Am i missing something??>

Yes - as was already pointed out in previous kibitzing, 33.Qg7! to prepare Rh5+. 33.f4+ exf4 34.e5 is also good enough to win; Black has an ingenious way to keep the game going for a bit longer by 34...f3+ 35.Kf2 Nd3+ 36.Bxd3 Qb6+ 37.Kxf3 Qf2+ 38.Kxf2 Rf8, but after 39.Bxg6 Rxf7+ 40.Bxf7 it's obviously hopeless for him.

Here's my computer (Fritz) checked breakout:

<29. Nxd6 !! Qxd6>

Relatively better for Black may have been 29... Reb8, when 30. Nxb5 Rxb5 31. Qxc3 leaves White with a solid extra pawn but no clear forced win.

<30. Rxh7! f6>

If 30... Kxh7, then White forces Black into a mating web after 31. Qxf7+ Kh6 32. Rh1+ Kg5 33. Qg7! when play might continue 33...Rh8 34. Rxh8 Rxh8 35. Qxh8 Bf1+ 36. Kg1 Kxg4 37. Bd1+ Be2 38. Bxe2+ Kg5 39. Qh4#.

If 30... Rf8, White forces a transposition back to the game continuation with 31. Rch1 f6 32. R1h6 .

If 30... f5, then White wins with 31. Rch1!, when play might continue 31...Qf6

(31... Nxe4 32. Bxe4 c2 33. Rh8+ Kg7 34. R1h7+ Kf6 35. gxf5 c1=Q 36. fxg6+ Qf4 37. Rf7+ Kg5 38. Qh5#;

31... Nd3 32. Bxd3 Bxd3 33. Qxd3 fxg4 34. f4 exf4 35. gxf4 Qxf4 36. e5 Qf3+ 37. Kg1 Kf8 38. Rh8+ Ke7 39. R1h7+ Kd8 40. Rxe8+ Kxe8 41. Qxg6+ Kf8 42. Rh8+ Ke7 43. Qe6#)

32. exf5 gxf5 33. R1h6 e4 34. Qf4 .

<31. Rch1 Rf8 32. R1h6! Rad8 33. Rxg6+! Kxh7 34. Qf5> 1-0

Black resigns as he cannot prevent a quick Qh5# (e.g. 34...Bf1+ 35. Kg1 any and 36. Qh5#).

< Ladolcevita: I just look back this puzzle,and i think white can survive by taking the rook,this is my line 31.Qxf7+ Kh8 32.Rh1+ Kg5
33.Rh5+ gxh5

Am i missing something??>

In many sacrifices, the victim is best to refuse, yielding the perpetrator a "mere" P with improved positional considerations. For some strange reason, upon seeing such consequences, I do not experience your apparent disappointment, <al>. Kotov's rule (as paraphrased by <UdayanOwen>): when you are a P up, with all else equal, you can cease calculation.

B Mathews vs J Subervi, 2008 (29.?)

White to play and win.

Material: Down a P. The Black Kg8 has 3 legal moves, all dark squares. White has Rh1 on an open file, with Rc1 ready to reload. The White Qf3 is on a semi-open file, but in <Once>’s happy phrase, Qf3 is a “tall R” hemmed in by the White K-side Ps. The White Nc4 attacks Pd6, tying down Qe7. The Black Bb5 is loose and quite vulnerable after Nxd6. The White Bc2 holds the Q-side, but requires activation. The White Kg2 is secure. To summarize, White has the h-file, which gives him an excellent base for operations, but Black has Q-side counterplay. The “normal course” for the game yields 29.Qxc3 [else, Bxc4 30.bxc4 with a strong Q-side initiative from the passed Pa5]. The h-file indicates the possibility of one of <Gilmoy>’s smiting moves, i.e., checks, captures, and threats.

Candidates (29.): Qxf7+, Rxh7, Nxd6

29.Nxd6 (threatening 30.Nxb5 or 30.Nxe8)

The Qe7 is overburdened and must prevent both Nxd6 and Rxh7, but cannot.

29…Qxd6 [else, face a passed Pd5 and then drop a P without compensation to Qxc3]

30.Rxh7 (threatening 31.Qxf7# and 31.Rch1 32.Rh8+ 33.R1h7#)

The Rh7 is untouchable:

(1) 30…Kxh7 31.Qxf7+ Kh3 [Kh1 32.Rh1#]

32.Rh1+ Kg5

<[Here I experienced chess blindness with 33.Qg7+ forgetting Pg6. Toga gives

33.Qg7 (threatening 34.Rh5+ Kxg4 35.Bd1, etc. winning).]>

(2) 30…Qd7 [or Qc7] 31.Qf6 then 32.Rh8# after reloading with Rc1 if necessary

(3) 30…Qe7 31.Rch1

Black must advance Pf7, so White wins at least Q+2P for R+N, with a huge attack to come.

(4) 30…Rf8 31.Rch1

(4.1) 31…f5 32.gxf5 gxf5

[else, White forces …Kf7 and plays fxg6+, effectively finishing the game]

33.Qh5 and the threat Rh8+ with control of f7 finishes Kg8.

(4.2) 31…f6 32.g5 (threatening 33.gxf6 Qxf6 [if Rxf6, Black is mated] 34.Rh8 35.Rh7+ 36.Qxf6)

The play in this variation centers on control of f7, a critical flight square when White has an h-file battery. If White controls f7, he can win with Rh8+ then Rh7#. Because the Black Rf8 must shelter Kg8 with the escape Kg8-f7-e8, it cannot defend f6, because Rxf6 permits Rh8+ then Rh7#.

32…Nd7 33.R1h6 (threatening 34.Rxg6+ 35.Kxh7 36.Qh5#)

33…f5 34.exf5 (threatening 35.fxg6 36.Rh8+ 37.Rh7#)

34…gxf5 35.Qh5

1) Al-ECK-en (youtube interview)
2) Al-uh-KEEN (I grew up hearing this one)
3) Ai-YECK-en (Larry Christiansen)

Vowel sounds in Russian change, depending on whether they occur in a stressed or unstressed syllable. For historical reasons, 'o' in a stressed syllable following a palatalized consonant (one with a little trailing 'y' sound, like the sound following the 't' in the British pronunciation of "tune") is written as 'e'. When Russians want to distinguish this from an ordinary 'e' (which they almost never do, since they already know how it should be pronounced), they put a diaresis -- identical with a German umlaut -- over the one that sounds like 'o'.