|
< Earlier Kibitzing · PAGE 204 OF 254 ·
Later Kibitzing> |
| Mar-25-12 | | Olavi: Well the Chessbase feature was based on a 13 ply search. It's debatable whether such an 'analysis' indicates anything at all, it certainly doesn't indicate whether a player manages to steer a game into a direction, which suits his/her particular talent. One can easily win lots of games (more) while playing 'worse' by this 'criterion'. |
|
| Mar-25-12 | | twinlark: 13 ply is wholly inadequate depth at which to analyse GM play. You need at least 18 ply, preferably more to get anything like reasonable results. 13 ply is the starting point within a couple of seconds for modern engines, and at that level is good only for blunder checks. |
|
Mar-26-12
 | | alexmagnus: Well, wasn't recently a study published suggesting that the search depth doesn't affect the results too much? |
|
Mar-26-12
| | alexmagnus: The result of the game analysis (with a sufficiently large game pool), that is. |
|
| Mar-26-12 | | twinlark: Think about it. This is saying that the first few seconds of an engine analysis will ultimately be sufficient to accurately describe GM play to several decimal places. When we watch live games, we know how inaccurate engines can be even if they run for a few minutes. I'd be interesting in seeing this article, as it sounds like so much horse's patootie. |
|
Mar-26-12
| | alexmagnus: < When we watch live games, we know how inaccurate engines can be even if they run for a few minutes.> Yes, but it's like in one move per 10-20 games... Not in the entire games. Gms don't think that far, according to Kasparov a normal thinking depth at this level is 4-5 moves (8-10 ply, that is), only in combos and endgames does one think longer (Kasparov: "my longest combo was 15 moves"). 13 ply is 6.5 moves. OK, computers are worse when it comes to the evaluation, but still... I meant this one: http://www.chessbase.com/newsdetail... |
|
| Mar-26-12 | | twinlark: <alexmagnus>
Thanks, I'll have a read.
<Gms don't think that far, according to Kasparov a normal thinking depth at this level is 4-5 moves (8-10 ply, that is), only in combos and endgames does one think longer (Kasparov:> Yeah but...we established long ago that humans are not calculating machines and that many operate from a profoundly developed sense of pattern recognition. This makes up for actual relative lack of computation compared with engines, as the heuristics in the GM's brain pick and choose from thousands of patterns stored in the mind, enabling an evaluation that quickly eliminates many variations. Call it an informed intuitive leap that is the equivalent of adding quite a few ply with most moves. Wetware can and does make howling blunders, and this is the main difference between engines and humans. Engines don't blunder, humans do. Anyway, I'd better read the article. |
|
| Mar-26-12 | | Pensive: <Engines don't blunder> I might argue that point.
Rybka vs Nakamura, 2008
Though, admittedly, the blunders were made under somewhat extreme conditions, Nakamura was under the same conditions and still managed to avoid blundering. |
|
Mar-26-12
| | alexmagnus: Engines occasionaly do blunder even without pressure, even one-move blunders happen, though extremely rare (most famous example of a one-move blunder by an engine is probably 19...Rfd8?? in P Lafuente vs Shredder, 2005, which was even reconstructed by <RandonVisitor>). But of course the probability of such a blunder popping up in an analysis is probably less than one in winning a lottery :) |
|
Mar-26-12
| | alexmagnus: And here another one-move blunder by an engine, Pandix' only loss in the last WCCC: The Baron vs Pandix, 2011 |
|
| Mar-26-12 | | Jim Bartle: That's probably mistake in the opening book, which I think is fed in by the programmers. |
|
Mar-26-12
| | alexmagnus: In Pandix' case - yes. But not in Shredder's case, where the comp - with given hash table settings - actually calculates the blunder to be the best move. |
|
| Mar-31-12 | | Dr. Yes: Met Karpov once. Nice fellow, don't you think? |
|
| Mar-31-12 | | ozmikey: <Dr. Yes> He certainly gave that impression during his visit to Australia in 1988. He behaved with great courtesy, generosity and good humour throughout (even when he was asked some barbed and even rude questions at the lectures he gave, or when Qantas managed to lose his luggage!). |
|
| Apr-09-12 | | swissfed: If only I had had my duel with Fischer, my fighting level would be of a higher order. Once I had attained and mastered such a level - a level which for Kasparov is completely unattainable - I would have recalled it whenever necessary.-by Anatoly Evgenyevich Karpov |
|
| Apr-09-12 | | JoergWalter: Is the quote below authentic?
