|
< Earlier Kibitzing · PAGE 6 OF 11 ·
Later Kibitzing> |
| Dec-25-04 | | csmath: This is a classic anti-computer style. Interesting that Kasparov adopted that since he used to be a bit more aggressive against computers but some bad experiences forced him to try to salvage his reputation using known recipes how to play against a machine. ;-))
This is not very exciting nor inovative game. |
|
| Dec-28-04 | | Kingsider: I can name that opening in three moves. ha |
|
Dec-30-04
 | | LIFE Master AJ: Black resigned because the Fritz team clearly saw the end coming ... as did their box. (I watched this game live on TV ... it is also deeply annotated on my web site.) |
|
Feb-12-05
 | | mahmoudkubba: Is crafty a machine?? |
|
| Feb-12-05 | | hintza: <mahmoudkubba> <Is crafty a machine??> Yes, Crafty is a chess engine which I think is available somewhere for free download, but I am not sure where. |
|
| Feb-12-05 | | csmath: At chessbase.com.
I used to play with it, it is pretty good but not as good as the retail engines like Shredder, Fritz, Hiarcs, or Junior. As far as Kasparov goes, I believe that today he would lose any match of say 10-12 games against a strong machine (good hardware). This particular match he held to a draw only by applying anti-computer chess strategy which is obvious in the position of this game. He learned the lesson from the past. There are more powerful computers today than the one he played here, and better programs too. Computers are better now. |
|
| Feb-12-05 | | aw1988: It doesn't really matter. Computers are supposed to be used as an analytical tool, not a main competition. It is like asking God to kill three people for you. Sorry, I promise I won't do that again. :P |
|
| Feb-12-05 | | WillC21: One major question that arose during the game was whether around move twenty-two black would play f5. Black's only chance of survival was to counterattack. Instead, Fritz blocked it's f-pawn with Nf6. Some say this is an unbelievable game and worthy of praise, but I beg to differ. Sure, Kasparov did what he had to do, but instead of an "unbelievable" game it really is just an example of a terribly confused machine not knowing how to strategically react to a closed position. While Fritz X3D was strong but not strong enough, I think on powerful hardware the most current versions of Shredder and Hydra could defeat any of the world's top grandmasters in a world championship format match. |
|
| Feb-12-05 | | csmath: Will, Kasparov played anti-computer chess (blockade position) which is what is known to work against machines. One needs to know how to do that, he apparently learned. On ICC you can occassionally run into people that are nowhere of Kasparov's stature but they play successfully against a computer using this type of approach. I am sure Kasparov wasn't too proud to do that but he had to salvage the match since he was losing it at this point. I do not know to play anti-computer chess (I tried though ;-))) but I can recognize when people do play that. This is what this game is about. Shuffling knights done by machine here shows you what happens when you successfully create a blockade. Machine just doesn't know what to do, it simply makes the "best" move according to algorhitm but has no plan for a long term. And since everything is blockaded there is no immediate tactical chance either, you need to create the position for it. This can be avoided, meaning you can tune program to avoid this. I too believe that Kasparov today would lose a match against Shredder 9 on quad Opteron 852. And more, since machine cannot lose concentration any longer match (more than 4-6 games) would be a disaster for a human opponent. |
|
| Feb-12-05 | | Minor Piece Activity: I don't think that Shredder 9 et al could beat the top players in a match today. Avoiding closed positions is not a bullet proof strategy because sometimes the best move is the move that leads to a closed position. If Shredder avoids the best moves in such cases, it will easily compensate and give humans more leeway. Maybe comp evasion of closed positions could even be turned into anti-comp strategy later on. Anyway in both Kramnik-Fritz and Kasparov-Junior, the humans played aggressively (yes including Kramnik) in all the games, something they would not do if they felt the computers were superior. |
|
| Feb-12-05 | | csmath: You wouldn't think so?
