Wikipedia:Reference desk/Archives/Entertainment/2010 December 1

= December 1 =

How many possible moves are there in Checkers, Chess, and Baduk/Go?
Moreover, how powerful will computers have to be in order to become unbeatable in those 3 games? In what years do we estimate computers to become advanced enough to become unbeatable in those 3 games?

PS: Is there a more complex game than Baduk? How many possible moves are on that one, how powerful will a computer need to be in order to be unbeatable on that Über game, and in what year do we estimate computers to become that advanced?

--Let Us Update Dusty Articles 01:32, 1 December 2010 (UTC)


 * I'm going to pass this question along to wikipedia's resident chess expert. ←Baseball Bugs What's up, Doc? carrots→ 01:52, 1 December 2010 (UTC)


 * Checkers - Computers have been able to play perfect checkers for a few years now, since 2007 according to Checkers. So they are unbeatable(subject to computer/power failures etc) in that.
 * Chess - In chess the best computers became better then the best human around 1997 (Kasparov best human at time), see the Deep Blue versus Garry Kasparov. Today, the best computers(Rybka(I think) are easily better. In chess there is Elo rating system that you can quite accurately work out ability and highest human is around 2800 Elo and highest computer around 3300 Elo. However, computers can still lose to humans even at fast chess where advantage of computer is greatest, mainly due to the Horizon effect, see Anti-computer tactics (gaming). It is thought unlikely computers are ever going to play perfect chess see First_move_advantage_in_chess.
 * Baduk - I think Baduk is like chess but I'm not sure but read here Baduk_(game). Regards, SunCreator (talk) 02:20, 1 December 2010 (UTC)


 * (ec) In 8x8 checkers/draughts, Chinook (draughts player) can't be beat. In chess they are still not perfect, but a human is very unlikely to beat the top programs.  The last I heard, a human can beat a computer if played at correspondence chess with plenty of thinking time (for each).  I don't know much about Baduk, but those games are generally ones where the computer doesn't do well compared to the best people.  When you say "beatable", do you mean by a person?  Because top programs do beat other top programs frequently.  Bubba73 You talkin' to me? 02:25, 1 December 2010 (UTC)


 * In chess you can't say the number of possible moves, but you can estimate the number of total positions and according to Shannon(Shannon number) it is ' estimated the number of possible positions, is "of the general order of $$\scriptstyle \frac{64!}{32!{8!}^2{2!}^6}$$, or roughly 1043". Although this number is likely a bit on the high side. In short computers are not going to approach being able to calculate enough by doubling in speed every 2 years for many years and in doing so current scientific barriers such as maximum speed of light would seem to come into play. Regards, SunCreator (talk) 02:41, 1 December 2010 (UTC)


 * Personally, I'm with Jonathan Schaeffer who said 'never underestimate the advances in technology' and so if I had to guess I'd say chess will be solved perfectly within 50 years. I'm sure almost everyone else would disagree however and I'm guessing based on observational information that could well be wrong. Regards, SunCreator (talk) 02:52, 1 December 2010 (UTC)


 * Computer advances are slowing down and there are physical limits as to how big and how fast computers could be. Bubba73 You talkin' to me? 02:55, 1 December 2010 (UTC)
 * That slowing of advances is due to reaching the limits of classic, binary, silicon-transistor based computing. The next big quantum leaps in computational speed are likely to be made by fundemental changes in the medium (such as light-based computers) or the fundemental paradigm of computers (such as neural nets or some as-to-yet unseen computer paradigm).  There were limits to how fast horses ran; no one forsaw breaking those limits until the automobile was invented.  Same situation here.  -- Jayron  32  03:08, 1 December 2010 (UTC)
 * Indeed Jayron32. A paradigm shift is due soon :) Regards, SunCreator (talk) 03:21, 1 December 2010 (UTC)
 * Those will still have limits. Light moves only so fast.  It is going to take a certain amount of matter to represent a bit.  Bubba73 You talkin' to me? 03:46, 1 December 2010 (UTC)
 * Bit storage or mass storage is doubling each year from Kryder's Law, read String theory for one possible idea where storage within atomic level may go. So what you end up with is a situation from the Wheat and chessboard problem, when constant doubling (computing power and storage) gives you massive leverage over time. Regards, SunCreator (talk) 04:02, 1 December 2010 (UTC)
 * It is not going to keep doubling indefinitely. In First_move_advantage_in_chess that you mentioned above, it talks about a 1965 paper saying that no conceivable computer could examine all branches of chess (of course, you can cut that down by a sizable factor, but it is still a large number). It goes on to say that 45 years of technical progress haven't significantly altered that evaluation.  Bubba73 You talkin' to me? 04:26, 1 December 2010 (UTC)

UTC)
 * Moore's law has been going fine since 1958. By saying it is 'not going to keep doubling' implies Moore's law is false but observationally evidence shows it is true. Regards, SunCreator (talk) 05:16, 1 December 2010 (UTC)

It can't go on indefinitely, see Moore's_law. The last time I checked, the doubling time had slowed to 26 months. I think it will fall short of what would be needed to solve chess. As the article you quoted in your first comment says, the evaluation hasn't changed significantly in 45 years. Bubba73 You talkin' to me? 05:44, 1 December 2010 (UTC)


