identical comptuers

This is not correct.

In fact, every contestant is allowed to use his own hardware.

Contestants which are not opting for their own hardware, get an Pentium-4 with 2,8 GHz from the Bar-Ilan-University.

Crafty (freeware) brings its own quad-Opteron machine with 2,4 GHz!

Two other contestants are playing on quad-Opterons, too. Fritz and Shredder are playing with 2,2 GHz resp. 2,0 GHz. Both machines are from the sponsor Transtec.

Re:Shades of WOPR

[Anonymous Coward]

Or it might result in win after win if one happens on the 'perfect' game.

Re:Shades of WOPR

[Portfolio]

> Crafty (freeware) brings its own quad-Opteron machine with 2,4 GHz!

Actually, it's a quad 850MHz.

Junior (the favorite) is on a prototype HP quad machine. Amir Ban is bound not to release any other hardware details.

Re:Shades of WOPR

[Skuto]

It's a quad OPTERON 850, not an 850Mhz machine.

The Opteron 850 runs at 2.4Ghz.

Re:Shades of WOPR

[Ayaress]

Assuming the computers are all the same power, there are still a few things that can effect it. As far as I know, at least some (if not most/all) chess algorithms aren't entirely fixed. Playing the same moves against the same program won't always trigger it to make the same counter moves.

Also, slightly different programs on each computer will make for different responses from each system. If all the computers were running the same chess program, then you might expect to see relatively static results (either infinite draws or very simmilar wins over and over).

Tic tac toe is a pretty simple game, and if both sides follow ideal rules, a win is impossible. The only real way to win is to hope you can trick your opponent into letting yo uset up a three-corner situation (take top-left and bottom-right first - if your opponent doesn't catch you at it, you can take top-right or bottom-left and have two possible wins, only one of which can be blocked in one move), but against somebody whos even slightly familiar with the game(not saying much), your third move will be needed to block them and not set up for the win.

For those who think chess needs a little variety

[Anonymous Coward]

www.chessvariants.com is a fun site.

(I'd make it a clicky but I don't know how to do so)

USA?

[magarity]

All five of the players listed as from USA have blatantly obvious Russian (and one Japanese) names. Looks like no native players in this one (again). Alas, purely intellectual pursuits are frowned upon in these here parts.

Re:USA?

[Ratfactor]

Just out of curiosity, what would you consider to be a genuine USAian last name? Washington, Jefferson, or Franklin?

Re:USA?

[jfengel]

That seems a little harsh. Americans are active pursuers of all sorts of intellectual activity, from art to literature to science to technology. Many open-source developers are Americans. Americans have a boatload of Nobel prizes.

So chess doesn't happen to be the obsession here that it is in Russia. They're not so good at soccer, either. BFD. It doesn't mean that they do nothing but watch reality TV.

Re:USA?

[F13]

yeah!

Where is John Redcorn.

Re:USA?

[Anonymous Coward]

"No native players"? You were hoping for Native American players?

In case you haven't heard, the USA is (almost entirely) a nation of IMMIGRANTS.

Glad the Russian-American & Japanese-American players have found a better life here.

Re:USA?

[tanguyr]

" Just out of curiosity, what would you consider to be a genuine USAian last name? Washington, Jefferson, or Franklin?"
Dances With Wolves?

Re:USA?

[zokrath]

Indeed; few other first world countries have such a stigma against using your mind for more noble functions. Having extended experience with computers and programming brings a label of 'pathetic nerds', while useless sports knowledge and statistics is often considered par for the course.

Luckily I was in the gifted program in Junior High and High School, where all of the teachers were dedicated Masters holders, and a much smaller percentage of students had an active criminal record.

No doubt partially responsible is the fact that intellectual careers, even in the upper echelons, are generally long term group based projects with little individual recognition for the scientists. It is the company or academic insitution that gets most of the credit.

Sure, there is the Nobel prize, but it is playing cath up and is so behind the times that when it awards someone for a contribution to everyday life, that contribution has most likely already become commonplace, so most people, if they even hear about it, think "Oh, he invented the microwave, that's nice, but those have been around forever..."

Meanwhile sports teams with annual grosses exceding countries demand that hundred of millions be spent on a new stadium or else they will abandon the city for green pastures.

I say that an additional 10% tax needs to be placed on all sports salaries greater than one million a year, with all of that tax revenue going towards funding for research and dedvelopment of things that will actually benefit mankind. Hell, put another 10% on salaries over two million and put that towards societal reforms.

Capitalism is great, but there is no reasonf or it to be unbridled like it is today. such a tiny percentage of people controlling a vasty majoority of wealth is pointless and harms the long term potential of our country.

Well that certainly go off track a bit.

Re:USA?

[Vellmont]

More like the US didn't (and doesn't) have state sponsored chess schools like the former Soviet Union did. If chess is such a great "purely intellectual pursuit", then why aren't all the great chess masters great geniuses and create wonderful things outside of chess? Chess may be a fun game and all, but all this connection between chess and "being intellectual" is just nonsense.

Re:USA?

[Anonymous Coward]

Idiot. The only natives are Native Americans. You are an immigrant.

Re: USA?

[Jes-ka]

What's in a name?

If you're from America, you're an American. By native do you mean Native American??

Re:USA?

[humblecoder]

All five of the players listed as from USA have blatantly obvious Russian (and one Japanese) names. Looks like no native players in this one (again). Alas, purely intellectual pursuits are frowned upon in these here parts.


No offense but this is one of the STUPIDEST comments I've ever read on slashdot. Actually, I take that back... I do mean to offend you.

Who's to say that the American players aren't fifth generation Americans? Just because they have an "ethnic" surname doesn't mean a thing? Surely you don't expect people to change their names to "Smith" or "Jones" upon obtaining American citizenship, do you? I mean, really!

Obvious "intellectual pursuits" like logic and rational thought are frowned upon in whatever parts you hail from, as well! If you are an American then maybe you have just proven your own argument, in which case I apologize.

Re:USA?

[NarrMaster]

I was with you, until you said "USAian". Sorry, its just stupid.

Re:USA?

[mondoterrifico]

That is a retarded argument. Why doesn't a physicist create wonderful things outside of physics.
Physics may be a wonderful pursuit, but all this connection between physics and "being intellectual" is just nonsense.

Re:USA?

[Anonymous Coward]

Indeed; few other first world countries have such a stigma against using your mind for more noble functions. Having extended experience with computers and programming brings a label of 'pathetic nerds', while useless sports knowledge and statistics is often considered par for the course.

There's an interesting
article written by Paul Graham about this problem.

Re:USA?

[HuguesT]

Chess grandmasters create wonderful chess, just the same as a great painter creates wonderful art. There is no fundamental difference.

Do we ask of great actors, writers, painters and musicians, or for that matter tennis people or rock stars to be good at anything else outside of their own area of competency?

Because you may not appreciate it doesn't mean it's worthless.

