Machines attack, humans strike back

10/8/2004 – The man vs machine team chess championship in Bilbao started with a depressing 0.5-2.5 loss for the biological systems. But in round two humans struck back – in fact at one point they were very close to a 3:0 whitewash. The president of the sanctioning body ICGA, David Levy, gives us a provocative assessment of the event.

ChessBase 14 Download ChessBase 14 Download

Everyone uses ChessBase, from the World Champion to the amateur next door. Start your personal success story with ChessBase 14 and enjoy your chess even more!


Along with the ChessBase 14 program you can access the Live Database of 8 million games, and receive three months of free ChesssBase Account Premium membership and all of our online apps! Have a look today!

More...

Man vs Machine World Team Championship
in Bilbao, October 6 – 9, 2004

This encounter between chess playing entities, biological and electronic, is taking place in the city of Bilbao in the Basque region of Spain. It pits three strong grandmasters against three top programs. The humans are Veselin Topalov (Bulgaria, Elo 2757, world number five); Ruslan Ponomariov (Ukraine, 2710, No. 13); and Sergey Karjakin (Ukraine, 2576, who at 12 was the youngest GM in the history of the game).

The computers are: Hydra, a hardware machine running on a 16-processer array that is located in Abu Dhabi, UAE; Deep Junior, the reigning computer chess world champion, playing on a remote 4 x 2.8 GHz Xeon machine located at Intel UK (Swindon); and Fritz 8, running on a Centrino 1.7 GHz notebook in the tournament hall. The organisers felt that this would provide interesting comparative data, for amateurs and press, about program performances on three different platforms.

The rate of play is two hours for the first 40 moves and 60 minutes for the rest of the game.

The first round ended fairly disastrously for the human players. Young Sergey Karjakin was dominated by Fritz, while his friend and mentor Ruslan Ponomariov turned a position that looked at least optically winning into a loss to Hydra. Only Veselin Topalov was able to hold his game with black against Deep Junior. Score: 0.5-2.5 for the machines.


Fritz vs Karjakin, Ponomariov vs Hydra and Topalov vs Deep Junior in round one

On the second day the humans struck back – almost devastatingly. There was a phase in the round when it looked like all three would score full points against the machine. Sergey Karjakin outplayed Deep Junior from start to finish, never leaving a doubt about who was in charge. Ruslan Ponomariov got a substantial edge against Fritz, one that looked as though it could be transformed into a win. But the little notebook kept its calm and held the ex world champion to a draw.


Fritz on a notebook, operated by author Frans Morsch, against Ruslan Ponomariov


Young Sergey Karjakin in fighting mode against Amir Ban and Deep Junior


Veselin Topalov at the start of his game against Hydra, operated by Chrilly Donninger

The most dramatic encounter was between Veselin Topalov with the white pieces against the 16-processor Hydra. The human did not seem to have anything in the opening, but demonstrated his mastery of the game when he took advantage of an unfortunate pawn advance by Hydra by picking up this pawn and getting a totally winning position. However the six-hour session took its toll on the human's ability to focus his full attention on all the finesses of the endgame and in the end he had to concede a draw to the machine, which showed absolutely no signs of fatigue. Score: 2:1 for the humans.


After the game Amir Ban (right), who was following the endgame with his program Junior, shows the exhausted and disappointed Veselin Topalov how he could have won the game, while tournament arbiter Jaap van den Herik (right) looks on.


A Great Show!

By David Levy, President of the ICGA

There were some lively discussions here in Bilbao over the dinner table last night, following the 2.5-0.5 lead taken by the computer team over the Grandmasters in the 1st World Man vs Machine Team Chess Championship. Inevitably the subject of the Kramnik-Leko match also came under discussion and some comparisons were made. My own view is that the games in Bilbao are far more interesting for the chess public than are those played in Brissago. I would go further, and say that in the game won by Hydra, as Black, against Ponomariov, there was more drama than in all of the first seven Brissago games put together. Let me explain.


A festive dinner after game one. Second from right: ICGA president David Levy

There are several obvious differences between the type of chess played in a match for the human World Championship and the type of chess played in a top class human vs computer event. Firstly, the players in Brissago are both inordinately afraid of losing the game each time they sit down at the chessboard. This is not a criticism, it is a simple fact, one that is perfectly understandable when we consider what is at stake for Kramnik and Leko.

