|Aug-29-03|| ||Shadout Mapes: 27.Rd4!! It amazes me how Tal could get two seemingly unrelated weaknesses (scewering the queen and king weaknesses) and exploit them like this. If black tries to save the knight with 27...Qc6 the obvious 28.Qxh7+! mates. |
|Jun-01-04|| ||Rowson: What a game! my two favourite players clash, Tal coming out on top. Shadout Mapes Rightly giving 27.Rd4!! as the the winning blow. When will we ever find another Tal? |
|Dec-16-04|| ||aw1988: What is the big deal about 27. Rd4 that it merits two exclamation points? |
|Dec-16-04|| ||DanielBryant: Yeah. It wins the queen, but it's a fairly obvious move. |
|Dec-16-04|| ||solstys: I didn't think 27. Rd4 was an obvious move and apparently neither did Andersson. I think it was a great move. |
|Jan-17-05|| ||Hinchliffe: <SadeoutMapes> Totally agree with your comments.Also like your pick of musicians. But if I may suggest another guitar player (of course there are 1000's) worth wrapping an ear around - check out Mister Jeff Beck. From his early stuff through to his most recent. I think you'll find something rewarding just like playing through a Tal game. |
|Feb-21-05|| ||offramp: The most obvious move - well, the one I would've played - is 27.Rh4, hoping for some kind of mate along the h-file. But that is a blind alley and after 27...Nf8 the attack is over and white is a pawn down.|
So I like Rd4.
|Feb-21-05|| ||aw1988: Rd4 is a standard move. |
|Sep-23-05|| ||ConfusedPatzer: The star move is 20. Qe3 after which every move is forced, not 27. Rd4.|
|Apr-21-07|| ||Tomlinsky: Unless I'm mistaken Andersson actually had a very simple defence here. Instead of 24...Kh8 the saving move is 24...Qh5! I believe with threats such as Bd6 and Rd4 now firmly off the menu at the very least.|
If white plays 25.Rg5 then 25...Qh6 and if the rook lifts to f3 intending h3 then black plays Rfe8 and the queen can retreat to f8 replacing it. White's attack is now an illusion and while Black looks on the backfoot the position is pretty darn solid and good for at least draw to me.
|Apr-21-07|| ||Gilmoy: <Tomlinsky>: Not enough -- White's control of g7 is still killing. <24 .. Qh5 25 Rg5 Qh6 26 Rf3 Rfe8 27 Rh3 Qf8> 28 R5h5 and Black can't protect h7. 28 .. gxh5 29 Qg5+ Kh8 30 Qxh5 Qg8 31 Rg3 wins Black's Queen: 31 .. Qf8 32 Rg7.|
That leaves material technically even, but White eventually has Qxf7, plus the h5 pawn whenever. In the 31 .. Qxg3 line, a later Qxf7 is with tempo (Rg8 Qxe6 etc.), so White can ignore the first threat on his Bishop, plus he has ridiculous f7+ threats -- all of which might force Black's Rf8, which leaves both Rooks passive. Black probably must lose a piece somewhere to stop the f-pawn.
|Apr-21-07|| ||Tomlinsky: <Not enough -- White's control of g7 is still killing. <24 .. Qh5 25 Rg5 Qh6 26 Rf3 Rfe8 27 Rh3 Qf8> 28 R5h5 and Black can't protect h7. 28 .. gxh5 29 Qg5+ Kh8 30 Qxh5 Qg8 31 Rg3 wins Black's Queen: 31 .. Qf8 32 Rg7.>|
Let's see... I play 32...Qxg7 33.fxg7+ Kg8 is forced and I have two rooks for the queen with White's attack now paralysed...
click for larger view
I'd rather be Black here any day of the week. There's at least a draw in hand in this position. Black can become far more active than White very quickly.
|Apr-21-07|| ||keypusher: <Tomlinsky> Interesting idea. So maybe 28. Qg3 threatening R5h5? Then if 28....Kh8 29. Qh4 h6 30. Qxh6 Qxh6 31. Rxh6+ Kg8. Now if 32. Rg3 e5 33. Rgh3 Nxf6!, so I don't see an immediate mate, but White must be winning. But does Black have better on move 28?|
|Apr-21-07|| ||Tomlinsky: <So maybe 28. Qg3 threatening R5h5? Then if 28....Kh8 29. Qh4 h6 30. Qxh6 Qxh6 31. Rxh6+ Kg8. Now if 32. Rg3 e5 33. Rgh3 Nxf6!, so I don't see an immediate mate, but White must be winning. But does Black have better on move 28?
