|
|
|
26. Rxa7 Rh6
|
The black queen is doing a sterling job on that long diagonal, giving me time to harass her opposite number. |
 
|
| |
|
|
27. Qg5
|
Forcing the queen off the f-file, though giving her a push on her way to g8. |
|
| |
|
|
27... Rd5
|
I wanted to play ...cb here, setting up a queenside pawn majority and freeing c5 for my bishop, but was worried about the white queen possibly swinging over to a5. |
|
| |
|
|
28. Qg8 Rd8
|
The rook gets back just in time. |
|
| |
|
|
29. Rfa1 Kb8 30. b5
|
The beginning of a little pawn march which underlined the cost of not having captured it. |
|
| |
|
|
30... Qf3
|
Intending Rf6, with pressure on both the white f-pawns. The threat to f2 keeps one white rook tied to the back rank. Of course I also had to keep the queen on the long diagonal to prevent Ra8 . |
|
| |
|
|
31. Rd7
|
An excellent move. I cannot afford ...Rxd7?, Qxf8 ! |
|
| |
|
|
31... Rc8
|
The only move. If ...Rf6, Rxd8 Kc7, the white rook gets back to d2. |
|
| |
|
|
32. b6
|
I had looked at Raa7, after which white has the opportunity to force a complicated endgame by Rdb7 . While I would have entered it nominally a point up, the lack of co-ordination of my rooks and pawns as compared with white's king's side pawns and queen made it a daunting prospect. After the present move I have the option of a perpetual check. |
|
| |
|
|
32... Rf6
|
Finally I have time for the not-so-subtle Rf6. |
|
| |
|
|
33. b7
|
If 33.Rf1, then ...Bh6, 32.Qh7 Be6 and white is in trouble. With mate imminent, I decide to set up the perpetual check. |
|
| |
|
|
33... Qxf2+
|
33 ...Qxb7 would have kept the game going, but I could not see any prospects for black. |
|
| |
|
|
34. Kh1
|
Unfortunately Gameknot's excellent site does not recognize perpetual checks, so we play out a threefold repetition. Don't bother to watch the rest. |
|
| |
|
|
34... Qf3+ 35. Kg1 Qf2+ 36. Kh1 Qf3+ 37. Kg1 Qf2+
|
1/2-1/2 |
|
| |
|
