Author: blass uri
Date: 12:25:09 07/18/00
Go up one level in this thread
On July 18, 2000 at 14:53:35, Graham Laight wrote: >On July 17, 2000 at 19:18:27, Amir Ban wrote: > >>On July 17, 2000 at 18:13:04, Graham Laight wrote: >> >>>On July 17, 2000 at 16:09:09, Amir Ban wrote: >>> >>>>On July 17, 2000 at 07:22:41, Graham Laight wrote: >>>> >>>>>I'm afraid I still feel that Junior could have come out ahead (instead of >>>>>level)in this tournament by beating Bareev and Khalifman - and possibly by not >>>>>losing with such apparent ease to Kramnik. Continuing the game against Anand >>>>>might possibly have gained an extra half point as well. >>>>> >>>>>I think that Amir has an aspiration to make his program demonstably better than >>>>>Deep Blue (this certainly comes across in his interviews published on the >>>>>Chessbase Website coverage of Dortmund (www.chessbase.com) before the Kramnik >>>>>game). If so, as a (hopefully!) impartial member of the viewing public, I'm >>>>>afraid to say that I've yet to be convinced. >>>>> >>>>>As evidence, I point firstly to the games against Bareev and Khalifman. On both >>>>>occasions when Deep Blue '97 gained an advantage over Gary Kasparov (who's a >>>>>better player than anyone at Dortmund was), it parlayed that advantage into >>>>>victory - whilst Deep Junior twice failed conspicuously to "slam in the lamb". >>>>> >>>>>I would also point to the game against Kramnik. Here we see Deep Junior lose >>>>>to a combination of blocked centre and king attack - classic anti computer >>>>>methods which have both been well known for a long time. They work because, in >>>>>this case, nothing short of truly massive search depth is going to help you to >>>>>make the correct moves. >>>>> >>>>>However, for both king attack and blocked centre, Deep Blue '97 demonstrated >>>>>that it's evaluation knowledge was able to adequately handle the challenge. >>>>>Indeed, in game 2 in '97, Deep Blue not only handled the blocked centre, it >>>>>turned it into a win! >>>>> >>>>>It took Deep Blue 2 attempts to beat Gary Kasparov, the world's best player - >>>>>maybe another year of work will push Deep Junior to a position where it can try >>>>>to win these tournaments, instead of settling for a middling position. >>>>> >>>>>But let's not be completely churlish - Dortmund 2000 was indeed a fantastic >>>>>performance by Deep Junior - and a landmark in computer chess history, since >>>>>here is both a computer and a program which one can buy in the shops! >>>> >>>>I disagree with most of this, but it's your opinion, and if experience teaches >>>>us anything, it's useless to argue. >>> >>>I _do_ change my mind about things. For example, earlier this year there was a >>>debate about whether the SSDF ratings were inflated. >>> >>>Here are the top 3 computers on the current (April 2000) list (using 450 MHz): >>> >>>Computer Rating Error Margin >>>======== ====== ============ >>> >>>Fritz 2721 + - 38 >>>Junior 2689 + - 30 >>>Tiger 2671 + - 32 >>> >>>I admit it took quite a debate, but I'm now willing to concede that, compared to >>>FIDE ELO ratings, these ratings are too high. Also, I am now willing to admit >>>that a stronger performance against other computers does not necessarily mean a >>>stronger performance against top human players. >>> >>>Sometimes, some people ARE willing to listen to what the others are saying! >>> >> >>I did not mean you in particular, and no offense meant. Experience tells that >>everything about Deep Blue's career of 12 games has already been said. Everyone >>made up his mind one way or the other, and that's it. >> >> >>>>For the record, I'm not trying to prove that I'm better than Deep Blue. I think >>>>I've already shown this some time ago, and I'm not the only one who can say so >>>>either. >>> >>>This may be true - but if it is than I, trying to be as impartial as I can be, >>>have to honestly say that it's not clear to me. >>> >>>>Looking at the (very few) games of DB, I don't see that it had either better >>>>evaluation or deeper search than today's top programs. >>>> >>>>Amir >>> >>>Fine - but at the risk of being repetitive, let me briefly restate the evidence >>>that points to it having better evaluation (and probably a deeper search): >>> >>>* It won a 6 game exhibition against the greatest chess player in history >>>(Junior came "equal" in 9 games against lower rated opposition) >>> >>>* Junior failed to win 2 games in which it had a significant advantage, wheras, >>>in 1997, once it was ahead, Deep Blue never failed to "slam in the lamb". >>> >>>* Against Kramnik, Junior fell heavily to the blocked centre, and the king >>>attack. Deep Blue actually won one of its games from a blocked centre position >>>(game 2, '97). >>> >>>As a mere hobby observer, you can surely see why I might not feel that Junior's >>>equality (or improvement) in evaluation or search depth have been clearly >>>demonstrated to me. >>> >> >>The answer is: look at the games, and study the moves. If you understand chess, >>and you understand how computers play chess, this can tell you a lot about the >>quality of the evaluation and how deep the program gets. I'm not saying >>everybody is qualified to do this, you obviously need to be some sort of expert >>to be able to judge. >> >>A grandmaster can identify a strong chess player by looking at his games. Do you >>believe he bases this judgement only on the result ? >> >>By the way, getting an advantage is quite an achievement in itself and you must >>play well to get it (unless it's handed to you as a free gift as in the DB match >>6th game). It's a fact of life in chess that not all advantages end up as wins, >>and this doesn't reflect on the level of the player with the advantage. What do >>you think of a game where you have a disadvantage and win, as against Leko ? > >Unfortunately (?!?), my girlfriend did not allow me to watch the game in full, >but from what I did see, Leko appeared to sacrifice material to gain a strong >attack. The GM commentator (Roman, I think) said that this was a reckless thing >to do against a computer, because they will defend against attacks in this type >of position very well. Therefore, the impression I got was that Leko decided to >gamble on a win so that he could possibly get (or share) 1st place. Clearly, in >the event, the gamble failed - and the price he paid was to lose the game. > >It may be that, between the sacrifice, and the result coming in, Peter did >obtain a winning position from which Junior had to fight back. > >>Also by the way, most of Junior's games in Dortmund were closed positions, not >>only the 2 losses. In at least two other games (Bareev, Akopian), it fought for >>a win, but got only a draw. This is a good description of what happened to Deep >>Blue in the 2nd game. > >But none of the grandmasters there knew it was a draw! In fact, none of them >even thought there might be a draw - the position looked so terminal that they >just assumed that Gary was correct to resign. I think that the opposite happened. The GM assumed after Gary resigned that it must be a loss for Gary because they believed that gary will not resign a draw position. Gary did not know deeper blue so he assumed that it is impossible to outsearch it so he even did not check the possibility of sacrificing a piece. Uri
This page took 0 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.