Re:USA?

[pommiekiwifruit]

Surely you don't expect people to change their names to "Smith" or "Jones" upon obtaining American citizenship, do you?

I thought that was standard practice there, for the 19th century anyway? An awful lot of US surnames are Germans pretending to be English anyway (Muller->Miller etc.), probably more than actual English surnames.

Re:USA?

[Anonymous Coward]

Wah wah wah, sports gets all the recognition. Don't forget movies! And TV! Those should probably also all have extra taxes to go toward funding Zokrath's R&D Fund to Benefit Mankind. Because then the money wouldn't go to waste. And a tiny percentage of the population wouldn't be controlling where that money is spent. Right?

Don't worry, you'll get individual recognition for your intellectual feat of creating the whiniest suckybaby post on /. for the month of July.

"Sports teams with annual grosses exceeding countries ..." come on. Name one.

Re:USA?

[Anonymous Coward]

Nakamura, the one with the Japanese name, is definitely American. The Russians, however, are mostly recent immigrants. Most of them have come to the US because, as lower-level professionals, there are many more opportunities to make money coaching/playing in open championships in the US than in Russia.

Re:USA?

[Ayaress]

And I have a baltantly Italian last name. Guess I'm not from around here? What's you're name? I bet its blatantly forign too.

Re:USA?

[Ayaress]

I won't point out the fact that you're wrong, and that there are chess masters who write, sing, compose, paint, or belch the Star Spangled Banner (which is intellectual enough to have a state championship here in Micghigan this month). Why don't great painters do anything outside of painting? Why don't great writers do anything outside of writing? Why don't great *insert sport here* players do anything great outside of shoe commercials? The artist does his own art, not all the other arts.

Re:USA?

[magarity]

No offense but this is one of the STUPIDEST comments I've ever read on slashdot. Actually, I take that back... I do mean to offend you. Who's to say that the American players aren't fifth generation Americans? Just because they have an "ethnic" surname

To you and all the other responders and moderators who've blasted me over this comment: I happen to know a LOT of immigrant families to the USA as my wife is from another country and we hang out with a lot of friends from the 'old country', as it were. This is how I just happen to know that the immigrant parents, in a flush of enthusiasm for their new home, overwhelmingly give traditional Anglo type types for their children so you get constructions like "John Wang" and "Richard Kasimdzhanov". What you do NOT find very often are given names such as "Hikaru" and "Sergey", which are samples of the GIVEN NAMES of the players listed from USA. Since I never mentioned SURNAMES your flippant comment about "Smith" and "Jones" is not relevant. I assumed that everyone would know I meant given names while I should have assumed hotheaded people who don't know any immigrant families would blast me as a troll posting "stupidest" comments.

Re:A chess posting on slashdot ?

[theJerk242]

How will the open source chess engine Crafty do against the proprietary closed engines?

Well...they did mention open source in the article.

Go is tougher

[Anonymous Coward]

Obligatory GO reference.
Put your own AI search tree bullshit here.

Re:Go is tougher

[IWantMyNickBack]

Of course it's tougher, mathematically there are around 10^43 to 10^50 Chess positions while there are 2.1×10^170 Go positions. In fact, humans usually have to take out large portions of their life to master the thing, and computers absolutly stink at it (although a GA or a neural net/GA combo could fix that)

Re:Go is tougher

[stephentyrone]

what makes you think that GA / neural nets would contribute anything at all to improving computer's go abilities? both ideas have been around for decades - if they were actually going to get us anywhere on the problem of playing go, it would have long since happened. go has too much global structure to be easily succeptible to GA or neural net approaches with current (or forseeable) hardware.

Re:Go is tougher

[Anonymous Coward]

Someone needs to build a better model of "how to play GO". Then the algorithms might have a chance. But it you look at Go from a purily tree point of view, it simply doesn't work. But human Go players don't think about Go in this manor. They generally think of the Go board in classifications (near a piece, open board, gray in between). When a piece is placed by itself, it generally is placed rather randomly (for a game anyway). This is just part of the difficult problem of writing a Go playing program. It is very very hard to be sure, but the model and not the algorithms are what need to be improved.

Re:Go is tougher

[chanceH]

I think current for current state of the art, Neural Nets are the best players. However, training time is measured in years, and the hardware (while general purpose) is a nightmare to maintain. (Its a special purpose neural hardware, with asynchrounous nodes.

Re:Go is tougher

[Portfolio]

> what makes you think that GA / neural nets would contribute anything at all to improving computer's go abilities?

Err... the fact that NeuroGo is one of the favorites in the World Computer 9x9 Go Championships? (which is being held now at the same computer game conference).

There is a lot of room for improvement (especially at the standard 19x19 board size), but neural nets have certainly proven themselves at Go, a game requiring lots of pattern recognition (the neural net's forte).

Why do they have to travel there?

[Osrin]

Can't they play over the internet? - it's a good job I don't have to transport my computer to CA to search for something using Google.

Re:Why do they have to travel there?

[nacturation]

Can't they play over the internet?

Can't I participate in your $1,000,000 trivia contest over the internet?

Most interesting part of computer championships

[civilengineer]

Computers do not settle for draws like humans do in face of complications. This will guarentee some extremely interesting endings.

Also, since Ken Thompson is making great progress on building endgame databases, the games might be all played to end.

Re:Most interesting part of computer championships

[Short Circuit]

Computers do not settle for draws like humans do in face of complications.

Sure they can, if they're directed to minimize losses.

Personally, I'm curious why game theory software doesn't have the kinds of export restrictions that encryption software or computing hardware does.

Re:Most interesting part of computer championships

[Anonymous Coward]

I would say the odds of globalthermonuclear chess arent as high as the odds, of someone using encryption to cover up their potentially illegal actions (can of worms opened).

I suspect the engines would be very chess specific in their logic, and to have the knowledge to change them, means having the knowledge to write your own.

Re:Most interesting part of computer championships

[Short Circuit]

Game theory goes beyond just chess. It's the whole concept of reacting to changing stimuli to reach a specific goal...

(IANAE)

Re:Most interesting part of computer championships

[foo fighter]

What are the best (insightful, informative, funny) sites out there for the latest happenings in the chess world?

Re:Most interesting part of computer championships

[dvdeug]

Personally, I'm curious why game theory software doesn't have the kinds of export restrictions that encryption software or computing hardware does.

Computer chess games have nothing on the fields of Wellington. This has all the bad effects of export restrictions, including the fact that the world outside the US has many good programmers and that it's impossible to stop software from being exported from the US, without being something you can point to and say "This is something only communist spys would use."

Re:Most interesting part of computer championships

[hunterx11]

Actually in the matches of Deep Fritz against Kramnik and Deep Junior against Kasparov, the computers were unable to offer draws or entertain draw offers. Instead the technician would decide based on the game and the computer's current evaluation of the position. I don't know if this is a computer shortcoming or was just part of the match rules, though.

