x
Chess - Play & Learn

Chess.com

FREE - In Google Play

FREE - in Win Phone Store

VIEW

Did Hikaru Nakamura ruin his game by playing too much blitz?

  • #61
    fewlio wrote:

    Faster time controls is the future.  This years WCC was DECIDED by faster time controls as the longer time controls was a draw.  That should tell everyone everything they need to know.

     

    That had a lot more to do with the match strategy due to the asinine FIDE rules for the match.  Karjakin's strategy was to draw every game unless there was an obvious win (i.e. never try to create complications to play for a win).  If he wins 6.5-5.5 that is all that matters.  That is why you saw the same opening used in about 8 of the 12 classical games.

     

    If you look back, such a strategy was exactly why Fischer avoided the 1975 WCC (he wanted a match to be first to 10 wins and must win by 2 wins to avoid a single win followed by a series of draws).

     

    Previously, the champion was given draw-odds to encourage the challenger press for wins.  Alternatively, there was the Karpov-Kasparov style match (first player to 6 wins), which ended up in over 40 very impressive games before the match was decided.

  • #62
    Morphysrevenges wrote:
    fewlio wrote:

    Faster time controls is the future.  This years WCC was DECIDED by faster time controls as the longer time controls was a draw.  That should tell everyone everything they need to know.

    well even that statement is misleading if you are not familar with history. 

     

    the classical format has been trimmed down to very few rounds as compared to past matches so it is much harder for one player to obtain the required plus score to clintch the match before they move to the fast controls. 

     

    anyone recall K-K 1984 which went something like 48 games???? no blitz all classical.

    I might be reading this wrong, but aren't you making Fewlios point for him? If tournaments used to be all classical and no fast, and now they include fast, doesn't that  mean back then fast time control was the future?

  • #63
    lfPatriotGames wrote:
    Morphysrevenges wrote:
    fewlio wrote:

    Faster time controls is the future.  This years WCC was DECIDED by faster time controls as the longer time controls was a draw.  That should tell everyone everything they need to know.

    well even that statement is misleading if you are not familar with history. 

     

    the classical format has been trimmed down to very few rounds as compared to past matches so it is much harder for one player to obtain the required plus score to clintch the match before they move to the fast controls. 

     

    anyone recall K-K 1984 which went something like 48 games???? no blitz all classical.

    I might be reading this wrong, but aren't you making Fewlios point for him? If tournaments used to be all classical and no fast, and now they include fast, doesn't that  mean back then fast time control was the future?

     

    No, he isn't.  The WCC was 12 classical games.  The faster time controls were only there as a tie-break.  Karjakin's goal was to get to the faster time controls because there is a higher probability of making a mistake, which favors the weaker player (i.e. randomness increases the odds of the weaker player winning).  While some people liked the method for tie breaks, there are those (and I'm among them) that prefer the Fischer/Kasparov method where you must reach a specific number of wins over the opponent (basically, wins are +1, losses are 0, and draws are 0 for both players).  The drawback to that approach is it can lead to a LOT more games being played, but in the search for the strongest player (and no, computers are not players - there is no psychology involved in computer chess) more games increases the likelihood that the stronger player will be victorious.

  • #64

    Fast games = too many mistakes to enjoy analyzing, reporting or discussing them.

  • #65
    Graf_Nachthafen wrote:

    Fast games = too many mistakes to enjoy analyzing, reporting or discussing them.

    In an era where chess is solved, and theory is known, the quick execution of that theory becomes the most interesting contest.  Even at one-minute, you see high-level moves being played, deep combinations, and flawless endgame technique.  The errors which were once a problem are now what makes it competitive.  The better player will still win, but it's much easier to run an event in a single day than over a weekend.

    Most chess tournaments got their free space by bringing visitors to a hotel.  A quick tournament can be held almost anywhere, particularly in a restaurant in between lunch and dinner.

     

  • #66
    StupidGM wrote:
    Graf_Nachthafen wrote:

    Fast games = too many mistakes to enjoy analyzing, reporting or discussing them.

    In an era where chess is solved, ...

     

     

    You can stop right there.  If you think chess is solved, you are demonstrating your clear ignorance.