Well, Fritz and Hydra have beaten Topalov and Ponomariov just recently in a combined match. A laptop strength PC has beaten Karjakin. What more proof you need? |
|
| Feb-12-05 | | csmath: Kramnik and Kasparov played 4, 6 games match respectively. They played against computers that are not extremely fast. Kramnik did play aggressively in one game, he lost that game after calling it "it could have been the best game in his life" or something like that. They both drew. In case of Kramnik this was 3 years ago (keep in mind how much faster computers are today, I have computer that is twice faster than anything I had 3 years ago). |
|
| Feb-12-05 | | Minor Piece Activity: If you think an out of practice ex Fide champion and a cute little kid are the best players in the world you might be a sandwich short of a picnic. ;-) And Topalov did not do bad. You do not remember his score? 3 draws and a loss. That's not bad at all especially since he played his usual style. No silly anti comp stuff. |
|
| Feb-12-05 | | Minor Piece Activity: < Kramnik and Kasparov played 4, 6 games match respectively> No, they played 8 and 4/6 games match respectively (Fritz/Junior). Kramnik was in a commanding lead until he blundered away everything in the second half, drawing the match. So was Kasparov in the Junior match. Against Fritz he lost because of some dumb error in the Berlin which he rarely played, overlooked a 2 move combination. |
|
| Feb-12-05 | | Minor Piece Activity: <keep in mind how much faster computers are today> There are diminishing returns for hardware improvements you know. 1 gigahertz more isn't very much when your ply is like 20. |
|
| Feb-12-05 | | csmath: Yup, Kramnik played anti-computer chess in games 2 and 3 he won. And then he blundered away 5th game (that is what humans do ;-))) and then was also soundly defeated in the best game of the match, 6th. Topalov didn't do that bad? ;-)) According to you. Is that a acknowledgment that he couldn't simply so any better?
Yeah, he just lost a game and drew three. Ponomariov is one of the top 10 players in the world, I don't see he would be out of form any more than any other top 10 player. Karjakin just won Corus B meaning he isn't a weak player by any standards. |
|
| Feb-12-05 | | Minor Piece Activity: Ponomariov refused to play for a long time after he was stripped of his title. When he returned he played terribly. Did you see his games? Karjakin is nowhere near his peak and not truly world class yet. His rating isn't even above 2700 and he just got obliterated in a match with Naka. Yes he's a top player... for his age group... Maybe we should start calling Nakamura "Fritz" instead. Beats the corny "Star Wars." Top players = Vishy, Kramnik, Leko, Kaspy. |
|
| Feb-12-05 | | WillC21: First, any of the top five players in the world on a very strong day, or their best day, could defeat Hydra, Shredder, or Fritz. But I think over a series of matches, like a WC format cycle, the humans performance would go down as the match progressed. With two humans this fatigue effect can cancel out, but with human against computer this is not so. Also, results of Kramnik and Kasparov versus Fritz two years ago mean little, as now both of these engines have improved at a rate faster than that of all of the world's top five players. Plus, Hydra is now on the scene, after appearing nearly unbeatable in its recent win over Shredder. I think it is fair to conclude Hydra, in a WC cycle match, would beat any player in the world, without exception. |
|
| Feb-12-05 | | WillC21: By "both of these engines," I mean Shredder and Fritz. |
|
| Feb-12-05 | | csmath: Well, let me inform you a bit that Anand has played match against computer program Rebel back in 1998 and lost 5-3.
Kasparov does not have a terribly good record against computers either. I'd like to see Leko, I would bet money on Shredder beating Leko, on my machine. |
|
| Feb-12-05 | | Minor Piece Activity: <Topalov didn't do that bad? ;-)) According to you.> He did bad? According to you only. ;) Remember that he scored the same in Bilbao as Junior, which won the comp championship. Must mean that computers really "suck." <soundly defeated in the best game of the match, 6th.> Did you read the anlysis of the game? Kramnik resigned in a complicated but drawn position. Ironically it is like when Kaspy played Deep blue and missed a perpetual check at the very end. |
|
| Feb-12-05 | | Minor Piece Activity: <Well, let me inform you a bit that Anand has played match against computer program Rebel back in 1998 and lost 5-3.> 6 of those games were blitz. The last two were classical. Results of the last two games? Vishy-Rebel 1.5-.5. Such a tragedy, Vishy only won by a point! How Terrible! |
|
| Feb-12-05 | | csmath: <<Did you read the anlysis of the game? Kramnik resigned in a complicated but drawn position. Ironically it is like when Kaspy played Deep blue and missed a perpetual check at the very end. >> You have absolutely no clue what you are talking about. Kramnik vs Deep Fritz, 2002
This is a totally lost position, my Shredder analysis shows -5 for black. There is no way anybody, Kramnik and Kasparov combined could save this position against a computer. As simple as that. |
|
| Feb-12-05 | | WillC21: Good point by Minor Piece Activity. I do recall from the website at the time that many of those games were blitz and rebel actually got bested in the classical games. However, that was in 1998, so please see my above posts concerning Shredder/Hydra/Fritz as they apply more to today. |
|
| Feb-12-05 | | Minor Piece Activity: <Kasparov does not have a terribly good record against computers either.> He has a plus or even score with every famous comp, including Deep Blue. |
|
 |
|
|
< Earlier Kibitzing · PAGE 6 OF 11 ·
Later Kibitzing> |
Kibitzing