 * This argument about Moore's law is an argument of semantics about what Moore's law applies to. Using solely integrated circuits, a maximum speed will soon be reached.  Similarly, if integrated chips never replaced transistors, a maximum speed would have been reached long ago.  If you consider Moore's law to apply to any computing device (not just computers with an IC CPU), a completely new technology may replace the integrated circuit, allowing for much faster speeds (just as the IC replaced transistors and transistors replace vacuum tubes and vacuum tubes replaced relays...). --  k a i n a w &trade; 16:54, 1 December 2010 (UTC)


 * Moore's law only needs to go less then another 50 years and chess computers will be unbeatable. The reason why 'the evaluation hasn't changed significantly in 45 years' and it will in the future is shown in Second half of the chessboard. Within the next 30 to 50 years you'll have chess computers playing at Elo rating system of 4800 to 6500 and then it will be 100% draws. Regards, SunCreator (talk) 15:38, 2 December 2010 (UTC)


 * No one has said anything about the solving of Baduk, and how many possible moves there are in that game. Besides, another computing paradigm shift could bring us to Quantum computing, and if QC takes off in 2015, how soon would it be able to solve Chess and Baduk from then on out? --Let Us Update Dusty Articles 07:10, 1 December 2010 (UTC)


 * Moore's Law can't go on indefinitely, no matter what the technology. Nothing moves faster than light and light has a finite speed.  Atoms have a finite size.  There are provable physical limits. Bubba73 You talkin' to me? 18:11, 1 December 2010 (UTC)


 * It is semantic. What are you applying Moore's law to?  The number of cores on a CPU?  The speed of data moving through a circuit?  The amount of data items stored in memory?  The number of calculations performed in a second?  Claims of physical limitations mean nothing until you state what you mean by "Moore's law". Within this thread, it is referring to the increase in a single computer's ability to perform more and more calculations a second.  That is not dependent on the speed of light.  It is dependent on the amount of data and the number of microprocessors you can place in a computer - assuming that a new technology doesn't redefine computers such that our current concepts of data and microprocessors is tossed out. --  k a i n a w &trade; 18:24, 1 December 2010 (UTC)


 * I don't think faster technology will ever be enough to play a perfect game of chess. If it is ever done, it will take a great deal of human analysis and work too.  Prove me wrong - if and when it is actually done then you can tell me "I told you so".  (I welcome being proved wrong on this.) Bubba73 You talkin' to me? 18:42, 1 December 2010 (UTC)


 * Our article Solved game is relevant. Comet Tuttle (talk) 19:59, 1 December 2010 (UTC)
 * It is, although note the original question asked when will computers become unbeatable in those 3 games - unbeatable is less then ultra-weak. You can be unbeatable without perfect play, you just need to have the ability not to lose.

LUUWDA -- You may be interested in Arimaa, which was designed to be a computational "über-game" using only a standard chess-board and chess pieces, and without pointlessly baroque rules which don't contribute to game play... AnonMoos (talk) 11:33, 2 December 2010 (UTC)

who mixes club hits givven cds?
Who mixes SPG music's club hits 2009, club hits 2k9, club hits 2010 and club hits 2k10? Is it wolfgang gartner, or deadmau5? it sounds more like their style of house, so i would like to know, is it either of them, and if it is someone else, who? N.I.M. (talk) 02:12, 1 December 2010 (UTC)


 * Email info@spgmusic.com for information they do not publish on their website. -- k a i n a w &trade; 14:44, 1 December 2010 (UTC)

Beatles and Oasis
in which year did the beatles gain fame?are they all dead?for which song did they gain fame for?why did the band oasis break up? —Preceding unsigned comment added by 122.179.22.237 (talk) 15:57, 1 December 2010 (UTC)
 * I gave your questions a title. The Beatles and Oasis (band) have the relevant information. Vimescarrot (talk) 16:11, 1 December 2010 (UTC)

Radio Nova Ireland
Does anyone have any info on an American disc jockey by the name of Jessie Brandon who worked on Ireland's Radio Nova in about 1985? Thanks.--Jeanne Boleyn (talk) 18:44, 1 December 2010 (UTC)
 * According to a google search, she seems to have worked for a number of radio stations, including KISW and Laser 558, and ended up back in the USA as a radio news editor for this producer of radio news and features for syndication in the late 1990s/early 2000s. She's pictured here attending a Laser reunion as part of Radio Day 2009 in Amsterdam.  Ka renjc 21:35, 1 December 2010 (UTC)
 * Thank you for the information and link.--Jeanne Boleyn (talk) 10:21, 3 December 2010 (UTC)

Re: en.wikipedia.org/wiki/List_of_social_networking_websites;
> I know, I know, I hear you, but I'm, (at this moment, clueless who to ask), thus, I'm asking you. . . > > Say, for Hermosa Beach, CA 90254, AND I do see: http://en.wikipedia.org/wiki/List_of_social_networking_websites, and I have (tried) to ask around via email unto local HB city office, newspaper, media, but thought ya might know better 'edge' for me to try. Me trying to find out what social networking website's are Hermosa Beach's 19,000 plus population on and/or using the most. I have tried to network/do homework, but, sigh. . . thanks. > > Me: http://www.jeffreydavidmorris.com  —Preceding unsigned comment added by 66.12.101.42 (talk) 21:20, 1 December 2010 (UTC)


 * Facebook is by far the most popular social networking website in America, and there's no particular reason to believe that people from Hermosa Beach prefer any other social network. Comet Tuttle (talk) 21:32, 1 December 2010 (UTC)