First "GO" Post

[Saeger]

Forget chess. To truly push the limits of computing and AI we should instead be trying to improve on the ancient game of "Go". No computer can even come close to besting a human here yet.

--

Re:First "GO" Post

[LiquidCoooled]

Having never played "Go", I can definately concede that a computer could beat me :)

Infact, a very poorly written script running on windows 3.1 could most likely beat me at the game!!

Re:First "GO" Post

[Anonymous Coward]

I'm not doubting the difficulty in playing Go, but I'm still yet to see a CPU win a game of I-Spy against me...

Re:First "GO" Post

[Dan Ost]

Give it time. Once computers are consistently beating the greatest human
players, the same clever people who worked so hard on building computer
chess players will find new problems to spend their time on. Go will certainly
be on of them.

Re:First "GO" Post

[wviperw]

Uh oh, here come the "Go" zealots.

And what happens when computers master Go? Then there will be those that say "To truly push the limits of computing and AI the computer should master the art of interpretive dance," or whatever you want to place in there. The truth is, people will ALWAYS try to come up with areas in which they are better than the computer (emotion, art, feelings, abstract thought, etc).

I believe Turing predicted something similar to this around 60 years ago.

Re:First "GO" Post

[Qrlx]

Having never played "Go", I can definately concede that a computer could beat me :)

That's exactly what I was thinking.

However I was surprised to learn that the top human checkers players can easily trounce the computer. I would guess that checkers would be orders-of-magnitude a "simpler" problem than chess. Maybe it's that chess gets all the buzz, since it's considered to be the ultimate thinking-man's game.

(Not trying to dis Go, maybe I should say chess is the Western Civilization's ultimate thinking-man's game.)

Re:First "GO" Post

[sokoban]

The real difference between go and chess though is that go can't be solved by computation (very well at least) and chess can. I think to really get a Go program to work well it will really just have to play and lose a lot of games. I am by no means a great go player, but I know that I have learned a hell of a lot more just playing Go and talking about it than I have from any book. Chess skill can be taught in strategy and tactics, but go can only really be learned from experience. Computing in Go is secondary to the learning process, intuition is key.

Re:First "GO" Post

[Russellkhan]

I am very glad that you've posted about Go. It's a beautiful game and more people in the western world should know about it.

That said, I've never understood why so many Go players treat Go and Chess as an either/or, one vs the other type of choice. I play both Go and chess, and while I can easily see that there remain much greater challenges in computerized Go playing programs than in Chess (For those who don't know, the best Go playing programs play at roughly the level of an intermediate amateur human player), this fact does not take away from the fact that chess is still an interesting game, both to play and, I'm sure, to develop better computer opponents for. Nor does the work being done on chess take anything away from Go development.

While I'm here, I may as well post a Go wiki link in a wiki that's all about Go. I realize you (parent poster) probably know about it, but just in case anyone is interested in learning more about Go, I figure it's a nice starting point.

Re:First "GO" Post

[Saeger]

Give it time.

It'll be about 25 years before our computers can match the parallelism and capacity -- but not speed, since transistors've always been faster than a synapse -- of the human brain's hundred billion+ neurons. But in 25 years there will be far more important things happening than Go :)

--

Re:First "GO" Post

[tyroney]

interpretive dance is a little more subjective than a two player game based on capturing territory. Next time come up with a better example, like baseball.

I'll admit that I like go, but I tend to be about as good as a computer at it, so I don't enjoy playing against anyone. Hence I'd just as soon they continue persuing chess so that I have a go opponent I can enjoy playing.

Re:First "GO" Post

[ChuckleBug]

I am very glad that you've posted about Go. It's a beautiful game and more people in the western world should know about it.

Yeah, we never heard of it until it was described here, for the very first time, on slashdot.

Go IS a great game. Still, we chess lovers get fed up with the Go zealots who, whenever chess is mentioned here, start with the "Chess sucks, Go rules!" stuff. I completely agree with you about the "either/or" mentality, but it's pervasive. You much choose your side, and fight all others to the death, you see.

Re:First "GO" Post

[neuroslime]

I think to really get a Go program to work well it will really just have to play and lose a lot of games.

Somehow I don't think losing a lot is how you get a program to get better. There has to actually be some algorithms for learning and generalizing for it to improve with play. Of course, the learning code is A LOT harder to design and implement than the playing and losing code.

Re:First "GO" Post

[ninja0]

Give it time. Once computers are consistently beating the greatest human players, the same clever people who worked so hard on building computer chess players will find new problems to spend their time on. Go will certainly be on of them.

I sort of doubt computers will play Go at the sort of level humans do for many years to come. Go plays particularly well to human's strengths. A large fraction of our brain is devoted to processing visual information. In Go (unlike in chess), this pattern recognition gives us a huge advantage over computers, which lack the hardware to recognize patterns in the same way.

Some major progress in AI as well as computer will most likely be required before computers can stand up to Go masters in the same way they currently match chess masters.

Re:First "GO" Post

[Saeger]

I'm no Go zealot, and in fact I rarely play either game. I was just posting the obligitory Go post (but I see I got beat by 3mins). About the only thing I could be called zeolous about is the rate of technological change we'll be seeing in the next few decades, and how many people are ignorant of this nearing Singularity.

--

Re:First "GO" Post

[JamesKPolk]

Computers haven't "mastered" chess in the sense that heuristics or rules have improved their play enough. The key ingredient is the brute-force search.

That's why IBM built Deep Blue, to show off their hardware doing the searches.

The 19x19 board of a standard go game combined with the fact that a play on most points is legal at any time basically rule out such searches.

Re:First "GO" Post

[JamesKPolk]

So why do you read slashdot if you can't handle "zealotry"?

Re:First "GO" Post

[nacturation]

However I was surprised to learn that the top human checkers players can easily trounce the computer. I would guess that checkers would be orders-of-magnitude a "simpler" problem than chess. Maybe it's that chess gets all the buzz, since it's considered to be the ultimate thinking-man's game.

Checkers has all but been solved. See this Mathworld article for more info. Basically, there's an estimated 10^12 to 10^18 different positions in a game, with a possibility for only having to solve 10^9 of them. With sufficient memory (Beowulf cluster, anyone?) checkers can be completely solved such that you can guarantee either a win or at worst a draw for the first person to move.

Re:First "GO" Post

[Anonymous Coward]

The ancient game of Go? Thats the one where you collect $200 right?

Re:First "GO" Post

[harlows_monkeys]

Go will certainly be on of them

Go is much harder than Chess for computers because brute force doesn't work even remotely as well. The branching factor is much higher (until the endgame, there are 100-360 possibilities per ply, compared to a dozen or two for chess), and the depth you might need to search is much deeper (consider a ladder starting on one side of the board whose outcome depends on the stones on the other side, 30 ply down the tree, and determines the life or death of a large group).

