EPTWorld Chess Championship 2013 - Page 4
| Forum Index > General Forum |
Chess discussion continues here | ||
|
Triggi
79 Posts
| ||
|
Cel.erity
United States4890 Posts
On November 05 2013 06:15 Grumbels wrote: I don't like the format, I read an article on chessbase that you need 20+ games for the match to be statistically significant. In that case 12 games is just very low. And of course a lower number of games favors opening preparation styles. I think the format is very good for Anand, and the games are in his hometown. Normally Carlsen should win this, but because of the format I give Anand a 40% chance. I agree, and I also think the world championship format overall is just terrible. I don't like the fact that someone can literally decide to play no games of chess for an entire year and still be called the world champion as long as he wins a single match. I'd prefer it greatly if there were simply a candidates tournament that cut to top 4 or top 2 for match play. That aside, I think the match will either be a slight victory for Anand or a landslide for Carlsen. I actually don't believe it's going to come down to Anand's opening preparation, because Carlsen can and will take risks to minimize that. Rather, I think it will be an issue of whether Carlsen presses too hard in objectively equal positions. I can honestly see Carlsen winning every single game as white, but I can also see him losing the first game and going a bit crazy trying to get his momentum back. This is a match that I believe will be decided in the middlegame and endgame, which is pretty exciting to me. On November 05 2013 11:04 Dogfoodboy16 wrote: Is Carlson considered to be a savant or is he just a very smart chess player? He's been touted as the future best player in the world for a good 7 years now, and he has been fairly dominant in tournament play for the past 3. Partially his popularity is due to his youth and demeanor, and the surge of new chess media, but he also has a chance to be the best player of all time. | ||
|
Dogfoodboy16
364 Posts
| ||
|
GolemMadness
Canada11044 Posts
| ||
|
Dogfoodboy16
364 Posts
| ||
|
GolemMadness
Canada11044 Posts
On November 05 2013 13:32 Dogfoodboy16 wrote: Its kinda surprising that the game of Chess hasent been solved yet. There are about 10^120 possible distinct games. You think that their would be an algorithm to solve the game by now. There are 32 pieces at the beginning of the game. In the early 90's, a tablebase that could solve any position with 5 pieces or less was made. It took until 2005 to bring that up to 6 pieces, and only recently was a 7-piece one released. The 6-piece tablebase required 7.05 GB, while the 7-piece tablebase requires 120 TB. That's over 17,000 times more space for just one extra piece. Now imagine how much it'd take to even go up to 10 pieces, which still isn't even close to the 32 you have at the beginning of the game. | ||
|
slowbacontron
United States7722 Posts
On November 05 2013 13:32 Dogfoodboy16 wrote: Its kinda surprising that the game of Chess hasent been solved yet. There are about 10^120 possible distinct games. You think that their would be an algorithm to solve the game by now. That's about 10^40 times more than the number of atoms in the universe. I'm not knowledgeable on what scale of problems can be solved with computers these days, but chess is significantly complex in any case. | ||
|
urboss
Austria1223 Posts
On November 05 2013 14:47 slowbacontron wrote: That's about 10^40 times more than the number of atoms in the universe. I'm not knowledgeable on what scale of problems can be solved with computers these days, but chess is significantly complex in any case. The Shannon number gives you the number of distinct games. There may be more than 10^120 distinct games but there are only about 10^45 possible positions. A lot less than the atoms of the universe. Positions may repeat themselves in many games, for a tablebase you only care about distinct positions. The number is still far too high to make it solveable. | ||
|
LittLeD
Sweden7973 Posts
| ||
|
Mr. Wiggles
Canada5894 Posts
