White's patience makes a lasting impression. After <9..e6> "returns" a pawn for open lines, White ignores it, resumes development, and creeps inexorably north, 1 or 2 squares at a time. In fact, after <7.Bxf7+>, White doesn't make another (3+)-square move until <31.Rxa7> he feels ready to attack :) It shows the depth of Mackenzie's prep: clearly he knew what the position had, and trusted its positional advantages, regardless of any material deficit.

(1) I played this line a couple of times, with great success: 1.e4 d6 2.d4 Nf6 3.Nc3 g6 4.Bc4 Bg7 5.Qe2 Nc6 6.e5 Nxd4 (I think 6...Nd7! is now considered good for Black) 7.exf6 Nxe2 8.fxg7 Rg8 9.Ngxe2.

(2) In the Saemisch King's Indian, there is a famous queen sacrifice, one example of which is Kasparov vs Seirawan, 1989.

(3) Here's another, by Black in the Robatsch: 1.d4 g6 2.c4 Bg7 3.Nc3 d6 4.e4 Nc6 5.d5 Nd4 6.Be3 c5 7.Nge2 Qb6 8. Nxd4 cxd4 9.Na4 dxe3?! Opening Explorer.

(4) There's also this line in the Gruenfeld: 1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.Nf3 Bg7 5.Qb3 dxc4 6.Qxc4 O-O 7.e4 Nc6 8.e5 Be6 9.exf6 Bxc4 10.fxg7 Kxg7 11.Bxc4. Opening Explorer

(5) Here's a queen sac Kasparov played in a King's Indian: http://www.chessgames.com/perl/nph-...

A few comments:
1) IMO the opening should be called the Bryntse Gambit of the Grand Prix Sicilian, as it is in several places on the Web. After 5. ... e6 Black would be fine. (White is fine too, it's just a game and I've played this several times.) 2) I was in home prep up to move 12, after that I was on my own. But I had played this *type* of position on the computer dozens of times and knew exactly what I wanted to do. 3) GM Jesse Kraai thought that the patient move 11. a3! was as mind-blowing as the original sacrifice. 4) GM Victor Mikhalevski wanted White to grab the b-pawn, 17. Bxb7?, thus showing that even a GM does not understand this opening at first glance. White is not playing for pawns. He is playing for a slow steamrollering of the entire Black position. 5) OTOH, IM Pruess said, "You keep talking about slow, slow, slow, but the one time you needed to play a fast move (20. b4!) you did." Guilty as charged! 8-) 6) Believe it or not, taking the exchange (27. Nxf7?!) was probably my worst move of the game. I should just increase my bind with b5, d4, etc. Black could have muddied the waters with 28. ... b5 29. c5 Nd5, although the computer still shows White as much better. 7) Pruess thought after the game that he could have drawn with 33. ... Ng4. But Mikhalevski redeemed himself for his earlier failure by immediately slamming down 34. f5+!!, a marvelous sweeper-sealer that (a) defends g3 and h2, (b) lures a pawn or a queen to f5 so that Rxg7 becomes a checkmate, and (c) if Black plays 34. ... Kh7 then he walks into a fatal discovery after 35. Rxg7+ Kh8 38. Rc7+. This is really the way the game should have ended. 8) Pruess was a perfect gentleman after this game, participating in all the analysis in spite of the fact that he had lost in spectacular fashion to someone rated 300 points below him. I will always be grateful to him for being such a classy opponent.

Two more things and then I'll turn the discussion over to the rest of you.

9) In response to vajeer, 9. ... e5? just blunders a pawn. After 10. Nf7 Qh4+ 11. g3 Black has to be careful not to fall into the trap 11. ... Qxg4?? 12. Nxe5+ and Black resigns. The infeasibility of 9. ... e5 is the first of many tactical points making this variation playable.

10) IM Emory Tate was playing on the next board. He claimed afterwards that he lost "because I was so distracted by the **** going on in Mackenzie's game"!