< Earlier Kibitzing · PAGE 8 OF 9 ·
|Aug-21-15|| ||anema86: <RookFile>, are you kidding? People don't play over games anymore. They just plug the moves into an engine and see what the engine says, then often parrot it back like it's their idea (see a comment further above). Self-analyzing games is obsolete.|
I'm being facetious, obviously. I see very little value in computer analyses except as confirmation and refutation for my own self-analyses. The vast majority of people prefer to just bow to the engine without knowing or caring why something is a good or bad move. "The engine says so" is good enough for them.
|Aug-21-15|| ||AylerKupp: <anema86> Well, in the end, what is analysis for other than confirmation or refutation of either your own or others' self-analysis? And I suspect that the vast majority of people just bow to a grandmaster's analysis without knowing or caring why something is a good or bad move. "The grandmaster says so" is good enough for them.|
So how are engine analysis and grandmaster analysis all that different? I'll tell you (of course I will!): <Properly conducted> engine analysis is more accurate. The key is <properly conducted>. The analysis should be of sufficient depth to have reasonably good confidence in the result, there should be a human review and/or forward sliding to guard against horizon effects, and it should be performed using multiple engines since each engine has a unique evaluation function and their evaluations could be either a little or a lot different. And, if a low-piece count endgame is within the range of the search depth, tablebases should be used to get a definitive assessment of the result.
Take, for example, Spassky vs Fischer, 1972 and the infamous 29...Bxh2. Several grandmasters, including R. Byrne and Larsen but most notably Speelman in his "Analyzing the Endgame", analyzed the endgame after 29...Bxh2 extensively (without the help of computers and chess engines; good ones were not available at the time) and concluded that Fischer could have held the draw with best play. But they were all wrong. Computer analysis (mine, with a "minor" assistance from Rybka) showed that after 29...Bxh2 Fischer was lost against best play by Spassky (see analysis starting at Spassky vs Fischer, 1972 (kibitz #455)).
That is the key however, "best play by Spassky". It's easy (though time-consuming) to sit in a chair in front of a computer, have it analyze line after line, review the results, and document a conclusion. But players OTB don't have that luxury, no matter how good they are they make mistakes, particularly understandable when in time pressure. So there is no guarantee that OTB Spassky would have always found the best moves.
But that works both ways. There is no guarantee that Fischer would have always found the best moves OTB either. Indeed, being continually on the defensive and struggling to get a draw, is almost a guarantee that the best move in each position will not always be found. And, indeed, some of Fischer's moves were criticized as throwing away the draw. But the draw was not there in the first place - the computer says so.
|Nov-14-15|| ||alisog: Shift the crucial black elephant takes the horse in c3|
|Jan-02-16|| ||Dionyseus: <AylerKupp> My computer analysis shows that Fischer could have drawn the first game with 31...Ke7. I posted my analysis at Spassky vs Fischer, 1972|
|Jan-02-16|| ||Howard: Just looked at it....doesn't look like very computer-like analysis to me.|
|Jan-02-16|| ||Dionyseus: <Howard> <Just looked at it....doesn't look like very computer-like analysis to me. >|
I assure you it is. I have experience as a correspondence chess player. If you find any improvements for white in my line let me know, and if you'd like to play against my 31...Ke7 line I'd gladly do so. We could play at the rate of a move a day if that's alright with you.
|Aug-17-16|| ||maseras: 11...Nh5?!
nice. In my opinion, fischer has already knew that 11...Nh5?! is too risky. 12.Bxh5 gxh5 13.Nd1! or 13.Kh1 could be the end of World Championship Match for Fischer.
|Aug-17-16|| ||RookFile: Yes, Gligoric found this recipe, it makes black look bad. Here's one example:|
Gligoric vs Browne, 1972
I think that Fischer had played over so many games of Spassky's that he started to think like Spassky. For that reason, Nd1 didn't occur to Fischer because it was not the type of move Spassky would play.
|Aug-17-16|| ||RookFile: Here's another:
Gligoric vs Kavalek, 1972
|Aug-17-16|| ||Petrosianic: Clearly if he played Nh5, he did not believe it was not too risky to play.|
|Aug-17-16|| ||Howard: Mednis once wrote in Chess Life that various players "chipped away" (that was the expression he used) at Fischer's strange knight move over the next several years.|
|Aug-17-16|| ||RookFile: It's not like Spassky was an idiot. He played Nc4 and Ne3 with the idea of controlling f5. Evidently, one of the virtues of the unexpected Nd1 is that it allows (after a4) Ra3 and if needed, the rook on a3 can swing over to the kinside. The other is that white can also play the bishop to c3 and challenge the g7 bishop (this happens all the time in the Benko Gambit).|
|Sep-13-16|| ||ZonszeinP: But he (Spassky) was playing that game under very odd conditions.