  • #67
    zac_howland wrote:
    StupidGM wrote:
    Graf_Nachthafen wrote:

    Fast games = too many mistakes to enjoy analyzing, reporting or discussing them.

    In an era where chess is solved, ...

     

     

    You can stop right there.  If you think chess is solved, you are demonstrating your clear ignorance.

    But chess IS solved. Not solved as in there is one and only one way to play the game and win, or one solution, but solved in that you can play an opponent and never, ever win. A computer will beat you 100% of the time at the problem of chess. That is a solved problem.

  • #68
    lfPatriotGames wrote:
    zac_howland wrote:
    StupidGM wrote:
    Graf_Nachthafen wrote:

    Fast games = too many mistakes to enjoy analyzing, reporting or discussing them.

    In an era where chess is solved, ...

     

     

    You can stop right there.  If you think chess is solved, you are demonstrating your clear ignorance.

    But chess IS solved. Not solved as in there is one and only one way to play the game and win, or one solution, but solved in that you can play an opponent and never, ever win. A computer will beat you 100% of the time at the problem of chess. That is a solved problem.

     

    That is not the definition of a solved game.  A game is solved when

     

    1) You can force a win by proper play, always.

    2) Proper play always results in a draw.

     

    To give you an example:  Checkers (8x8 - not 10x10) is weakly solved (meaning if both players play perfectly from start to finish, the result is a draw and that can be proven).

     

    Chess is NOT solved (and likely never will be).  It is assumed that proper play always leads to a draw, but that is not proven.  In fact, to be more specific:  it is "partially solved" - meaning the endgame tablebases show wins and draws for limited remaining pieces on the board.  There is nothing that proves a win or a draw from the start of the game.

     

    Engines are purely tactical beasts.  There is an article just a week or so ago that shows a position that an engine (even the strongest ones) gets wrong, even after thinking about it for hours, but a human gets correct instantly.

  • #69
    zac_howland wrote:
    lfPatriotGames wrote:
    zac_howland wrote:
    StupidGM wrote:
    Graf_Nachthafen wrote:

    Fast games = too many mistakes to enjoy analyzing, reporting or discussing them.

    In an era where chess is solved, ...

     

     

    You can stop right there.  If you think chess is solved, you are demonstrating your clear ignorance.

    But chess IS solved. Not solved as in there is one and only one way to play the game and win, or one solution, but solved in that you can play an opponent and never, ever win. A computer will beat you 100% of the time at the problem of chess. That is a solved problem.

     

    That is not the definition of a solved game.  A game is solved when

     

    1) You can force a win by proper play, always.

    2) Proper play always results in a draw.

     

    To give you an example:  Checkers (8x8 - not 10x10) is weakly solved (meaning if both players play perfectly from start to finish, the result is a draw and that can be proven).

     

    Chess is NOT solved (and likely never will be).  It is assumed that proper play always leads to a draw, but that is not proven.  In fact, to be more specific:  it is "partially solved" - meaning the endgame tablebases show wins and draws for limited remaining pieces on the board.  There is nothing that proves a win or a draw from the start of the game.

     

    Engines are purely tactical beasts.  There is an article just a week or so ago that shows a position that an engine (even the strongest ones) gets wrong, even after thinking about it for hours, but a human gets correct instantly.

    I think that we are talking about two different things. I am talking about a solved problem, you are talking about a solved game. I agree a game cannot be solved, because it's a game. Monopoly and poker and solitaire and yahtzee will never be solved. Besides the point I didnt know a solved game even had your definition. But the problem of chess is solved, and it will likely be solved even more definitively as time goes by. I agree that there are probably positions a human can beat a computer. But since the problem of chess starts at move 1. can a human beat a computer at that problem? No. The problem is solved. There is always a forced win by one side (your definition). And the forced win will likely be even more dramatic as time goes by.

  • #70

    Nakamura played far too much blitz chess and is now just a shadow of his former self, banished to a life of shameless red bull advertisements.

