Author: Robert Hyatt
Date: 11:34:31 02/15/03
Go up one level in this thread
On February 15, 2003 at 13:33:39, John C Collins wrote: >On February 15, 2003 at 13:06:55, Robert Hyatt wrote: > >>On February 15, 2003 at 10:30:00, Amir Ban wrote: >> >> >>> >>> Opening preparations: Kasparov is no. 1 in this regard, and he showed what >>>this means in game 1. However, Boris put up an excellent fight and in the second >>>part of the match it appeared that we were getting the upper hand. >>> >>> Tactics: Strangely enough, towards the end of the match I realized that Junior >>>is not outplaying Kasparov tactically. I have seen enough games against >>>grandmasters where this happened to realize that nothing of the sort is >>>happening against this particular opponent. For this reason I was less >>>enthusiastic about game 5 than others arguing that early complications lead us >>>nowhere. >> >> >>A note before I comment: (1) I think you guys did great in the match. I >>can't imagine being disappointed with drawing with arguably the best player >>in the world. (2) I don't think any _good_ chess player is surprised with >>the concept of the computer getting out-tacticaled... Computers excel at >>wide "bushy" tactics with lots of pieces doing lots of things. They are _not_ >>good at narrow/deep tactics, which is where a GM excels. >> >> >> >>> >>> Positional understanding: perhaps even more surprising than tactics, Junior >>>was holding its own in positional play, the best examples of which were games 3 >>>& 4. We have had our experiments in the past with so called anti-GM strategies >>>and with avoiding closed positions and everything else that according to >>>conventional wisdom is supposed to be unsuitable for computers and all these >>>were left in the wastebin for this match. >>> >>> Endgames: This aspect of the game was hardly discussed in this match. >>> >>> In the end Deep Junior played not only as strongly as a super-GM (a debate >>>that has been going in this forum for several years too long), but actually as a >>>super-GM. I've seen opinions expressed that the programs are 2500 in >>>understanding and 2900 in tactics, so you get an overall 2700 performance. I'm >>>not buying it. The same for some comments like "typical computer move" which in >>>some cases were so unthinking that they were seemingly generated by a computer. >>> >> >> >>I disagree with the "played like a super-GM" player, however. I doubt you >>will find _any_ 2200 FIDE player that would play as badly as DJ played in >>the first three games, up until move 30 or so. Game 1 would not have been >>played by any 2000 player I know, myself included. So saying that it has >>super-GM positional understanding is _way_ _way_ offbase. Yes, it played >>good moves at times. But it also played _horrible_ moves at times. And I >>am not just talking about tactically horrible moves such as the blunders that >>Kasparov dropped on the board, I am talking about moves such as taking the >>g-pawn and getting exposed to a horrific attack. >> >>So while I certainly think DJ played very well, I would be hard-pressed to >>think it played like a super-GM positionally. I think its "super-GM" skill >>is great resiliance, the ability to play game after game, at the same "level" >>repeatedly. Compare Kasparov at game 1 vs Kasparov at game 5-6. That is a >>factor I had not considered, and it is a factor that I don't _know_ will apply >>to all human opponents, but at least in watching Kasparov and Kramnik, it seems >>to be a silicon advantage that offets a lot of tactical/positional weaknesses. >> >>But to think that a "super-GM" would play game 1 as black is a stretch I'm not >>willing to make. To think that a "super-resiliant-GM" would play like DJ did >>is perfectly believable... >> >>I think it was an _outstanding_ result, regardless of mistakes made by either >>the program or the human... >> >> >> >> >> >> >>> I believe that the two different ways of playing chess: human chess and >>>computer chess are starting to converge at their highest level. To be sure, the >>>programs are from time to time conceding their silicon origins, but in the same >>>way we humans must admit that when events exceed our understanding, then >>>something mechanical in our own thought processes reveals itself. >>> >> >>I think it might be more related to biological fatigue than anything else. I >>would have hated to play the same kasparov that played game 1, five more games. >>IE the typical golf-cart can't touch a 100M sprinter, but it can keep on >>going until the human drops, and it will eventually win the race. > > > > > > How Deep Junior achieves it's rating is irrelevant, Results are all that >count, the statement that Deep junior did not play like a supergrandmaster is >meaningingless since first we all must agree what a Supergrandmaster plays like, >certainly you don't know, since you aren't one yourself. So you have to be one to understand the idea? In that case, how did the first digital computer get designed? Because the first one certainly came from someone who had not doen it before. I think there are _many_ signature GM characteristics. From simple positional play in the middlegame (and opening of course) to endgame play. I don't think I have to be a GM to recognize moves that are better than what +I+ am capable of playing. Nor do I have to be a GM to see some basic positional flaws in a computer's play. Mine certainly has enough of 'em to keep me busy looking. :) > There is no General >Supergrandmaster playing style, since humans vary as much as computers in the >way they play chess, everyone has a unique style that is almost like a >fingerprint. I disagree. GMs find most reasonable tactics. 2200 players miss quite a bit. GM players find the right positional idea most of the time. 2200 players miss quite a bit there _too_. > Answer me this, What Difference Does it make if you play more >positional chess, if you cannot defeated me?? If that were the case, I would agree. But by the same token, do you want your program to play 30 brilliant moves and one lemon move, over and over? That one lemon will drag your performance _way_ down at the top of the rating scale. >Then the question becomes is your >positional chess more important then my tactical chess? How many people play >chess for pure aestetics? Most play to Win and Achieve better results, and a >Higher rating. So whatever or not Computers mimick the play of Humans is largely >irrelevant, finally explain to me why it is so Important for computers to >emulate the style of humans? Isn't the final Goal of Chess to Win the Game? I don't quite follow the "tactics" issue. Are you saying that computers are so tactically superior to humans that positional play doesn't matter? If so, I certainly don't agree. Look at the exchange sac Kasparov offered in game two. Did any computer realize how bad that was for white to accept it? None that were giving analysis understood, yet Kasparov had calculated it much farther and saw that he was generally better, where before the sac he was certainly worse.
This page took 0.01 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.