129 ( +1 | -1 ) The mystery goes onI have found a flaw in Kai's deepest game, 3...Nc3 is an illegal knight move, the quest will continue!
In all seriousness, I don't think there will be a definitive "deepest" game of chess, correspondence or otherwise, not until the game has been solved (wake me in 500 years when this is done), objectively speaking. A nice positional move may have consequences (beneficial or otherwise) on the position for many dozens of moves and a move in response to this type of move may be seen as being extremely "deep".
Still another example may be carefully manoeuvering your pieces and/or accepting a certain type of piece trade in anticipation of the coming endgame, this type of play could be seen also as extremely deep, particularly piece manoeuvers based on endgame theory (most of which has been proven remarkably correct with recent studies into endgame databases).
And so as to not leave out the opening, theoretical novelties designed to exploit psychological aspects of the players involved and to bring about a position, perhaps not proven universally best, but rather not to the liking of the abilities of your opponent may be considered deep.
It is safe to say any game played by a correspondence master may be considered sufficiently deep from which to glean instruction.
172 ( +1 | -1 ) One more point to noteI suppose maybe I did not fully answer the question, but my opinion is there may be a point in positional analysis where it is not practical to continue with analysis. For example, it is practical to have analysis on every key move, but to analyse every move to checkmate is not practical as it is unlikely the game will actually continue as expected for very long (save long tactical sequences). People make mistakes and overlook things, especially in deep analysis on a complex position.
I am forced to imagine olympio coming to a web page advertising the world's deepest game, finding thousands of lines of analysis, presuming by quantity this must be the deepest game every created. This game would likely be played by computers as they are not hindered by fatigue or disinterest. Their analysis may well be valid for the first couple of moves in the analysis but then stray away with an oversight, leaving 90% of the analysis pointless. They were not told to stop and so continue on without a clue what happened. This analysis would not be nearly as englightening or useful as that given by a real player, not analysing as "deeply".
In other words, at some point, you have to say "that looks good," write down your move and put it in the mail drop, you can analyse deeper later when your opponent's response arrives in the mail.
I've found myself getting carried away with analysis and end up playing games against myself from a position, at some point you need to "snap out of it" and go back to the position you must play from. These tangents players run off, I don't believe should be considered "deep analysis", it is not sufficiently objective in my opinion.
62 ( +1 | -1 ) i agreewith everything you said except for the computer point..
because of the way game tree generation works.. a computers oversights basically only occur at the "leaves" of the tree. it wouldn't happen at the top leaving the 90% below useless.. though the incorrect scoring of the leaves due to horizon-effect may trail back up and have a profound impact on the quality of the top nodes.. but due to singular extensions and the like.. the best programs don't have that problem
and basically if the incorrect scoring of the leaves did screw up the analysis at the top lvl.. allowing the computers to continue on would only fix the problem cause then a ply or two later the problem would be fixed
33 ( +1 | -1 ) and i thinki should rephrase my question.. what was the deepest tactical combination every calculated in a correspondence game..
i'm pretty sure kasparov v. topalov featured the deepest tactical combination in an OTB game.. but i'd like to know for correspondence games.. since i'm sure the best in that would dwarf the depth of his Rxd4 move
4 ( +1 | -1 ) this is no jokeyou've put an end to chess...
that's no laughing matter sir kai
204 ( +1 | -1 ) hmm, deepest combinations..First, as for the computer move selection routines, I don't know how they work today but even some of the latest Fritz analysis I've seen on chessbase, there are some simply bad moves in its analysis. I would trust them to work out the tactics involved in a sound sacrifice but I would not expect them to understand a piece sacrifice for initiative, or in other words, a positional sacrifice.
I am not familiar too much with modern correspondence championship matches but from what I have seen, they are incredibly deep, in all aspects. As for the change in style of OTB play, the days of Fischer have long since been over. In his day, they were more oblivious as to what could happen and would accept the type of complications modern players love, when in fact they simply could not keep up. This allowed players like Fischer and Karpov to build up a position (in Fischer's case, in an aggressive style; in Karpov's, a seemingly defensive style, inviting attack) that would reach a sort of critical mass, allowing suddenly and from nowhere a constant supply of combinations and tactics.
By the time of the 1980's, most players (GMs) discovered how this process worked and played more to defuse these positions long before they could reach any critical mass. To me, these games played between modern GMs show a deep understanding far surpassing even Fischer in his prime, and to a greater degree, those by correspondence GMs.
As for a single combination featuring a steady stream of tactics, it may be a matter of opinion. A 10 move combination featuring a windmill motif is certainly deep from a quantitive standpoint, while a 5 move Queen and Rook sacrificial mating attack out of nowhere may be qualitatively deep (especially if you've never been exposed to the theme).
If I were to submit a deepest game, I don't think I could stop with just one.
Everyone wants the perfect game to feature an opening from their repertoire.
55 ( +1 | -1 ) ...It would be easier if it was to find the longest combination. Even there it would be difficult. I have beaten computers and humans in endgames after what at first seemed to me to be simple blunders but because of abbreviated analysis enabled by knowing specific K+P formations actually were 40+ moves deep.
PS This is one problem that computers will still have for quite some time. There are many types of positions in the endgame where a human's knowledge enables them to calculate 40+ moves deep in seconds.
for every 3 fold increase in processing power.. you get about a 1-ply deeper search and endgame databases are out there.. a program with access to ALL endgame databases.. could for the most part handle itself in most practical endgames..
also there is better programming.. that is enabling softwares to determine good endgame moves.. but this is the slowest line of developement..
all in all.. programs will tend to suck in the endgame like bogg said
i have literally seen positions where players who have been playing chess for a week could find the mate and a program couldn't.
namely: rook+king vs. king
55 ( +1 | -1 ) ComputersIf computers keep getting faster at the current rate, and they have been getting about 50% faster each year for the last 25 years so this seems likely, in about 8 years a cheap desktop will play open middlegames as well as the world champion! In an additional 5 years they will be able to play 40/2 better than the world champion's home analysis. It may be centuries before they can play either side of the French defence properly. Long live closed positions!
in 8 years.. Go programs may be able to play on the level of the average club player..
7 ( +1 | -1 ) But that is onlyBecause there are more pieces (= more legal moves).
38 ( +1 | -1 ) Kasparov-TopalovEveryone who want to take a look at this great game, please take a look at my website www.schaakpartijen.nl. It's a new website and at the moment only in Dutch, but the game annotations are in English and from the master himself. To watch the game go to "Partijen" (games) in the menu at the left, select the game and enjoy!
9 ( +1 | -1 ) hoi optimistleuke site, maar bij mij bewegen de stukken niet. wel gaaf, dat commentaar:) groetjes.
11 ( +1 | -1 ) kasparov-topalovis defienitely my favourite game ever..
but tell me..
if topalov had played 24. Kb6.... would this game be famous?
20 ( +1 | -1 ) Now that I understand your question, ...... Bogg is right, it is easier to analyze simple endgames. The record for an announced mate is 45 moves by the Liverpool Chess Club in a corrsepondence game against the Edinburgh Chess club in 1901!
19 ( +1 | -1 ) Problem solvedYabbadabadoo! It works now! =)) Please come by my website www.schaakpartijen.nl to watch and play through the Kasparov-Topalov game.
18 ( +1 | -1 ) Direct linkPlease, be aware the direct link of loreta doesn't work anymore. It's now www.schaakpartijen.nl/pgn/kastop.htm