Game Of Chess

The Game of Chess is one of the most cerebral of sports. ChessIsLife!

Want to play online? There are many many opportunities:

If you want to watch GM or IM games, ICC typically has at least a couple GM's and several IM's logged on at any one time. Most major chess tournaments can also be watched live from ICC and FICS.

See: PatternsInChess, PostalChess, IrregularOpenings, also ThreeDeeChess, ChaTuRanga?


Computer Chess

The GameOfChess has been thought of as the fulcrum point to indicate that computers can think. If a computer could play a game of chess, and win, then by golly the computer can reason and evaluate the nuances of the game. Unfortunately, the computer has generally ended up using BruteForce by evaluating all possible moves in response to its proposed move, then choosing the move which has the best advantage that it can see. This is not thinking, but calculating. As computers continue to win at chess over masters (they just don't tire out...), it still does not prove that computers can think. They likely never will. (According to your calculations?)

Chess computers overtook humans in 1997, when DeepBlue beat World Champion Kasparov. From 2006-2010, the computer champion Rybka on a PC was so strong it offered odds matches to GMs! In 2011 many programs surpassed even Rybka: Houdini, Critter and even the OpenSource CeePlusPlus masterpiece Stockfish. By 2014, they were joined by Komodo, Gull, and Fire (begot from Ivanhoe, begot from Robolitto, begot from Ippolit, which was probably reverse-engineered from Rybka, which might have stolen code from Fruit). Even HIARCS running on a cell phone has won strong human tournaments!

This sort of combinatorial attack is much less effective in GameOfGo (and indeed the best Go programs are still very weak, compared to the best chess programs), so maybe ability at Go might be a better marker for computer thinking.

All sorts of discussion on computer chess takes place on the Computer Chess Club forums: http://talkchess.com/. Many chess programming experts actively participate, such as Robert Hyatt, Marco Costalba, Ed Schröder, and Gian-Carlo Pascutto.


AFAIK, no one has applied the XP methodology to writing a chess program. How do you write a UnitTest for the user story "can beat Kasparov"? That would be a CustomerTest. RegressionTests could tell you whether that last experimental change made the program weaker.


There is a chess openings wiki: http://chess.sourceforge.net/openings/index.php?FrontPage

There is also a Danish wiki on chess at http://www.caissa.dk/Wiki.jsp?page=CaissaKolding (in Danish) which also covers chess for younger people and children. If you want to edit, you need to ask for access though, since the foreman doesn't like everyone to edit without his approval. The RssFeed? is at http://www.caissa.dk/rss.rdf and the news page at http://www.caissa.dk/Wiki.jsp?page=SkakNyheder actually brings some interesting news. The engine used is JspWiki.


CategoryGame


EditText of this page (last edited November 11, 2014) or FindPage with title or text search