Note that I'm not saying the good Chess programs are pure brute force. They are basically a combination of brute force and good AI working together, but the brute force is a critical component of their success. With Go, the AI has to pull all the weight, and it isn't nearly good enough.

Re:First "GO" Post

[Anonymous Coward]

It seems to me that what is being demonstrated is:
Brute Force [computer] + Rules [developer] = Brute Force [Chess Champion] + Rules [Chess Champion]
Someday there will be no need for rules, the computer will win everytime, and thus mastered the game as has been done for Tic-Tac-Toe or checkers

Re:First "GO" Post

[Veramocor]

Or guarantee a loss for the first person to move. Since it still hasnt been solved, although highly unlikely(I would put millions to one odds against) and counterintuitive going second may be the more favorable position! It just seems that going first is the best postion based on current experiences and limits of calculations.

Re:First "GO" Post

[Anml4ixoye]

Yeah, we never heard of it until it was described here, for the very first time, on slashdot.

Actually, I hadn't heard of it ever before before it was described here, in this thread, on slashdot.

Re:First "GO" Post

[Quirk]

I've always thought Go Maku presented an easier introduction to the complexities of Go. Strangely Google provided only one precursory listing for Go Maku. Are Go players distainful of Go Maku?

Re:First "GO" Post

[Anonymous Coward]

One of the most significant problems of go is that a sufficiently suitable evaluation function has not yet been found.

With checkers and chess, it is significantly easier to look at the board, and determine which player is currently in the lead. IBM's earlier attempts at programming chess AI even relied on specialized hardware for board evaluation.

Once a suitable evaluation function exists, it becomes orders of magnitudes easier to optimise for it.

One of the largest problems with the game is that it can be broken down into numerous subgames that can be easily computed, but the outcome of the subgames is almost entirely independent from the outcome of the final game.

Re:First "GO" Post

[s1234d]

Actually GnuGo 3.4 beats me every time on level 1. I assume you mean a *good* human player.

Re:First "GO" Post

[Russellkhan]

I have played Go Moku (that's the spelling I'm familiar with, but I'm sure there's more than one way to spell it), and while it shares board and pieces with Go, that's about where the similarities end. Go Moku is much closer related to Tic-Tac-Toe than to Go in terms of how it's played and its objective (not in origin).

For those that aren't familiar with it, Go Moku is a game played on a Go board, with the objective of getting 5 stones (pieces) in a row. Pente is a variant of Go Moku.

The best way to simplify Go (IMO, of course) into easier to digest bits is to play it on a smaller board. The standard board size is 19x19, but the game also commonly played are 13x13 and 9x9. Much of the strategy of the game only happens on the larger board sizes, smaller boards emphasize tactics more, which is easier for beginners to grasp.

Re:First "GO" Post

[WalksOnDirt]

Ever try a 4x4 board?

It doesn't really give you a feel for the larger game, but it's amazing how hard it is to compleatly analyse even so small a board.

Re:First "GO" Post

[Russellkhan]

I've played 3x3, and have tried 6x6, but never 4x4. The boards smaller than 9x9 seem almost novelty games to me - Yes it's true there is still some depth to it, but it's just not a very interesting game after the first time or two.

Re:First "GO" Post

[Kiryat Malachi]

I don't know.

Every computer Go program I've played against has kicked my ass, and I'm certainly a human.

Re:First "GO" Post

[dallaylaen]

AI is also very-very bad at bughouse

Re:First "GO" Post

[Anonymous Coward]

>Uh oh, here come the "Go" zealots.

Yer damn right!

>And what happens when computers master Go?

Um. Nobody knows. Computers haven't mastered go.

>I believe Turing predicted something similar to this around 60 years ago.

What - so there's no merit to the discussion that a computer task may be better or worse for evaluating AI? Man, ever since computers could count faster than humans the meat entities have just been whiney bitches!

Re:First "GO" Post

[StressedEd]

...Go zealots who, whenever chess is mentioned here, start with the "Chess sucks, Go rules!" stuff.

Do you mean like Linux zealots? Just because they're zealots doesn't mean they are wrong. :-)

I play chess and have started learning to play Go.

It is a truly sublime game. It takes literally no more than 3 minutes to learn the rules (all five of them) and you are up and playing straight away. But it is madenningy hard to be any good at! I think getting computer Go to the same level of play as computer chess will teach people about AI. I can easilly see how a typical chess algorithm would work but wouldn't have a clue with Go, the "merit function" is just too hard!

Re:First "GO" Post

[VendingMenace]

Not to be overly argumentative, but i belive that GnuGO is rated in the high single digit Kyus -- definately in what is considered to be intermediate strength in GO.

Re:First "GO" Post

[Anonymous Coward]

Mod Parent Up!

I'm an infrequent slashdot visitor, but an avid go player and tinkerer with computer go! The most surprising thing about computer Go is that when it's played on a 9x9 grid (Go can be played on several standard board sizes, of which 9x9 is the smallest), the game has a branching factor more or less equivalent to chess, but the strength of Go programs does not significantly improve! Even on a 9x9 grid, the static evaluation function problem is so huge that even the best Go programs at small-scale infighting, which is prevalent on the small board, still stall out at the level of a weak club player.

Re:First "GO" Post

[Abcd1234]

Err, more like intermediate amateur strength, and barely that. GNU Go is, supposedly, in the 8 kyu range (not quite the high kyus, but getting there) on KGS, but any decent amateur player could trounce GNU Go. Heck, Martin Mueller, our local honorary 1p, managed to beat Many Faces of Go (one of the strongest Go AIs out there) with a 29 stone handicap!

I hate computerized chess, the server cheats

[Anonymous Coward]

It always ends up AWP'ing my queen! I suspect an aimbot.

Re:Yay, first post

[grandbonheur]

brute forcing millions of... positions Heh. Sounds a bit rude, duzznit.

computer vs human players

[spacerodent]

One of the main things that will never really be present in computer players is human reaction. A human may bluff, or try to call a bluff, or deliberatly do somthing retarded to cause you to underestimate them. Until the devlopment of true Ai there will never really be an "exciting" computer match. Currently computers simply calculate the most "efficient" move and take it. Thats like listening to a recording of music instead of playing it yourself.

Re:computer vs human players

[lpp]

But why would a "true AI" be susceptible to bluffs? Would a veneer of humanity laid atop an overgrown calculator somehow limit its ability to perform said calculations?

Or, to put it differently, why would I include sufficient ego in an AI to cause it to be able to underestimate an opponent?

No, I for one welcome our new AI overlords.

Re:computer vs human players

[romango]

There's no reason a computer couldn't be programmed to make trick moves occasionally.

Re:computer vs human players

[canadian_widget]

I used to play chess a lot when I was younger competitvely, and the rise of the computer chess masters was an area of great interest.