On November 05 2013 13:32 Dogfoodboy16 wrote: Its kinda surprising that the game of Chess hasent been solved yet. There are about 10^120 possible distinct games. You think that their would be an algorithm to solve the game by now. There are several algorithms that can solve the game, but the problem is that solving it with them is infeasible due to resource constraints, i.e. it would take much too long. The problem isn't really that you can't solve the game with current search methods, it's that solving the game with those algorithms isn't efficient. For example, Checkers took ~15 years near-continuous parallel computation to solve, using (I believe) a Proof Number search variant. The number of positions in Checkers is roughly the square root of the number of positions in Chess, so using similar algorithms and the same computational resources, it would take maybe 200ish years of computation to solve Chess the same way. Sources: http://webdocs.cs.ualberta.ca/~chinook/ http://www.sciencemag.org/content/317/5844/1518.abstract?keytype=ref&siteid=sci&ijkey=jVmVcXy2/NTnY | ||
|
Blargh
United States2103 Posts
Though, I believe Anand actually has more wins against Carlsen in all the games they've played. Plus, he is on home turf, which is a bit lame, in my opinion. | ||
|
pebble444
Italy2497 Posts
Where can we watch this? what are exact times of the first game? | ||
|
Xiphias
Norway2223 Posts
 | ||
|
Aelfric
Turkey1496 Posts
On November 05 2013 10:46 marvellosity wrote: Houdini is pretty much unanimously the best. There are still Rybka fanboys running arround and komodo looks really strong lately, it is not that clear. Also are there gonna be any stream? | ||
|
Grovbolle
Denmark3805 Posts
| ||
|
Grumbels
Netherlands7031 Posts
On November 05 2013 15:57 Mr. Wiggles wrote: There are several algorithms that can solve the game, but the problem is that solving it with them is infeasible due to resource constraints, i.e. it would take much too long. The problem isn't really that you can't solve the game with current search methods, it's that solving the game with those algorithms isn't efficient. For example, Checkers took ~15 years near-continuous parallel computation to solve, using (I believe) a Proof Number search variant. The number of positions in Checkers is roughly the square root of the number of positions in Chess, so using similar algorithms and the same computational resources, it would take maybe 200ish years of computation to solve Chess the same way. Sources: http://webdocs.cs.ualberta.ca/~chinook/ http://www.sciencemag.org/content/317/5844/1518.abstract?keytype=ref&siteid=sci&ijkey=jVmVcXy2/NTnY (off-topic) + Show Spoiler + Here is a game that could happen if some computer had solved chess and was playing a beginner: 1. Nc3 - Nf6 2. Nb1 - Ng8 3. Nc3 - Nf6 4. Nb1 - Ng8 1/2 - 1/2 Because the computer knows the game is theoretically drawn it won't necessarily understand how to win. If you want the computer to win you have to program new algorithms simply to make use of your new database. On November 05 2013 17:31 Grovbolle wrote: Armageddon style tie breaker sounds pretty sick. Anyone knows what player will play as black or as white for this tie breaker? I think blitz games at that level will rarely draw so it's probably an advantage to the white player, but perhaps they came up with the five versus four minutes number to ensure it's perfectly even statistically. (I wonder how though) | ||
|
urboss
Austria1223 Posts
Anyone knows what player will play as black or as white for this tie breaker? I think blitz games at that level will rarely draw so it's probably an advantage to the white player, but perhaps they came up with the five versus four minutes number to ensure it's perfectly even statistically. (I wonder how though) For the armageddon match, the player who wins the drawing of lots may choose the color. | ||
|
warcralft
Singapore609 Posts
On November 05 2013 17:41 urboss wrote: For the armageddon match, the player who wins the drawing of lots may choose the color. I thought armageddon games are normally bid upon. Both players bid for white how much time they want. The lower bidder wins. | ||
|
Orome
Switzerland11984 Posts
| ||
|
urboss
Austria1223 Posts
On November 05 2013 18:05 warcralft wrote: I thought armageddon games are normally bid upon. Both players bid for white how much time they want. The lower bidder wins. From the rules: 3.7.3 If the score is still level after five matches as described in Article 3.7.2, the players shall play a one sudden death game. The player who wins the drawing of lots may choose the color. The player with the white pieces shall receive 5 minutes, the player with the black pieces shall receive 4 minutes whereupon, after the 60th move, both players shall receive an increment of 3 seconds starting from move 61. In case of a draw the player with the black pieces is declared the winner. | ||
| ||
