The 5 Minute Surprise

I've been a life long Grunfeld player, ever since I discovered that I could take the rook on a1 with my g7 bishop. Of course to real chess players this isn't a reason for choosing an opening (pressure against the centre, latent piece activity they say), but for a hacker like me, it is good enough. So in a small lightning tournament at the ANU Chess Club last night I was shocked by a move I had never seen, or considered, before.
Playing Black against Mario Palma I reflexively spat out the standard moves 1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.cxd5 Nxd5 5.e4 Nxc3 6.bxc3 Bg7 and was taken aback when my opponent played 7.f4 (D) "Hang on" I said, "you cant play that" but in lightning distraction is often death, so I quickly played the thematic 7. ... c5 Now given my befuddled state it was hardly surprising the game went quickly south.
8.Nf3 [ 8.Bb5+ Bd7 9.Bxd7+ Nxd7= was given by Fritz while 8.Be3 Qa5 9.Kf2 0-0 10.Qb3 cxd4 11.cxd4 Nd7 was given by Hartston in The Grunfeld Defence (1971)] 8...0-0 9.e5 Nc6? 10.d5 Nb8 11.c4 and White just has the massive, and well defended centre.
It then got even worse (then better!), as I dropped a piece through an elementary blunder (Nd7-b6xc4 with the bishop still on e2) before I was able to win the d pawn, and push my queenside pawns up the board, and win.
What then surprised me was when I checked my database at home, how little this move has been played (I found about 30 games). While White does score badly I haven't been able to find a clear and outright refutation of 7.f4 (Hartston simply says "with good play for Black")
So if you are looking for a shock weapon against the Grunfeld then Mario's 7.f4 may fit the bill.