|
< Earlier Kibitzing · PAGE 3 OF 3 ·
Later Kibitzing> |
| Oct-28-10 | | CHESSTTCAMPS: <Once> Genius! |
|
Oct-28-10
 | | patzer2: Black's 19...Kf8 is an awkward looking move, allowing the clear winning shot 20. Rxg7! However, it would appear Black is lost at this point anyway. Black's best chance to put up a defense appears to be 19...h5!? when 20. Qh6! Ng4 21. Qxh5! (not 21. Qd2? f6!  ) 21...Nxf2+ (21...Bxh4 22. f6!  ) 22. Kg1 Nxh3+ 23. Rxh3 f6 (23...Bxh4 24. f6! ) 24. Bc4! gives White a winning attack. Other 19th move possibilities for Black, give White an even easier win: If 19... Kh7, then it's mate-in-five after 20. Rxg7+ Kxg7 21. Qxh6+ Kg8 22. Rg3+ Ng4 23. Rxg4+ Bg5 24. Rxg5#. If 19... Nh5, then 20. Qxh6! wins when play might continue 19...Bxh4 21. Qxh5 Kf8 22. Rxg7 Kxg7 23. f6+ Kxf6 24. Rf3+ Kg7 25. Rxf7+ Kg8 26.
Qh7#
If 9... Ne8, then 20. f6! wins when one possibility continuation is 20...Bxf6 21. Bxf6 Nxf6 22. Qxh6 Kf8 23. Rxg7 Ke7 24. Rf3 . |
|
| Oct-28-10 | | bengalcat47: And the poisoned pawn variation has claimed yet another victim! |
|
Oct-28-10
 | | Once: <AylerKupp> I generally give it somewhere between 30 minutes and an hour, including trying to solve the puzzle, reading the previous kibitzing and sometimes chucking the game into Fritz to see if he comes up with anything interesting. Glad you enjoy them. |
|
| Oct-28-10 | | riverunner: well yeah, if he takes it. |
|
| Oct-28-10 | | Patriot: <CHESSTTCAMPS> I have to agree with <LIFE Master AJ>. "Resigns" is the best move for black rather than trying to minimize the time it gets mated or playing giveaway chess. Among human players it's considered rude to play on in positions that are obviously lost so computers must sometimes be forced to resign during matches against the top world players ("IF" they are losing!). I've played against people that do that and find it annoying because it wastes my time and theirs when I could be taking a break and waiting for the next round. It's also true that sometimes a computer's #1 numerical choice is not always the best human choice. In a losing position, computers try to minimize damage by playing moves that are obviously lost but may give no counterplay. Against a human player, it's much better to play a move that is dead lost that offers counterplay over a move that is just lost without any remote chance in hades. It's like a bluff. If it doesn't work then you were lost anyway and had nothing to lose; otherwise you may be able to turn the game around. |
|
| Oct-28-10 | | Naugh: Actually got the 20. ♖xg7! move with some sample contiuations, but my guess is that it wasn't hard to see. |
|
| Oct-28-10 | | Patriot: <CHESSTTCAMPS> I should clarify that the first "silly" move is after 20.Rxg7 Ke8 21.Qxh6 Ng4?. Umm... 22.Rxg4 is "good enough". |
|
| Oct-28-10 | | rapidcitychess: Good Evening: Easy and awesome. |
|
| Oct-28-10 | | CHESSTTCAMPS: <Patriot> <<CHESSTTCAMPS> I should clarify that the first "silly" move is after 20.Rxg7 Ke8 21.Qxh6 Ng4?. Umm... 22.Rxg4 is "good enough".> I understand that 22.Rg4 is good enough - that's why I put '??' on 22.Qh5. My primary point was that the computer is playing by its algorithm and it has concluded that the knight is lost anyway and/or the other alternatives are otherwise worse. Because black's position is lost anyway, do you have an objective basis for giving Ng4 a '?', i.e. can you find an alternative that doesn't lose even faster against best play from white? As for the argument about what's rude or not, it's rather irrelevant for us, because we're voluntarily playing the engine for fun or to test our understanding of the position. I do agree with <LIFE Master AJ> that it would be nice to have Crafty vary its response. However, you can deal with this by setting up Crafty End Game Trainer with an alternative position that you want to study. |
|
| Oct-28-10 | | Patriot: <CHESSTTCAMPS> <Because black's position is lost anyway, do you have an objective basis for giving Ng4 a '?', i.e. can you find an alternative that doesn't lose even faster against best play from white?> My main point is that Ng4 is just a silly move that gives a piece away. This makes sense if black has counterplay after 22.Rxg4 but here, what is there? The move deserves a ? because the best move is "resigns". But I do think you have a point. What you're talking about is practice in "winning a won" game (i.e. technique of winning). In that sense I completely agree with you. It is really great practice playing a winning position against a computer. Maybe 21...Ng4 is the most stubborn move that Crafty could find. |
|
| Oct-28-10 | | sfm: When clicking on Black's move 14 and then move 18, Black's position didn't change at all, while White has moved two rooks into attacking position and opened the c1-h6 diagonal by playing f4-f5. Can that have something to do with the speedy end? ;-) |
|
| Oct-28-10 | | CHESSTTCAMPS: <It is really great practice playing a winning position against a computer.> It is indeed. If you haven't already done so, I recommend that you try playing out the "won game" that resulted from the solution of Monday's puzzle, using the link set up by <David2009>. |
|
| Oct-28-10 | | TheBish: S Matthews vs P Seegolam, 2010 White to play (20.?) "Medium"
I got this last night fairly quickly, which must mean I'm improving, since I was tired after coaching chess and being stuck in a traffic jam, and I was looking at it while watching the World Series -- Go Giants! Fear the Beard! (Giants won game one 11-7 in a wild one, which was supposed to be a pitcher's duel!) 20. Rxg7!!