After 28.Qg3 Black has 28...e5! Now the f-pawn drops and Black is winning.
|Apr-21-07|| ||keypusher: yep, that's a better 28th move, alright...|
|Apr-23-07|| ||Gilmoy: <Tomlinsky>: Indeed, your position looks like an oasis for Black ... but I recall Tal's won with similar pawns. He could steer for a set-up of h4, Qg5, Be3, and h5-h6. So Black probably can't ever get both Rooks forward.|
Another tactical shot is an immediate Be3, when Black can't take the g-pawn yet, because of Bh6+, Qxf7 Rg8 etc., with pressure on the (a,b)-pawns.
Another idea is (instead of Rg7) Rf3, then Be3-h6 to boot the Queen out. The B is getting kicked anyways -- so it has two good reasons to move, and on the h6-f8 diagonal it becomes immense.
|Apr-23-07|| ||Gilmoy: Bah -- this is why he's Tal, and I'm not. <24 .. Qh5 25 Rg5 Qh6 26 Rf3 Rfe8 27 Rh3 Qf8> 28 Rg4! (to reach h safely). Also, I find it helps when I set up the board properly :) <Tomlinsky: Instead of 24...Kh8 the saving move is 24...Qh5!> i.e. the King is still on g8, not h8. This matters; Qg8 is impossible, hence h7 is Tal-chow:|
28 .. e5 29 Rxh7 (not check, per above) Kxh7 (else Rgh4) 30 Rh4+ Kg8 31 Qh3 and Black is 1 move too late to remove the f6-pawn *and* vacate f6: 31 .. Nxf6 32 Rh8+ Kg7 33 Qh6# and Black's Knight and Queen are helpfully part of the mating net. Note how this line properly addresses the danger to the Bishop by simply ignoring it -- as befits a Tal.
28 .. h5 29 Rgh4 e5 30 Rxh5 and Black either opens g and dies to Qg5+, or Nxf6 and the same mate as above.
28 .. Qd6 (losing about 5 tempi) and I think 29 Rxh7 still works. 29 .. Kxh7 is beautiful: 30 Rh4+ Kg8 31 Rh8+ Kxh8 32 Qh6+ and that pawn is killing, a lovely variant of Damiano. 29 .. Nxf6 fantastic 30 Rh8+ Kxh8 31 Qh6+ Kg8 32 Rxg6+ Kf8 (Kh8 34 Bxf6#) 33 Qxf6+ and 34 Qg7#.
Man, that Tal is a genius or something :)
|Apr-23-07|| ||Tomlinsky: <Gilmoy>: The key to what Tal may have had in mind as a response could lie within a combination of what you and <keypusher> have posted. After 24...Qh5 25.Rg5 Qh6 an immediate 26.Qg3 with the intetion of following it up with h4/Be3/h5 looks a very strong and Talesque(?) continuation. |
I had mostly concentrated on basing the Qh5/Qh6/Rf8/Qf8 manouvre/defence around white playing the rook-lift in the sequence but I think 25.Qg3 is the correct response to 24...Qh5 here. If only that chain-smoking genius were around today to see if what he had in mind as a response OTB corresponds with what it took me 30 years to come up with!
|Apr-23-07|| ||Gilmoy: Oops, left out a move-pair: 32 Rxg6+ fxg6 33 Qxg6+ etc.|
|Jun-04-10|| ||Everett: Imagination in Chess, puzzle 587 White to move 24..Kh8|
I kept on trying to sac the rook on g6, but this is easily refuted by a timely ..e5, blocking the bishop diagonal.
The h-file kill doesn't work because of 25.Qh6 Rg8 26.Rh4 Nf8, thus Tal played to deflect the knight with 25.Bb6 and followed with the now available square 27.Rd4. The queen can't move off the 5th rank 28.Qxh7+ and 29.Rh4#
|Jun-04-10|| ||whiteshark: After <20...Kh8> the technical department is rather cluecless how to proceed and break the 0.00 ewalluations. |
click for larger view
|Jun-07-10|| ||Everett: <whiteshark>
Black is up a pawn, but the evaluation is 0.00. If there is no forcing line that balances material, then the computer is weighing white's "compensation" as 1 pawn. How accurate is the computer's understanding of compensation?
|Jun-07-10|| ||whiteshark: <Everett> I don't know how they measure "compensation", not to mention their accuracy of understanding it.|
For the <0.00> evaluation after <20...Kh8>, it goes <21.Nd6 f5 22.exf6 Bxf6 23.Nxf6 Nxf6 24.Qa7!>
click for larger view
when playing against points f7/g7 leads to perpetual/repetition, e.g.
<24...Rd7 25.Qxd7 Nxd7 26.Rxf8+ Nxf8 27.Nf7+ Kg8 28.Nh6+>
click for larger view
|Jun-07-10|| ||wierba: why not 26... Nxf6? 27 Rxf6 Qd1+|
|Jun-07-10|| ||Sastre: If 26...Nxf6, then 27.Bd4 e5 28.Rxf6 exd4 29.Rh4.|