  • #71
    lfPatriotGames wrote:
    zac_howland wrote:
    lfPatriotGames wrote:
    zac_howland wrote:
    StupidGM wrote:
    Graf_Nachthafen wrote:

    Fast games = too many mistakes to enjoy analyzing, reporting or discussing them.

    In an era where chess is solved, ...

     

     

    You can stop right there.  If you think chess is solved, you are demonstrating your clear ignorance.

    But chess IS solved. Not solved as in there is one and only one way to play the game and win, or one solution, but solved in that you can play an opponent and never, ever win. A computer will beat you 100% of the time at the problem of chess. That is a solved problem.

     

    That is not the definition of a solved game.  A game is solved when

     

    1) You can force a win by proper play, always.

    2) Proper play always results in a draw.

     

    To give you an example:  Checkers (8x8 - not 10x10) is weakly solved (meaning if both players play perfectly from start to finish, the result is a draw and that can be proven).

     

    Chess is NOT solved (and likely never will be).  It is assumed that proper play always leads to a draw, but that is not proven.  In fact, to be more specific:  it is "partially solved" - meaning the endgame tablebases show wins and draws for limited remaining pieces on the board.  There is nothing that proves a win or a draw from the start of the game.

     

    Engines are purely tactical beasts.  There is an article just a week or so ago that shows a position that an engine (even the strongest ones) gets wrong, even after thinking about it for hours, but a human gets correct instantly.

    I think that we are talking about two different things. I am talking about a solved problem, you are talking about a solved game. I agree a game cannot be solved, because it's a game. Monopoly and poker and solitaire and yahtzee will never be solved. Besides the point I didnt know a solved game even had your definition. But the problem of chess is solved, and it will likely be solved even more definitively as time goes by. I agree that there are probably positions a human can beat a computer. But since the problem of chess starts at move 1. can a human beat a computer at that problem? No. The problem is solved. There is always a forced win by one side (your definition). And the forced win will likely be even more dramatic as time goes by.

     

    You are mixing games of chance with games of skill and have your definitions all screwed up.  Chess is a game, not a problem.  For a list of games that are solved (and their levels of being solved), you can look here:  https://en.wikipedia.org/wiki/Solved_game

     

    You are making the flawed assumption that because computers are perfect tacticians that they are unbeatable.  This is simply incorrect (and there is a video on this site where Maxim Dlugy does just that:  https://www.chess.com/video/player/dlugy-defeats-stockfish).  In fact, there are even strategies for beating engines in various time controls (many of the top blitz and bullet players on this very site use those strategies to avoid engine users - as Dlugy describes it in his video, it is effectively avoiding all tactics and playing quiet, positional, suffocating moves).

     

    Also, you are making the flawed assumption that because engines score better than most humans, that they are the "solution".  However, it is easy to see this is not the case when you put an engine against another engine.  The one playing white may start with e4 every game, and get all 3 results (win, loss, or draw).  This clearly demonstrates that chess is not solved from move 1.  In fact, you cannot see who will win after white plays his first move (i.e. a strongly solved game).  In fact, you cannot see who will win at the end of most common (sound - not counting blunders like Fool's Mate, Scholar's Mate, etc.) opening sequences.

    Finally, you misunderstand what "forced win" means.  Put simply, it means no matter what you do, you are forced into a losing position.  This is clearly not the case as you can easily start a game with Stockfish and go through the first 20 moves in a Sicilian variation with completely even chances for both sides (i.e. the opposite of "forced").

  • #72

    Zac

    I will assume everything you said is true. It changes nothing. When people talk about chess being "solved" they are talking about a problem, not a game. A game cannot be solved, but a problem can. You will never ever beat the best computers at chess. Ever. It has solved the problem of playing against you. It has not solved the game of playing against you because part of the game, for you, may be to not win. The game may be to play fast, lose as many pieces as possible, learn, have fun, etc. Thats what makes a game. But the computer doesn't care about the game, it only cares about the problem, which is to win (or not lose). And that is solved. It will not lose 100% of the time. As I said, this is likely to even be more definitive over time since even now in their relative infancy the best computers are about 600 points better than the best human. Imagine 200 years from now what the margin of superiority will be. 200 years from now a 5000 rated computer playing against another 5000 rated computer may figure out a single solution for chess, but until then, the problem of finding a solution for winning (or not losing) against you are anyone else is solved.