When I saw the white rooks lined up in an unusual formation (g3 and h3, pretty good if you can get it!), I figured that the h6 square would be a key attack point -- but h6 is defended by g7, so maybe we can lure the black king out? Now: A) 20...Kxg7? 21. Qxh6+!
(Only one exclam for this, as you had to see it before playing 20. Rxg7!!) 21...Kxh6 (or 21...Kg8 22. Rg3+ Ng4 23. Rxg4+ Bg5 24. Rxg5#) 22. Bxf6 mate, quite a combo! B) 20...Ng8 (defending h6) 21. Rxg8+! Kxg8 22. Qxh6 Bxh4 23. Rxh4 exf5 24. exf5 and Black is hard-pressed to stop mate. There are probably a couple of better defenses, but not using an engine (or moving pieces) and that's all I have time for now - checking it out before I have to go coach a class. |
|
| Oct-28-10 | | BOSTER: During three days in row <CG> represented the puzzles,where the solution was to find only one (but "killing") move.
Monday-f4
Tuesday-Qc2
Wednesday-Nh6.
Nevertheless, some players including <dzechiel> still count <Material even> or not, missing one nuance-that "The worth of material at all times is related to the position" -Horowitz.
Being able to look one move ahead I'd like to see the <CG> puzzle with zero moves for solution, when White have checkmated his opponent's King,and diagram has White to play.
The question is . Does anybody count <Material even> in such position? |
|
| Oct-28-10 | | wals: Analysis Rybka 4 x 64
depth 13 : 4 min :
White blunder
(-1.15):12.Kh1. Best, Qe1, -0.68.
depth 18 : 60 min :
Black blunder
(+10.37):19...Kf8. Best, Ng4, +6.25.
and the end of the line for Black. |
|
| Oct-28-10 | | Patriot: <CHESSTTCAMPS> I tried today's problem against Crafty and it took somewhere around 50-60 moves to checkmate it. Basically I played "simple chess", trying to make sure Crafty had little to no counterplay. I made several mistakes along the way and dropped a few pawns but in the end I traded my knight for his last pawn and went into a simple king and pawn winning endgame. Crafty allowed a stalemate trick (my king was on f7, my pawn on g5, black king on h8-- 1.g6?? stalemate) but I played it correctly and Crafty still played on against king and queen until it was checkmated. ---
Here's some advice for kibitzer's who don't know how to win in winning positions: It's good to know the technique of "simple chess" (not playing into complicated positions whenever winning and trading down pieces) and another of Dan Heisman's recommendations, "Think defense first", which basically means your thought process changes when playing a winning game such that you look for your opponent's threats before your own. If you can take away your opponent's counterplay you should be able to win by attrition. |
|
| Oct-28-10 | | Patriot: <BOSTER> When dealing with puzzles, it's a good habit to consider material imbalances before delving into analysis. The reason is that, for example, white is ahead a piece and black has a strong attack it's not unreasonable to give a piece back with no further compensation. However if your point is "Why look at material imbalances or the other player's threats before looking for a forced mate?", then you may be right. If white has a mate in three and it is white's move then who cares what the material is or if black is threatening mate in one. Just force the mate! |
|
| Oct-28-10 | | Bignevermo: HEY ONCE!!!! i second(third actually) AylerK.....i ALWAYS look to see what you have to say!! you explain things so succintly!! :)which REALLY helps!! cause after being here a while now and trying to play this game....i found i aint as smart as i once thought i was!!! :(... :) |
|
| Oct-28-10 | | CHESSTTCAMPS: <Patriot: <CHESSTTCAMPS> I tried today's problem against Crafty and ...> Did you mean today's puzzle or Monday's puzzle (much tougher)? The Crafty link for the Monday puzzle is:
http://www.chessvideos.tv/endgame-t... You can win a piece, but you must play accurately to win. |
|
| Oct-28-10 | | Eduardo Leon: After some time away, I am back.
<20.♖xg7!>
Now 20...♔xg7? loses to 21.♕xh6+ ♔xh6 22.♗xf6#, or 21...♔g8 22.♖g3+ ♘g4 23.♖xg4+ ♗g5 24.♖xg5#, so the rook is invulnerable. <20...♘g8 21.♖xg8+ ♔xg8 22.♗xe7!> Exploiting the fact it is impossible for black to defend h6. <22...♘xe7 23.♕xh6> Threatening 24.♖h8#, 24.♖g3+ and 24.f6.
<23...f6 24.♖g3+ ♔f7 25.♕h5+!> But not 25.♖g7+? ♔e8!. Now 25...♔f8 allows 26.♕h8+ ♘g8 27.♕xg8+ ♔e7 28.♖g7#, but the "clever" alternative does not save black from mate. <25...♘g6 26.♕xg6+ ♔e7> Still trying to be "clever". Of course, 26...♔f8 allows 27.♕g8+ ♔e7 28.♖g7#. <27.♕g7+ ♔e8 28.♖h3!> And 29.♖h8# is unavoidable. |
|
| Oct-28-10 | | njchess: Fun puzzle. Black must have died when the Queen took the pawn! |
|
| Oct-28-10 | | Patriot: <CHESSTTCAMPS> Actually I was referring to today's game. |
|
| Oct-28-10 | | stacase: A poisoned Rook if ever there was one. |
|
| Oct-29-10 | | LIFE Master AJ: The links are always nice, my point (the human perspective) is that engine chess sometimes has no relevance to what us CBL's do. |
|
 |
|
|
< Earlier Kibitzing · PAGE 3 OF 3 ·
Later Kibitzing> |
Kibitzing
sac: 20.Rxg7