What you don't take into account is that the computers can be programmed to "bluff". Although they ARE using brute force calculations to figure out what the best move may be, they do have databases of moves. If they have a "bluff" in their database, and the other computer opponent doesn't see it... a game could be one.

I agree, it's not AS interesting as watching people play. But, I never liked watching people play chess anyway. It's boring! I liked beating people ;)

Re:computer vs human players

[exp(pi*sqrt(163))]

As computers get better and better at chess humans will go on finding more and more dumb aspects of the game that they claim is what makes it interesting. If programmers wanted a machine to bluff they would - but why bother, bluffing against a grandmaster is generally a very bad strategy.

Re:computer vs human players

[medelliadegray]

i have no links, but i seem to recall after one of the deep blue matches a few years back, the human commented to the effect that some of the plays the computer made--it made the player wonder if there was truely an intelligence playing against him, because the route (trap) the computer setup became so utterly unique to what this player had experienced before in any computer or human chess game.

Re:computer vs human players

[fw3]

At this level, the human players are decidedly *not* doing anything to intentionally cause the opponent to underestmate thier skill.

In chess as in most competition the first step to *losing* is assuming your opponent will make a mistake (including the mistake of not recognizing *your* mistakes).

'True Ai' exists. AI methods have been tried in chess and so far they simply do not work as well as brute force evaluation based on material gain.

That may change someday however the progress / work to date is no indicator that it's going to be any time soon.

On the other hand games between humans and computers have been considered 'interesting' by master-level players so if that's your criteria for 'True Ai' ...

Re:computer vs human players

[Stray7Xi]

Bluffs aren't useful when being underestimated, but when being overestimated.

If I throw my bishop before your pawn, and you KNOW you're better then me, and you don't see any complications, you'll take it.

But if you KNOW I'm better then you, it causes hesitation. The bluff only works because that presumption I know something you don't.

I assume what he means by "true AI" is that it'd be able to self-learn more general knowledge then chess (or other things that succumb to brute force algorithm) and is based on pattern forming (similar to life). This would mean the computer would be capable of recognizing the opponent is much better after playing many games (and recommend movies to him based on his game... well maybe not). Bluffs like this may cause the computer to misclassify moves as more critical then others (afterall if my opponent is so good, why would he sacrifice that piece unless he had a plan!) and waste time on it. However, the greater part of the bluff on humans is at the emotional level and this AI should be immune to it.

Disclaimer: IANACM (chessmaster) IANAAIP (AI programmer)

Re:computer vs human players

[lpp]

However, the greater part of the bluff on humans is at the emotional level and this AI should be immune to it.

I guess that's what I was responding to. Most folks have a tendency to misrepresent AI as being somehow more human like, an error brought on, I would think, by the fact that humans are the most (only?) intelligent species we are familiar with. An AI doesn't necessarily inherit our foibles (unless we screw up and add them in).

ChessGML animated with SVG

[wombatmobile]

.

Max Froumentin of W3C shows how to animate chess games by converting ChessGML to SVG with XSLT.

Re:ChessGML animated with SVG

[Ninja Programmer]

In the scholars mate demo they have 2. ... Kc6 which is a little bit illegal

Re:A chess posting on slashdot ?

[hunterx11]

Acutally, a closed-source engine has been banned over accusations of copying. I guess with open-source there's no worry :)

The Russians

[Ratfactor]

Can Robot Jox be far off?

The Russians pretty much dominate human chess. Now that things have shifted to machine chess, robots with chainsaws in the crotch are an obvious next move.

Re:The Russians

[winkydink]

And they'll hang around outside until the chess tournament ends and kick the shit out of the chess robots and take their lunch money.

Re:The Russians

[Anonymous Coward]

In Soviet Russia, chess plays you!

C'mon, you all wanted to say it.

Libya is our friend, Eurasia has always...

[fantomas]

Hey, Libya is our friend, Libya has always been our friend. Oh, I meant Eurasia. Or did I mean Eastasia?

Huh, politics, I just leave that to politicians, they tell the truth, they know what's best ;-)

Breaking News....

[k4_pacific]

One of the competing teams announced today that they had secretly patented the interface by which the chess programs compete with one another. As such, the other teams forfeited under threat of lawsuit and the team that held the patent won.

Re:Breaking News....

[TastyWords]

"One of the competing teams"? Tell the truth. It was Microsoft, announcing their patent of the day.

No GNUChess ?

[Anonymous Coward]

Why no GNU Chess ?

Re:No GNUChess ?

[Anonymous Coward]

GNUChess is not strong enough.

It has this going for it though, "This release features simplified chess code and modern data structures, which make it more pedantically accessible, easier to modify, and more understandable for skilled chess experts who are not necessarily programmers."

overly critical /. again

[Anonymous Coward]

>How will the open source chess engine Crafty do against the proprietary closed engines?

This is about as ridiculous as saying:

How will the communist Chinese do against the capitalist Americans?

And you sir,

[Anonymous Coward]

Are a moron. Perhaps my great great grandparents were immigrants, but I'm American.

Some results

[ninja0]

Some results are at Chessbase.

Crafty managed to draw Shredder, one of the big-name computer programs, in the first round. That makes it tied with a bunch of other programs in the middle of the pack.

Personally, I'm always excited to hear about the progress that has been made by chess engines. Nowadays, the top programs can compete with all of the top humans, without requiring a supercomputer.

Re:you lose - here is the firstest GO post

[Saeger]

I concede technical defeat. But at least I won the popular vote!

--

Crafty prediction

[Anonymous Coward]

Crafty is strong, but it does not have the staff of the other engines. They have paid staffs that work to strengthen and bugtest their engines. Crafty has some people that do this for free of course, but they don't spend nearly as much time as paid full-time staffs.

Re:Crafty prediction

[Portfolio]

He doesn't know Robert Hyatt very well, does he? Dr. Hyatt may be the most experienced chess programmer still active today. He has been programming world champion programs (Cray Blitz) since before many of the posters here were born!

"Staff" has little to do with engine strength. (That said, creating a good opening book is almost as much work as the engine itself. A full time book maintainer can give you the winning edge in a tournament like this.)

Re:Crafty prediction

[Anonymous Coward]

He doesn't know Robert Hyatt very well, does he?

;-)
I have been on the Crafty mailing list for some time now actually.* Dr. Hyatt has extensive experience and has written many papers on chess programming, but he still has to split his time between Crafty and the rest of his work. Crafty has started out the tourney with 2 draws, which is not bad at all, but I still don't expect it to get in the top 3.

I still say that staff has something to do with engine strength. Other engines can pay GMs to test out their product. Well, at least Crafty has Roman. ;-)

I love Crafty so don't take any of this the wrong way.

*If you like Crafty or have trouble compiling Crafty check out the list. Dr. Hyatt is very helpful and so are others on the list.

Re:Crafty prediction

[WalksOnDirt]

Crafty won a major chess tournament at the beginning of 2004, CCT-6.