  • #73

    Zac,

    I should have looked at the sources you mentioned before I responded. That is my fault. The video that you mentioned of GM Dlugy defeating a computer shows that it could be done. The problem is that he won two of 220 games, 10 years ago. 10 years ago, I'll bet there were a HUNDRED people who could beat the best computers. I doubt that's the case today, or ever again going forward.

  • #74
    lfPatriotGames wrote:

    Zac,

    I should have looked at the sources you mentioned before I responded. That is my fault. The video that you mentioned of GM Dlugy defeating a computer shows that it could be done. The problem is that he won two of 220 games, 10 years ago. 10 years ago, I'll bet there were a HUNDRED people who could beat the best computers. I doubt that's the case today, or ever again going forward.

    Chess is "practically" solved, for all intents and purposes, and much closer to being completely solved than during the pre-computer era, where it was still art.

    I believe one day humans will digest all the computer knowledge and defeat them, or at least pull nearly even with them.  That we haven't done that yet is a sign that the top players are in a very fragile position, protected by mutual incompetence.  It's not unreasonable to expect a 3200-rated human to emerge at some point.  That will be your "next Fischer."

     

  • #75

    fischer would have beaten deep blue!

  • #76
    lfPatriotGames wrote:

    Zac

    I will assume everything you said is true. It changes nothing. When people talk about chess being "solved" they are talking about a problem, not a game. A game cannot be solved, but a problem can. You will never ever beat the best computers at chess. Ever. It has solved the problem of playing against you. It has not solved the game of playing against you because part of the game, for you, may be to not win. The game may be to play fast, lose as many pieces as possible, learn, have fun, etc. Thats what makes a game. But the computer doesn't care about the game, it only cares about the problem, which is to win (or not lose). And that is solved. It will not lose 100% of the time. As I said, this is likely to even be more definitive over time since even now in their relative infancy the best computers are about 600 points better than the best human. Imagine 200 years from now what the margin of superiority will be. 200 years from now a 5000 rated computer playing against another 5000 rated computer may figure out a single solution for chess, but until then, the problem of finding a solution for winning (or not losing) against you are anyone else is solved.

     

    Again, this is entirely incorrect.  For more answers on the same question:  http://chess.stackexchange.com/a/13525

     

    Additionally, the estimate for a "perfect" engine's ELO is currently 3600 (and the highest rating an engine currently has is around 3300 - which should not be confused with the FIDE ratings as they are not exactly comparable).

  • #77
    fewlio wrote:

    Faster time controls is the future.  This years WCC was DECIDED by faster time controls as the longer time controls was a draw.  That should tell everyone everything they need to know.

    Yes as the games are becoming too drawish at the top level of this game the future competition will all about who can think better and faster. A six hours board game is great for players but not for average spectators rapid and blitz  will keep this game alive and funny.  

     

  • #78
    The_Phenominal wrote:
    fewlio wrote: Faster time controls is the future.  This years WCC was DECIDED by faster time controls as the longer time controls was a draw.  That should tell everyone everything they need to know.

    Yes as the games are becoming too drawish at the top level of this game the future competition will all about who can think better and faster. A six hours board game is great for players but not for average spectators rapid and blitz  will keep this game alive and funny.  

     

    It's great for hotels too, since the players fill up any place with an overnight tournament.  Just move them to restaurants in between lunch and dinner if you want a better venue for this.

    It's like basketball's 24-second shot clock, a time period determined when someone figured out that, even without a shot clock, 95 percent of shots were taken within 24 seconds, excluding deliberate stalling.

    Perhaps we could watch some GMs play without a clock and see how long their moves take.  My guess is 5-10 seconds a move, or maybe 3-5 minutes for the time control.

    I'm training at Bullet so that not many can cheat (and those who do are caught), but also because then I just slow down to whatever the proper time control is when I reach my peak.  I already win most "slower" games (even 3 minute) when in time pressure.  Time pressure is "the other endgame."

     

Top
or Join

Online Now