Being afraid of losing a game in a world championship match is, in many ways, the mark of a true chess professional. For those Grandmasters who play chess at the highest professional level a single loss in a title match is much more than the words suggest, it could mean the loss of the title, the loss of what might be one's last chance of gaining the title, the loss of that nice apartment that one would like to buy on the Boulevard Saint Germain in Paris, or even the start of a career nosedive. So we should not blame Kramnik and Leko for producing games that fail to delight the vast majority of chess enthusiasts. They must play at the highest level they can, to maximize their chances of success in the match, and without any regard whatsoever for the loss of chess artistry that results.


The computer people's end of the table: Frans Morsch, David Levy, Ali Nasir Muhammad, Chrilly Donninger, GM Christopher Lutz (who works for the Hydra team) and Jaap van den Herik

But although one might accept the approach of Kramnik and Leko as being the epitome of professionalism, insofar as they are both trying their very best to end their match as the holder of the World Championship title, this does not mean that at the same time one has to applaud games that are devoid of artistry. Imagine watching a boxing match in which the two antagonists dance around on opposite sides of the ring, afraid to go too near each other more than once every few rounds in case of a lucky blow coming in the opposite direction. If that was the approach of World Heavyweight Championship contenders, how long would boxing last as a spectator sport? It might be considered by some to be perfect or near-perfect professionalism, but does that interest the crowd? No, it does not. In boxing the crowd wants blood, just as the chess crowd wants to see blood on the chessboard.

In a Grandmaster vs computer game the computer program knows no fear and never shows any. The program is there to win if it can and the Grandmaster is therefore under pressure from the start of the game. So it is a completely different type of struggle, and this difference makes for more dramatic games from the perspective of the vast majority of chess enthusiasts.


Ponomariov vs Hydra in round one [Photo Ali Nasir Muhammad]

The Ponomariov vs Hydra game is, in my view, an example of high drama at the chessboard. Watching the game develop we see White taking complete control of the a-file, establishing a rook in a menacing position and eyeing a backward black pawn on c6. Meanwhile Black, the program, appeared to be floundering, for example when it played its bishop to g5 attacking a very well protected white pawn on e3. But look what happened over the next few moves. Black's feeble pawn on c6 was eaten up, while Black lashed out on the king side with the advance of its h-pawn, which looked at first sight like desperation. But it was not desperation. At that point in the game, despite being a pawn down, Hydra was displaying a plus score – it was searching so deeply that it already knew that it stood better. And within another couple of moves what had, at first, seemed like clutching at straws, was revealed to be a blistering attack against which it was almost impossible to suggest a defence for White. Maybe there was one, of sorts, by sacrificing the exchange with Rxf6, but whether it would have enabled White to survive is highly doubtful.


ICGA president David Levy in discussion with ICGA Journal editor Prof Jaap van den Herik

The opinion of the grandmasters here in Bilbao appears to be that I am completely wrong, and that the games in Brissago up until now (game 7) are definitely more interesting than those in Bilbao yesterday. Maybe this is true for them, but what about for the much larger number of chess fans throughout the world whose Elo ratings are hundreds or even a thousand or two points below the rarified level of the super GMs? If chess is to become a media success then the games watched by the masses of chess fans need to interest them. Do the games in Brissago interest the masses of chess fans? No they do not. They may interest many GMs and super GMs, but not vast majority of the less brilliant amongst us who number tens of thousands or more for every holder of the GM title.


Veselin Topalov not agreeing with David's arguments. Amir Ban considering them

Roll on the day when every World Chess Championship match is contested between the strongest human player and the strongest computer program. “But what happens...” I hear you ask, “...when the computers can regularly defeat the strongest humans? Isn't that the end of the story?” No, it most certainly is not. Kasparov's matches against computers have enthralled the chess world and many outside it, far more so than any of his matches against Karpov. If, ten years from now, Kasparov or his successor regularly succumbs to the leading programs, let us not despair but instead become creative. Perhaps pawn odds will come back into fashion, as it was in the 19th century. Who knows?

Pictures by Frederic Friedel

Links


Discussion and Feedback Join the public discussion or submit your feedback to the editors


Discuss

Rules for reader comments

 
 

Not registered yet? Register