Crafty was running on the best computer used in the tournament, a quad Opteron, which must have helped. Crafty is close enough to the best programs that over eleven rounds it has an outside chance of winning.

still human against human

[sick_soul]

>Will the computers play more interesting chess than their human counterparts?"

#define PACMAN "ProgrammerAlgoristChessmasterMAN"

I think it becomes a game of PACMAN against other PACMAN, so I always see this as human vs human.

The games are interesting, not because they are "played" by the machines, but because they are indirectly played by the programmers.

Re:still human against human

[Tojosan]

while your observation is mostly true today, ie its really programming team against programming team, I have a feeling that the next generation of programs will take a step up from that.

I've noticed on the chess programs I've played against that the same program will pick the same move in the same position over and over again, even with good alternatives. What I invision in the future is that the programs won't just brute force algorithm themselves to the mate, but actually apply rules of fuzzy logic (shoot me someone for the buzzy words ehre) and pattern recognition.

I don't write chess programs, and to be honest my programming problems aren't near that complex, but we are often being tasked with not just simple data mapping, or totalling or scanning, but being asked to make the programs able to recognize that something is wrong via patterns. And of course there is the age old problem of guessing what the user wants to do and balancing it against what you know they should be doing. This sort of thing is just too hard to program directly but a learning system, one that could evolve its own heuristics over an extremely large sample set, would have a much better success rate. This same thing applies to chess, assuming some computer doesnt' figure out all possible positions and solutions before then.
Pattern recognition and distilled heurisitics combined with brute force is going to take us beyond programmer vs programmer because the programmers won't be able to easily predict how their system will behave after its played thousands of games. That is unless they disect and dump each part of the system.

Of course, this is all just buzzwords and pipedreams since peeps have been talking about this for ages! :)

Final note, and sadly no link, but there is more than one team working on alternates and enhancements to blind brute force.

Anyway, rant off, mod me down if you want for too many buzzy words!!!

tripoli

[minus_273]

hmm be careful not to give the material poweing thos super computers to the Lybians make no deals either or doc might get killed.

Junior has an impressive track record...

[drclaw007]

... keeping Kasparov to a 3:3 draw in one challenge. Interesting bishop sacrifice it used in one of the games - one of the better AI moves I have seen I must admit :) http://www.chessbase.com/shop/product.asp?pid=170& user=&coin=

Sorry... had to

[Frogbert]

Well its obvious chess would be an obsession in Russia. Why?

Because in Soviet Russia Chess plays you!

HMM --

[Grommet - Space Cade]

Knight to rook 4.......You have performed an illegal move your computer has become unstable please lay down your king and rebbot the system. We are sorry for this inconveinience if problems persist please contact your system administrator

Being open source doesn't hurt Crafty

[mebon]

Last November one of the programs, List, was disqualified for being suspected of using some of Crafty's source code.

Crafty may be open source but it looks like the rules won't allow competitors to use substantial parts of another competing program's code. So having the source available to everyone isn't a liability for Crafty.

Mebon

Re:Being open source doesn't hurt Crafty

[SLi]

Except that Crafty is far from being open source. From the license:

No part of this program may be reproduced in any form or by any means, for other than your personal use, without the express written permission of the author. This program may not be used in whole, nor in part, to enter any computer chess competition without written permission from the author.

Re:Being open source doesn't hurt Crafty

[Skuto]

How exactly is that not open source?

Re:Being open source doesn't hurt Crafty

[SLi]

Simple.

It fails at least OSD #1, #3, #4 and #6.

Its source is available, but open source it's not. Of course there are always some people who try to twist language enough to claim that "open source" == "source available", but then if you wanted to be understood you'd say "source available". Open source did never, and will never, mean simply "source code available for viewing under some arbitrary conditions which may or may not include paying a billion dollars".

Re:For those who think chess needs a little variet

[NarrMaster]

Thats a lot of variants on that site. I personally liked a variant with a 10x8 or 10x10 board, and two new pieces called the Cardinal and the Marshall. Cardinal moves like a bishop and knight and the Marshall moves like a rook and knight. Its pretty fun.

Re:For those who think chess needs a little variet

[ZzzzSleep]

Clicky!

No, Intuition is not the key to master-level play

[fw3]

Yes a human chess player has a few neat tools. The primary one is called:

Positional play

Algorithms / heuristics which have attempted to capture this 'intelligent' side of chess players' methodology have uniformly failed and the winning programs continue to primarily rely on simple evaluation of material.

This means that a master-level player has a strong advantage in offering a computer opponent some material in exchange say for superior control of the center of the board.

Advanced chess play has very little to do with 'intuition'. The specific tools that come to bear are:

exhaustive study of openings and endings
solid tactical evaluation (stupid mistakes still lose games)
positional evaluation

generally, for instance it's suicide to allow a game against a machine develop into an 'open' vs a 'closed' position. Tactical evaluation is less effective in closed positions; in open positions the machine's greater depth-search works extremely well.

Hissyfits?

[FlyingOrca]

...I think you've been playing too much chess at the Zone!

Or, if not, you certainly have the typical players pegged...

Um, something does prevent that from happening

[Anonymous Coward]

The fact that we don't know how to program it. Sure, there must be a way for a powerful enough computer to emulate human intuition. But that fact isn't much help in figuring it out. Sure, we know what has to go into the decision making process. But we don't understand how to make the analysis work once data is collected.

A litmust test is that if we knew how to program computers to be good at recognizing patterns, then we'd probably also know how to program computers to be good at Go. Which we don't.

Chess is highly tactical. A single "Ah, hah!" can spell the difference between a good position and a bad one. Given the ability to process the game farther versus provide a more sophisticated analysis of each position, going farther usually wins. Which is why computer chess programs have focussed more on efficiency than sophisticated analysis. Which makes it a poor battleground for human intuition's ability to provide a very sophisticated analysis of a small number of positions.

By contrast Go is very strategic. It takes a long time for anything to resolve, so you really have to rely on intuition about how things will turn out far down the road. This is a much better battleground for human intuition, as is shown by the fact that computers don't do well at it.

World Computer Go Championships also

[Portfolio]

I'm surprised no one has mentioned that there are other World Computer game championships at the same event, and Go is one of them. It is even another Open Source vs. Commercial slugfest, with GnuGo the favorite from the open source community.

Re:you lose - here is the firstest GO post

[Anonymous Coward]

This time perhaps - you darn meddling kids.

Re:Interesting computer Chess? Follow the Money...

[TygerFish]

-- that may be one of the reasons that computers aren't as good at Go as they are at chess, Go appearing (to a rank beginner) to depend more on pattern recognition and less on straight-forward deductive analysis.

A simpler answer than one of 'intuition versus analysis' to resolve questions about machines playing Go to machines playing chess is to be found in a statement borrowed from a completely different field of study: 'follow the money.'