To put it mildly|
|Dec-02-16|| ||Grbasowski: Unbelievable 11th Fischer's move!|
|Dec-02-16|| ||Petrosianic: Not good, just unbelievable.|
|Dec-02-16|| ||steinitzfan: I remember when this game was played. It was Fischer's first win against Spassky -- and what a game! For some reason, turning that 2-0 to 2-1 changed my whole outlook.|
|Dec-02-16|| ||Howard: But, wasn't 11...Nh5 later found to be overrated? Objectively, Fischer had better options, but it was the surprise value of that move that made it work---not the accuracy of the move itself.|
|Dec-08-16|| ||An Englishman: Good Evening: Do you think we shall ever see another Benoni in a World Championship match again? An Alekhine? A Pirc? This match had them all, and the variety of openings stands in stark contrast to Carlsen-Karjakin. Not that one can say this is a bad thing; it just seems odd that the last match seemed so fixated on just a few openings.|
|Dec-08-16|| ||Petrosianic: <Howard: But, wasn't 11...Nh5 later found to be overrated? Objectively, Fischer had better options, but it was the surprise value of that move that made it work---not the accuracy of the move itself.>|
Not just the surprise value of the move, but the fact that Fischer got into a shouting match with Lothar Schmid while the game was in progress must have rattled Spassky at least a little. His opening play is very shaky, with moves like 15. Bd2 and 18. g3. He still manages to hang in for a long time, but he's got the short end of it all the way.
|Dec-08-16|| ||Olavi: While 11...Nh5 made a big impression, it wasn't such a novelty. An earlier game, a spiritual forerunner I think, was Timman vs Ljubojevic, 1972|
And in the Kings Indian similar moves had often been seen. Szabo vs Boleslavsky, 1950 first comes to mind though that one is not so close.
|Dec-08-16|| ||offramp: <An Englishman:...the variety of openings stands in stark contrast to Carlsen-Karjakin. Not that one can say this is a bad thing; it just seems odd that the last match seemed so fixated on just a few openings.>|
I thought <most> world championship matches were limited in their openings. The QGD is knpwn as the world championship opening.
The most opening-fixated match I can remember is not a world championship but a Candidates' Final. Can you guess?
It is the 1984 Candidates' Final between Kasparov & Smyslov. You can see the games at this excellent but unusually-monickered collection by User: whiteshark:
Game Collection: 99999_Kasparov-Smyslov 1984 Candidates final
|Dec-08-16|| ||perfidious: As I have noted elsewhere regarding this match, the great variety of openings adopted was a practical necessity for Fischer, given that he was facing the might of the Soviet chess establishment and its immense ability at ginning up theoretical innovations. As it was, some of Fischer's specialities came in for a hard time; see the fourth and eleventh games of the match.|
|Dec-09-16|| ||An Englishman: Good Evening: Excellent point about the Soviet chess machine <perfidious>, but they could not have felt any surprise concerning the Alekhine or the Benoni, given that Fischer first played these opening in the middle Sixties. He first played the Pirc in this match, and *that* must have shocked them, given his ridiculous success vs. 1...d6 and 1...g6. <offramp> is certainly right about the Alekhine-Capablanca match in terms of using the same opening, but in Botvinnik-Bronstein the players used different openings for the first six games!|
|May-02-17|| ||Wedontgetchess: I think it's hilarious how everyone is claiming that Nh5 is a mistake or an inaccuracy. Now in 2017 with stockfish 8, my computer claims Nh5 is the Best move with the other top moves being your regular mainline a6 and Ne5. So even if Nd1 is the apparent refutation (Stockfish 8 apparently doesn't like it) Nh5 is still the best move. Honestly I just hate engine analysis on positional moves such as Nh5 and wanted to see if Fischers move was actually not that good... apparently it is so that's that I guess. (If other moves are played first like a6 a4 or the other mainlines then yes Nh5 is an inaccuracy; subtlety and move order matters apparently).|
|May-02-17|| ||Helios727: What happens after 42. Ke3 ?|
< Earlier Kibitzing · PAGE 8 OF 9 ·