|
< Earlier Kibitzing · PAGE 1 OF 2 ·
Later Kibitzing> |
| Jun-13-03 | | dev321anand: great game |
|
| Oct-18-03 | | mj29479: the depth of 22...Qxc3 is really superb not only it is a good tactical move but what is amazing about it is the calculaton that goes as far as 33...R8c3+.Ivanchuk might have thought 21...bxc3 as a bluff buth it was certainly not one. |
|
| Oct-18-03 | | mj29479: what i was looking forward to was 21...bxc3.22.Kb1 Nd5 23.Rxd5 cxb2 24.Re3 Ba3 25.Ra5 Bb4 26.Rd5 though the blacks with advantage but still an open game, or may pull it off as a draw. |
|
| Oct-19-03 | | drukenknight: you really think these guys see 15 moves deep? What is the follow up to: 26 Rxh4? HOw deep do you have to see that a pawn capture is matched by a pawn capture, a R for a R. Isnt that what you are supposed to do when you are ahead in material? |
|
| Oct-19-03 | | refutor: i guess you're trying to spark a debate on the chess vision of these super GMs...i'll bite...not only can they analyze a position much better than you, they can also calculate twice as far as you :) in all seriousness, they can only see about 4 or 5 moves, longer if there are forcing moves in there. some can see further, some can't. the real strength of a GM lies in his/her ability to analyze a position. what do you think they're doing when they're sitting there for 20 or 30 minutes on a move? calculating variations? no they're analyzing the position and deciding on the best strategy |
|
Oct-19-03
 | | Eggman: Regarding the calculating ability of elite GMs: do you remember Tony Miles’ famous comment, after losing a 1986 match (-5 =1 +0) to Kasparov? He said “I thought I was playing a world champion, not some monster with 28,000 eyes who sees EVERYTHING.”
As for strategy, Kasparov, Kramnik, Anand, etc. are time and again encountering positions so familiar to them that I doubt they have to think too much in terms of evaluating and coming up with plans, etc. Much of this is done at home ahead of time. Even an unfamiliar position can be fairly quickly assessed by comparing it to a familiar one and noting the significance of the differences. I remember Shirov a few years ago saying that he is able to calculate so far ahead that pure strategy practically doesn’t exist for him, and that his style is more the product of concrete analysis of variations. I also distinctly recall Seirawan talking about once looking 30 moves ahead in an endgame. Indeed, why ask if a move makes sense in strategical terms when you can calculate many variations 10 moves deep and see that the move definitely leads to an inferior position? And what better way to win game after game after game than to calculate very quickly, thoroughly, and accurately, thus living little to chance? Naturally one has to evaluate the final positions at the end of one’s calculations, so no, it’s not all about calculating, but no doubt calculating predominates. As for the idea that GMs are only seeing 4 or 5 moves ahead, or longer for forcing variations, I have to say this is way, way, way off the mark. To me that sounds like a 1700 player, not a GM. How far ahead these guys look depends on the position, but when you see a GM sitting there thinking for 30 minutes, you’d better believe he’s (she’s?) calculating variations. If you don't trust me to speak on behalf of these super-GMs, then let me say this from my own experience: once when I was about 2100 strength I spent 35 minutes on an endgame puzzle, and after finding the solution I went back and out of curiosity counted how far ahead I had seen. It turned out to be 23 moves. Admittedly, there were only a few pieces on the board, but still if someone 2100 can do that ... |
|
| Oct-19-03 | | refutor: <35 minutes on an endgame puzzle...23 moves> an endgame is very different from the middlegame type positions that drukenknight was talking about. i doubt if kasparov checks more than 4 or 5 moves ahead when looking at a general middlegame position. if it is sharp or contains checks or captures, no doubt he looks a little deeper. for instance look at the position in Geller vs Gligoric, 1953 after move 16...is there any reason to look deeper than 4 or 5 moves in a position like this? or is it even possible? both sides have a number of almost equally good choices at every turn here, nothing is forced. |
|
| Oct-19-03 | | John Doe: I don't know... Strategic play often seems like deep calculations(how did he know to place his knight there and thus foild my mate in 8!)
Well, it was a good place for a knight? |
|
| Oct-19-03 | | Diggitydawg: <Indeed, why ask if a move makes sense in strategical terms when you can calculate many variations 10 moves deep> Because the clock is ticking and it takes time to calculate. I think that GMs look only a few moves ahead in general but when the game comes to a critical juncture, (you can see this when they take 20-30 minutes on a move) they are calculating like demons. They may have a few plans in mind which requires calculation each 4-5 moves deep. Or it may be one concrete plan that requires looking 10-30 or more moves ahead. I'm sure it's also quite exhausting to calculate so deeply so GMs would want to save their reserves of energy strictly for the critical points of a game. |
|
| Oct-19-03 | | Benjamin Lau: My guess is that a GM looks ahead about 4-5 moves during most of the middlegame, except when he/she sees a possible combination, mate in X>5, or a speculative positional sacrifice. |
|
| Oct-19-03 | | drukenknight: do you think that guy that sees 23 moves ahead could post a follow up to Rxh4? |
|
| Oct-19-03 | | ksadler: "...Asked how many moves ahead he can think, Kasparov replied that it depended on the positions of the pieces. "Normally, I would calculate three to five moves," he said. "You don't need more.... But I can go much deeper if it is required." For example, in a position involving forced moves, it's possible to look ahead as many as 12 or 14 moves, he noted...." From: http://www.maa.org/mathland/mathlan... |
|
| Oct-20-03 | | Calli: "What is the follow up to: 26 Rxh4?"