Chess has fascinated western thinkers since before there were computers and the problems of how to make a machine play chess have been approached by everyone from great players (M. Botvinnik, himself a world-champion) to corporate technology research centers.

'Intuition versus analysis' might be useful in talking about computers playing chess or playing Go, but one thing you have to take into account is the sheer volume of readily available and examined research on the subject.

Chess as an intellectual pursuit is a part of western culture and has had vast intellectual and material resources poured into it for decades, were the same true of Go, you might see more machines that played it better.

Re:For those who think chess needs a little variet

[Anonymous Coward]

Here is you post as it would appear with html tags and the link made clicky: (Just remember to select post type as HTML formatted... (and now you know!)

<a href="http://www.chessvariants.com">www.chess.vari ants.com</a> is a fun site.
<br>
<br>
(I'd make it a clicky but I don't know how to do so)

Matter of time, next FRC or variant?

[Manic Ken]

I would like a computer to play the FRC way or some varian of that. THAT would be interesting.

Re:Matter of time, next FRC or variant?

[Anonymous Coward]

No problem, you can do that, with some UCI or WB engines under the chess GUI "Arena" : FullChess-Engines

Re:Matter of time, next FRC or variant?

[Manic Ken]

Good stuff!! Seems to be working under wine as well, have to verify this..later...gotta go out. Cheers.

why is it some jack-off

[dh003i]

has to post a comment about "Go" everytime that Chess is mentioned? Go is a great game, but obviously not as popular as Chess. Maybe it won't ever be, maybe it will. Get over it.

Re:why is it some jack-off

[Anonymous Coward]

In Asia, Go is far more popular than chess, so I'm not sure what your point is. In fact, I'd be willing to bet that if you totalled the number of people worldwide who regulary play chess and compared it to the number of people world-wide who regularly play go, the go players would match if not outnumber the chess players. Asia is over 1/3 the world's population, afterall - and go is much more prevalent in Asian society than chess is in Western society.

Re:why is it some jack-off

[Anonymous Coward]

And in Asia's "Srashdot.org", there is always some fucker who posts about Chess in a Go-related article.
BFD.

Re:why is it some jack-off

[hunterx11]

Actually in Japan, Shogi is much more popular than Go, but many of the Go players consider themselves more cerebral (just like with chess). Shogi really is a violent game compared to western chess, though I rather enjoy it; pieces that are captured can return to the board under enemy possession, and draws are unheard of (in the very rare event of a draw, they replay immediately).

Re:why is it some jack-off

[Anonymous Coward]

Because they are related topics. Really, it's OK.

Lasker was a very good player

[dh003i]

Lasker was probably the best chess-player ever, better than Kasparov, better than Fischer. Translated into today's rankings, he would have ranked about 3000. In tournaments of all of the strongest chess-players in the world at the time, he dominated brilliantly. He was the world champion for, what, 28 years? And chess wasn't even his main profession. I think that if Lasker had played Fischer or Kasparov, he would have won...and I don't think it would have been very close either.

Re:Lasker was a very good player

[hunterx11]

Lasker may have been higher above his opponents than Kasparov, but so was Morphy. Heck, so was Philidor. Back then there weren't as many people aspiring to be masters at chess. If Lasker were alive today he would have a lot of theory to catch up on, and not just novelties like new openings or hypermodernism. Kasparov has been above 2800 for some time now, and he has plenty of close games with people 200 Elo below him.

Brief summary

[Anonymous Coward]

There are really two classes of computers in the tournament: the good and the bad. Crafty has established itself as the former with draws against Shredder (one of the top programs), Deep Sjeng (not Shredder, but still in the good category), and a win against one of the bad programs. The tournament is typically won by the program which beats all the bad programs and manages to beat some of the good ones. Since the good programs are so hard to earn a win against, giving up a draw to one of the lesser entrants is very, very bad. Crafty has not done this yet, which bodes well for our open source hero.

Re:For those who think chess needs a little variet

[hunterx11]

Chess variants is definitely Eurocentric and culturally offensive. Shogi and Xiangqi and Changgi are chess variants? No, chess is a Xianqi variant! A Shatranj variant! A cheap ripoff of Makruk! A copyright-infringing Chatarunga clone!

Crafty is _not_ Open Source or Free Software

[Anonymous Coward]

The Crafty license is not an Open Source or Free Software license. It only permits personal use of the program, and also requires changes to be published, prohibiting private non-distributed modifications.

Here is the Crafty license, from Debian's debian/copyright file database:

All rights reserved. No part of this program may be reproduced in any
form or by any means, for other than your personal use, without the
express written permission of the author. This program may not be used in
whole, nor in part, to enter any computer chess competition without
written permission from the author. Such permission will include the
requirement that the program be entered under the name "Crafty" so that
the program's ancestry will be known.

Copies of the source must contain the original copyright notice intact.

Any changes made to this software must also be made public to comply with
the original intent of this software distribution project. These
restrictions apply whether the distribution is being done for free or as
part or all of a commercial product. The author retains sole ownership
and copyright on this program except for 'personal use' explained below.

personal use includes any use you make of the program yourself, either by
playing games with it yourself, or allowing others to play it on your
machine, and requires that if others use the program, it must be clearly
identified as "Crafty" to anyone playing it (on a chess server as one
example). Personal use does not allow anyone to enter this into a chess
tournament where other program authors are invited to participate. IE you
can do your own local tournament, with Crafty + other programs, since this
is for your personal enjoyment. But you may not enter Crafty into an
event where it will be in competition with other programs/programmers
without permission as stated previously.

Re:Crafty is _not_ Open Source or Free Software

[Skuto]

>The Crafty license is not an Open Source or
>Free Software license.

It's not because some zealots redefined what
"open source" or "free software" are supposed
to mean that the rest of the world has to follow
their perverted definition.

The source is available for anyone. Hence open
source.

Re:Crafty is _not_ Open Source or Free Software

[Anonymous Coward]

Open Source is a registered service mark, hence *not* open source.

Re:Crafty is _not_ Open Source or Free Software

[Anonymous Coward]

> The source is available for anyone. Hence open source.

By that logic, all books are open source, because anyone can read them, and all music is open source, because anyone can listen.

What is missing is the freedom to distribute copies and changes.

Requires Closed-Source (Win32) Client Software!

[privaria]

How will the open source chess engine Crafty do against the proprietary closed engines?

I'd check out some of the action if I didn't have to use playchess.com's proprietary closed client software to do so...

I think I'll stick with FICS, thanks.

good & bad people

[Anonymous Coward]

i understand your point. though imo no such
thing as "good" or "bad" people. all of
us have good AND bad qualities. "black-and-
white" personalities only exist in comic books.

computer Vs human

[tommeke100]

The funny thing about chess computer programs is that, while they can beat human players, it's possible that they perform so-so against other programs that lose against humans.
I read something about how they can adapt they're play-style to the style of the humans, but this turns out really different against other computers.

