06 Oct 09
1 edit
Originally posted by heinzkatsuppose you're playing a game and it's your turn to move against very strong opposition. that's it. if the site tried to explain what this game of "chess" is, you would try to refuse every single suggestion anyway.
[pgn][FEN "8/2q2k1p/1r2pppP/3p2P1/p1nP1P2/P2Q4/1PR1N3/1K6 w - -"]1. Ka2 Qb7 2. Qc3 Rb3 3. Nc1 Rxc3[/pgn]
This is an example of why I REALLY do not UNDERSTAND what CT wants from me. In any puzzle environment you can stop looking at the consequences of 1. ... Qb7/Qb8 quite quickly, since they are two different moves that reach the same thing. (I discarded ng on.
I really do not understand what kind of moves CT wants from me - how annoying 😕
I haven't seen a single time the owner's comments weren't accurate. if both Qb7 and Qb8 are winning, both are accepted. If Rb3 doesn't work, it doesn't work. you just got the problem wrong, it's not the site's fault, nor the owners'.
06 Oct 09
2 edits
Originally posted by heinzkatthere's no such thing in "puzzle environments." you were wrong to assume that in the first place. that, and you also failed to find the refutation to your move.
I did not try Qb7 nor Qb8 because they are the same - which to me means that none of them can be correct in a "puzzle" environment where only one correct move can be accepted.
06 Oct 09
Originally posted by philidor positionProper chess puzzles only have one winning continuation, come on now. Otherwise they are called "cooked". SwissGambit knows all about it.
there's no such thing in "puzzle environments." you were wrong to assume that in the first place. that, and you also failed to find the refutation to your move.
06 Oct 09
1 edit
Originally posted by heinzkatI don't agree. we aren't talking about compositions. (I don't like compositions by the way.) a puzzle already gives too much information. having only one move is too much to know. a puzzle would be cooked if the winning lines have a refutation.
Proper chess puzzles only have one winning continuation, come on now. Otherwise they are called "cooked". SwissGambit knows all about it.
and in the puzzle, Rb3 gives time to the queen to defend the pawn with Qd1&c1. if kramnik had made it, I'm sure it would get a ??.
06 Oct 09
1 edit
Originally posted by philidor positionHmm... that's one of the other factors why I do not understand CT - if I have found something that looks reasonable (+1 to +2), is that what I am looking for?
I don't agree. we aren't talking about compositions. (I don't like compositions by the way.) a puzzle already gives too much information. having only one move is too much to know. a puzzle would be cooked if the winning lines have a refutation.
and in the puzzle, Rb3 gives time to the queen to defend the pawn with Qd1&c1. if kramnik had made it, I'm sure it would get a ??.
06 Oct 09
2 edits
Originally posted by heinzkatRb3 definitely isn't the best move. Qb7 or b8 kills right on the spot. there are very rare cases in which that situation is open to discussion, but this is not one of them. you've just missed the defensive idea of white.
Hmm... that's one of the other factors why I do not understand CT - if I have found something that looks reasonable (+1 to +2), is that what I am looking for?
some tips for what CT wants from you: to be precise, the threshold is 1.75 pawns.
from a human point of view: having a piece for two pawns is almost always not the right answer, there's a better move. there are some cases where even having a piece for a pawn doesn't count, depending on the endgame. an exchange is sometimes the best answer, but in those cases you also have to evaluate the position strategically. sometimes it's not sufficient. you'll have to decide if you're a "healthy exchange" up, and find refutations for other alternatives. goes on like this.
it's all about being able to prove your move wins decisively, without question.
I'd also like to tell that I'm studying dvoretsky's tactical play at the moment, and I feel at home, seeing no difference from my training in CT.
06 Oct 09
If you claim Chess is Art, then you should above all appreciate Composed works of Art.
Or is it that there is no Rybka that composes 3000+ chess problems. Just kidding, there are very strong solving programs, for example Popeye & Euclide, so you will feel just at home. 😉
06 Oct 09
1 edit
Originally posted by philidor positionI was not claiming Rb3 is the "best" move at all, but a basic axiom when solving a chess puzzle is that there should not be an alternative that works just the same way
Rb3 definitely isn't the best move. Qb7 or b8 kills right on the spot. there are very rare cases in which that situation is open to discussion, but this is not one of them. you've just missed the defensive idea of white.
some tips for what CT wants from you: to be precise, the threshold is 1.75 pawns.
from a human point of view: having a pi lay at the moment, and I feel at home, seeing no difference from my training in CT.
06 Oct 09
1 edit
Originally posted by philidor positionThanks for that, so the difference between what it evaluates as "best" and "second best" is at least 175 centipawns? That makes sense. Should help a little. What is the minimum "advantage" after the first solution move then, +1.75 too, or might this be lower?
Some tips for what CT wants from you: to be precise, the threshold is 1.75 pawns.
06 Oct 09
Originally posted by heinzkatyou mean alternative? I don't understand. I also didn't get the joke about popeye and feeling at home.
What is the minimum "advantage" after the first solution move then, +1.75 too, or might this be lower?
all accepted moves have to be above 1.75. I think there are other details like if there is a mate, a non mating move isn't accepted in Standard even if it's greater than 1.75, but is accepted in blitz, or that if the difference between two moves that are above 1.75 is greater than x, the lower one isn't accepted, etc.