Xtreme Toy Zone

Feature: Brute force or intelligence? The slow rise of computer chess

Well, all of the above—or below, actually. In the rear bottom interior of the box sat a flesh-and-blood operative (by necessity a small one) who followed the human contender's moves from below and maneuvered The Turk's right hand across the table board. Nonetheless, the machine became "the most famous automaton in history," Standage notes, commented on by Charles Babbage, Edgar Allan Poe, Benjamin Franklin, and Napoleon Bonaparte.

More importantly, The Turk whetted the West's appetite for real devices that could do such things. Over two centuries later, this project culminated in Deep Blue—the IBM computer that bested Russian chess champion Gary Kasparov in 1997.

But what's most fascinating about "Mastering the Game," the Computer History Museum's computer chess exhibit, is that it frames the rise of the automated chess playing as a debate between two philosophies of computing. One emphasized the "brute force" approach, taking advantage of algorithmic power offered by ever more powerful processors available to programmers after the Second World War. The other has foregrounded the importance of teaching chess computers to select strategies and even to learn from experience—in other words, to play more like humans.

Source: Ars Technica

Getting better thanks to technology - Chess

A recent study suggests that there are more great chess players now than there ever have been – and that players continue to improve. The authors of the study noted and confirmed that the ELO rankings used to determine how good you are at chess have been stable over time — that is, unlike grades and health care costs, they haven’t seen any inflation. (Indeed, they may have seen a slight deflation.) Despite this, the number of high-ranked players is significantly higher:

In the 1970’s there were only two playerswith ratings over 2700, namely Bobby Fischer and Anatoly Karpov, and there were years as late as 1981 when no one had a rating over 2700. In the past decade there have usuallybeen thirty or more players with such ratings. Thus lack of in?ation implies that those playersare better than all but Fischer and Karpov were.

The authors don’t suggest a mechanism, but if I were to guess, I’d suggest that the reason for this has to do with competitive chess software and online play. When you can play chess on the computer, you don’t have to waste time setting up the board, finding other players, or anything else. The net result is that in the same amount of time, you can more games under your belt. Moreover, a lot of chess software will analyze your games for you and allow you to record them so you can examine your own play. All of these factors, I suggest, lead to a greater number of people having superior chess expertise....

Read more...

Q&A: If the world champion chess players from 100 years ago played ...

by chess-learning

I believe the compter would win, especially if you’re talking about Hydra or Rybka.

They might be able to manage a few draws here and there, but I don’t think they would win very many games against these monsters.

In a long match, the computers would have an advantage, since they would not be subject to fatigue.

I think guys like Lasker and Steinitz (as well as others like Capablanca, Alekhine, Marshall etc) could probably beat most of the good desktop PC programs, like Fritz, etc, even in a long match.

But when you’re talking about something like Hydra, with all the chess-specific processors, coupled with the fact that much chess knowledge ‘discovered’ by Steinitz, Lasker et al has been incorporated into these programs, the machine has a huge advantage.

Plus, computers like that can’t blunder in the endgame because of their endgame databases…

And finally, Hydra would not be intimidated ‘sitting’ across the board from Lasker, Capa, et al.

Read more...