Re:Interesting computer Chess? Follow the Money...

[Tyler Durden]

Chess as an intellectual pursuit is a part of western culture and has had vast intellectual and material resources poured into it for decades, were the same true of Go, you might see more machines that played it better.

There probably has been just as much intellectual resources poured into Go as there has been with chess over the years as Go is just as old if not an older game, and the hemisphere those resources come from doesn't matter one whit.

The only problem from a programming standpoint is the analysis of what constitutes good moves in chess is much more specific and direct while in Go they are more hazily defined. This is partially because chess is a much more direct game (I may lose a pawn in this move, therefore I must justify it with a comparable gain in positional advantage or material later on) than Go (I may lose a stone here - *feh*.) The reasons for this are that there are so astronomically many more viable possible moves in Go that direct analysis is simply impractical, and the more intuitive skill of pattern recognition is more helpful to a Go player than it would be to a chess player.

So I think it would be more accurate to say that the nature of intellectual inquiry of both games may explain why one plays better on computers than the others. But even then, it's likely that it's simply a natural result of the natures of the games.

Obligitory Futurama quote

[Marshall Banana, Esq]

Deep Blue: Pawn to D8!
Al Gore: Not all problems can be solved with chess Deep Blue. Someday you'll understand that.

offtopic

[Anonymous Coward]

Thanks. As a former Danzig employee, I must say that you made my day.

Unfortunately, you have also caused www.northsidekings.com to be slashdotted.

A two-fer!

I'd rather concentrate on Go than Chess

[wfolta]

Not dissing chess players -- I was on the high school chess team myself -- but it feels to me that chess programs are in beat-a-dead-horse mode now. I mean, the best programs can beat all but a handful of humans and from my brief reading on the subject it would appear that most of the progress that's been made in the last 20 years has been faster computers searching move trees to greater and greater depths.

By comparison, the best Go programs in the world play at intermediate amateur levels. Why not explore this fertile territory?

For those not familiar with Go, there are many differences that make it much harder to make a good Go program...

1. Chess starts in a fixed position, with a full set of pieces of which the most powerful are blocked in, while Go starts with an empty board. In fact, a Go game might be thought of as 9 skirmishes that flow over into each other. (Battles usually start first in the four corners, then spread to the four edges, then into the center.)

2. A chess board is 8x8 while a Go board is 19x19, which means large search trees for any brute-force search.

3. In chess, there are 6 different kinds of pieces, each with a different role and capabilities. In Go, all pieces are the same. In fact, pieces are placed and do not move (though they can be removed by capture) and basically have no power on their own. It is the configuration of groups of pieces that determines their safety, power, and influence.

4. Go has an aesthetic element to it that is hard to describe. It's amazing to me how close to strategically correct you can get by looking at the black and white stones aesthetically and saying, "Looks like my piece would look good here." Of course, "close" is not usually good enough and there's the matter of timing.

5. In Go, the board is filling up throughout the game and territorial control is being solidified. This means that timing is critical and there is always a tension between making an important move now and making a more urgent move first. This wrestling over iniative ("sente" in Japanese) is a key factor in the game.

6. I don't know how to describe it, but Chess seems much more constricted than Go. The ultimate symbol of this, for me, is zugzwang, which is a point in chess where it is a disadvantage for it to be your turn. Any move you make is bad. In Go, by comparison, it is always to your advantage to move right up until the end of the game. In fact, you're free to pass and skip your move at any time and the game ends when both players pass.

Others can add more observations, but I think these are some of the areas where Go is somewhat more "squishy" than chess and a much better challenge. I mean, Chess has basically already been conquered in one sense. Move along people, nothing to see here.

(As a side note, one of the beauties of Go for humans is its handicap system. You can play against opponents up to 9 levels better or worse than yourself, with even expectations of winning. The way this handicap is implemented is elegantly and seamlessly part of the game, allowing the stronger player full use of their toolbox of experience yet still balancing the game. By comparison, chess handicap games literally strip pieces from the stronger player, changing the very nature of the game, IMO.)

Re:I'd rather concentrate on Go than Chess

[Anonymous Coward]

By comparison, the best Go programs in the world play at intermediate amateur levels. Why not explore this fertile territory?

Because then computers will master it and there will be no point in playing Go either!

WON'T SOMEONE THINK OF THE HUMANS?!??!?!

Remember Wargames?

[Anonymous Coward]

Remember the old Matthew Broderick movie, Wargames? That movie taught me that two knowledgeable players would always tie at tic-tac-toe. As I was 5 years old at the time, that was quite a revelation. I played around enough and learned how, myself.

I see the same happening in computer chess. The programs will get better and better but the results will lead to more and more draws. Eventually, the game will be ruined. I know I see no more reason to play tic-tac-toe.

go vs chess

[chanceH]

curious how much of go versus chess complexity is due to inherient nature of game, and how much is just due to size of the board.

I.e. how good is best computer algorithm (relative to 90%th percentile human player) at 8x8 go?

Or consider this made up chess like game, played on a 19x19 board. Back rank looks like this:

54321RNJBKQBNR12345, and front rank is still all pawns.

5 = King + Rook (can move like Rook or a King)
4 = King + Bishop
3 = King + Knight
2 = Rook + Knight
1 = Bishop + Knight
R = Rook
N = Knight
J = Jester. Moves just like king, but just a piece.
Q = Queen
K = King

Would same-ole-same-ole chess methods in current use be as good at this game (relative to humans) as in normal chess?

My guess that, given N, the size of one side of the board

a) (given that N > than some 'small' N), go is in fact harder.

but:

(b) both games have as a part of their 'hardness' a quadratic relation to N, and a lot of these comparisons aren't really fair since 19^2/8^2 ~= 5.6

For personal playing purposes I like go a lot more than chess, but I wonder if I might like my made up chess game above better than either.Probably not .... probably would still always have that feeling of "why can't I just write a program to check my moves for obvious blunders".

well, playing the what-if game

[dh003i]

If Kasparov was alive when Lasker was playing, he wouldn't have had the benefit of all that had been developed since Lasker.

Lasker has had a bad reputation, due to a comment made about him by Bobby Fischer ("a bad player...a coffee-house player"). However, if you analyze the games he played, you'll find they are extremely brilliant.

or Futurama...

[SeanDuggan]

During the episode 2.2: Mars University courtesy of The Neutral Planet

[In another room two other fratbots sit at a table with a chessboard. The room is a mess, littered with empty pizza boxes, cans and books. A sock hangs over the moose head and there is a dartboard hanging on a door with darts jabbed in the wall around it. There is a pin-up of a Fembot and a L&#246;brau poster with a human woman on it on another wall. One of the other fratbots looks like he has glasses painted onto his face and the other one is very fat. The glasses one looks at the chessboard. The game hasn't begun.]

Fratbot #2: Mate in 143 moves.

Fratbot #3: Oh poo, you win again!