Rublevsky

Originally posted by Sicilian Smaug
Bxf6+

Wouldn't it be Rxb6? Because it's the same game as the original position, and white still has his bishop. He needs it to take the pawn if he black takes the rook.

Originally posted by Sicilian Smaug
Oh well, my move still wins but in more moves.

I'd say white should win from:



(white to move)

But, white still has some opportunity to go wrong, while the mate in three shown by z00t is clear. If black plays Kxf6 after the initial Bxf6, mate follows in four, but again not by grabbing the rooks.

Perhaps the whole game should be viewed. Note how black gulped the pawns while his king was still in the centre

[Event "4th Karpov It"]
[Site "Poikovsky RUS"]
[Date "2003.04.26"]
[Round "6"]
[White "Rublevsky"]
[Black "S Lputian"]
[Result "1-0"]
[WhiteElo "2670"]
[BlackElo "2638"]
[EventDate "2003.04.20"]
[ECO "C03"]
[PlyCount "52"]

1. e4 e6 2. d4 d5 3. Nd2 Be7 4. Ngf3 Nf6 5. e5 Nfd7 6. Bd3 c5 7. c3 Nc6 8. O-O a5 9. Re1 cxd4 10. cxd4 Qb6 11. Nb1 Nxd4 12. Nxd4 Qxd4 13. Nc3 Qb6 14. Qg4 Kf8 15. Nb5 Nc5 16. Be3 Bd7 17. Nd6 Bxd6 18. exd6 Qxd6 19. Qg3 Qc6 20. Rac1 b6 21. Rc3 h5 22. h4 Rc8 23. Bd4 f6 24. Bg6 Re8 25. Rf3 Ke7 26. Rxf6
1-0

Another theoretical position. This one from Rublevsky's annotations to Rublevsky-Asrian, Moscow 2004, published in Informant 90.

white to move

This position is much more interesting than the last. The first thing I looked at here was the crazy Qxe5??, looking to follow ..Qxe5 with Bf4, Nd6+ and Nxf7, but that doesn't work due to ..Qh5!, when f7 is protected. So then I moved on to Rxf7, which looks absolutely devastating here. A line I looked at was Rxf7!! Qxf7 Qxe5 (Threatening Nd6+) Qf8 Bg5! (Threatening all kinds of stuff: Rf1, Nd6+ (forcing the queen to give herself up)) and Black can resign here. I might have missed something dumb here. Maybe there is something better than Qf8, though I am pretty sure Rxf7 is the move.

Originally posted by !~TONY~!
This position is much more interesting than the last. The first thing I looked at here was the crazy Qxe5??, looking to follow ..Qxe5 with Bf4, Nd6+ and Nxf7, but that doesn't work due to ..Qh5!, when f7 is protected. So then I moved on to Rxf7, which looks absolutely devastating here. A line I looked at was Rxf7!! Qxf7 Qxe5 (Threatening Nd6+) Qf8 Bg5! (Thr ...[text shortened]... dumb here. Maybe there is something better than Qf8, though I am pretty sure Rxf7 is the move.

Rublevsky gives Rxf7, but then Bxh2+ before Qxf7