Only need to see one move ahead! After 26...0-0, the position is pretty much the same as the game. |
|
| Oct-20-03 | | drukenknight: It IS isnt it?
Hmm, chess is a funny game. Did you ever lose a game by one move? Hmm come to think of it most games are lost by one move. Just one more move. ONE MORE and I would have won. But instead I lost. Gee, one move. It seems so insignificant.
So the R is on one square different from where he was in the original game? Hmmmm.... |
|
| Dec-29-04 | | Whitehat1963: "I only look one move ahead, but it's the right move." Jose R. Capablanca |
|
| May-07-05 | | Karpova: Kotov says that you should try to find the right strategy while your opponent's clock is running and calculate variations while your clock is. It's not that clear if Capablanca ever said < only look one move ahead, but it's the right move.>
some people say it was lasker, etc.
If you read his <Chess fundamentals> it gets pretty clear how Capablanca thought. Amazing game, btw. |
|
| May-07-05 | | Kingdumb: My theory on looking x number of moves ahead has always been, how can you really do it? Nobody can predict a person's reply to a move unless it is a forced move. If Kasparov tried to calculate 5 moves ahead against me he'd have a stroke as my play wouldn't be anything close to what he'd expect. All he can do is look for the best possible chances for him based on me making what he believes is the best move. If I make a different move then it's back to the drawing board. Granted, GK can probably beat me blindfolded anyway but my point is there is just no way to "guess" what move someone will make when theoretically there are so many possible moves for each position. Think about it. How many times while watching these games live were you sure of someone's move to be totally taken by surprise by a move you never expected. A good example is Topalov's Nh4 move in the last game against GK just before GK retired. Everyone thought he was out of his mind with that move. Even Fritz couldn't figure it and most people has Topy written off, including myself. Funny how he ended up winning that game. Nh4 turned out to be a key move in the game. When I play I try to look a few moves ahead based on my plan, not what I think someone else is going to do because 9 times out of 10 I guess wrong and I'm not a strong enough player to go through a hundred different variations when the clock is running. I'd rather go down from a blunder than go down because I ran out of time. The debates will go on I'm sure. Guessing someone's next move? I'd sooner take my chances playing the lottery. |
|
| May-07-05 | | WillC21: <Kingdumb><Guessing someone's next move? I'd sooner take my chances playing the lottery>. And that's why you are a patzer. |
|
| May-07-05 | | Kingdumb: Thank you for your kind remarks Will. That's what I love about the people at this site. They're all so friendly. |
|
May-07-05
 | | tpstar: <Kingdumb> Friend, let's spin this topic into a different direction. Predicting your opponent's next move is not guessing; it is calculating. As you improve at chess, your ability to objectively assess each position also improves. Therefore you will make the "best" move more often, plus you expect (anticipate) the "best" response more often. (If these move choices are not "best" yet, at least they should be sound. ;>D) Simultaneously, you are completely surprised by your opponent's response less often. Trick moves on either side will stand out in your memory, but more often the bump and grind of chess (based on both sides playing the "best" move) goes unnoticed. I'll bet these pros see at least 3-5 moves ahead all the time. You and I cannot, at least not yet. And that's why we are amateurs. =) |
|
| May-08-05 | | aw1988: Speak for yourself. Ahem. |
|
| May-08-05 | | Karpova: predicting your opponent's moves should be calculating not guessing. if either you or your opponent is really bad it might be guessing but you should always consider your opponent to make good moves.
it's hard if you go for secure lines because a little sac or a combo may have flaws in it but after the game is finished and you analyze with your opponent you may experience him to have no idea of what was going on in the position at all... so it's up to you to decide if you want to go for dubious lines hoping for your opponent not to find the right defense or consider him to be at least as good as you are. |
|
| May-09-05 | | Kingdumb: Well, I guess everyone missed my point. Most likely because I didn't state it clearly enough. I'm not suggesting that we don't try to "calculate" our oppenents next move. Believe me, I do. All I'm saying is tbat we cannot with 100% certainty know that the move we're counting on is going to be the move played. If that were the case there would be no reason to play the games. We just state our openings to each other, play book for the next 10-15 moves and then when someone decides to break out of book just reference a database to see who's going to win the game based on what moves "should be" played next. But it doesn't work that way. There will always be surprise moves no matter how good you are and no matter how well you think you can calculate, especially in the more complicated positions where I've seen Fritz calculate as many as 30 possible continuations, all of them being viable. As humans I seriously doubt many of us can calculate that many possible variations. Maybe GK and a few others can but I'm sure the majority of us can't. Please understand, I am not suggesting we abandon the attempts to calculate our oppenents moves. All I'm saying is don't be surprised when he doesn't live up or down to our expectations. |
|
| May-10-05 | | Karpova: <Kingdumb>
Very intersting point though quite obvious. It's useless to discuss it since it would take a mindreader to guess the opponent's moves 100% correctly. |
|
| Jul-14-05 | | nikolaas: What about 22.exf6 Qxh2 23.Rxd7 Kxd7 24.fxg7 Bxg7 25.Qxf7? |
|
 |
|
< Earlier Kibitzing · PAGE 1 OF 2 ·
Later Kibitzing> |
Kibitzing
sac: 22...Qxc3