<This match cannot end normally. Either I’ll be taken to hospital or else he’ll be taken to the insane asylum. – Anatoly Karpov (on the potential match against Fischer in ’75.)> http://www.chessville.com/Quotes/mo... |
|
| Apr-09-12 | | swissfed: <It is even more absurd to compare Fischer's chess strength with that of Kasparov, in whatever way, than it is to compare Fischer and me.> by Anatoly Evgenyevich Karpov |
|
| Apr-09-12 | | Everett: <my fighting level> this is one crucial point not easily transferable by ratings precisely. Karpov had tremendous fight, really a never say die attitude. This is precisely why Korchnoi stayed strong for so long; he never lost the passion to fight. |
|
| Apr-10-12 | | Everett: <swissfed: If only I had had my duel with Fischer, my fighting level would be of a higher order. Once I had attained and mastered such a level - a level which for Kasparov is completely unattainable - I would have recalled it whenever necessary.-by Anatoly Evgenyevich Karpov> Of course this is a bizarre quote, for if Kasparov had to fight a Karpov <who had the experience of fighting Fischer>, Kasparov would have every chance to develop amazing fighting strength (which he of course developed anyway against the "lesser" Karpov) I am not so enamored with ratings but with styles. In a way I feel bad for the new batch of stars, because it is very hard to create something new or to have a style that doesn't somehow take from the previous stars. Their individual stamp, outside of results, is hard to define. Perhaps there are players who play better than Karpov, but Karpov had a unique mix a patience, accuracy, tenacity and prophylactic play. Despite having no one to really push him in his youth (save Korchnoi, who was no slouch of course) he basically helped forge the greatest chess player of our time in Kasparov. Each player is of an era, and each "great" has given immensely to this game. |
|
Apr-10-12
 | | HeMateMe: sour grapes from Karpov. I think its ridiculous to think he would be significantly better, after a single 24 game match. Did Petrosian somehow become a much better player, after defeating Spassky in a match? He complained bitterly about the playing conditions at some of his 24 game matches with Kasparov. And--Kasparov claimed that there was a "secret grandmaster" hiding in another room evaluating candidate moves for Deep Blue, in the match which Kaspy lost. What was that GM quote? "I've never won a game against a healthy player". |
|
| Apr-10-12 | | Lambda: <I think its ridiculous to think he would be significantly better, after a single 24 game match.> Kasparov became significantly better in far fewer games against Karpov. Different sport, but I remember Ray Reardon saying he learnt everything he needed to be champion from a match with Fred Davis. |
|
| Apr-10-12 | | ughaibu: Ray Reardon clears the table: http://www.youtube.com/watch?v=lsw0... |
|
| Apr-10-12 | | Everett: <Lambda,< Kasparov became significantly better in far fewer games against Karpov.>> Well, this was a particular strength of Kasparov's. He really learned as much as he could from each tough opponent. Part of it had to do with how thoroughly his preparation was. |
|
Apr-10-12
| | HeMateMe: Considering how young Gary Kasparov was, he was gaining playing strength each year, just by being in GM events. I think he would have beat Karpov a year later even without the previous 63 game match, of which only the first 20 games or so are really relevant. The rest is a load of crap, where Kasparov played quiet draws with white every other game. Then, he
also played for draws when Karpov had white. Kasparov's stated goal was "to exhaust Karpov". This doesn't sound like anyone was getting much instructional chess. It was this poor performance, on both sides, that ended the no draw world championship matches, and brought back the 24 game matches. Karpov had probably been playing exclusively against grandmasters, the very best, for the past five years. Would 20-30 games against Fischer have changed Karpov's skill level? It may be true, but I am skeptical. Chess players tend to rewrite events a bit, in the way in which they would like to be perceived by the chess public and by history. I'm sure Anatoly Karpov would love to believe that a long match against Fischer would have turned Tolya into a juggernaut, capable of holding off Kasparov. The reality is somewhat different. |
|
| Apr-11-12 | | Everett: <HeMateMe> of course, but I imagine you agree that it's not just strength or skill level, per se. It's intensity. Matches with Kasparov, Karpov, Fischer, Alekhine, Lasker, Korchnoi, etc. are going to be at another intensity level... Not just ideas and accuracy at the board. Kasparov himself had a massive deficit in handling nervousness in '84, which is why he didnt deserve to be champion then. Otherwise, we do not know how Karpov would have responded to a match with Fischer, win or lose, in '75. I do know that he got his share of intensity in '78, which is likely a big reason why he was +4 =5 in '84. |
|
 |
|
|
< Earlier Kibitzing · PAGE 204 OF 254 ·
Later Kibitzing> |
Kibitzing
