During our game last year we reached the position
1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.Be3 e5 7.Nb3 Be6 8.f3 h5 9.Qd2 Be7 10.Nd5 Nxd5 11.exd5 Bf5 12.O-O-O Nd7 13.Kb1 Rc8 14.Bd3 Bxd3 15.Qxd3 Bg5 16.Bf2 Qc7 17.c3 b5 18.Qf5 Bh6 19.Qe4 Qc4 20.Qxc4 Rxc4 21.Rhe1 Nf6 22.Na5 Rc8 23.Nc6 Kd7 24.b3 Rhe8 25.c4 Ng8 26.Na7 Rb8 27.c5 g6 28.Nc6 Ra8 29.cxd6 f6 30.b4 Bf8 31.Na5 Bxd6 32.a3 Nh6 33.Nb7 Nf7 34.Ka2 Reb8 35.Nc5+ Ke7 36.f4 a5 37.Kb3 axb4 38.axb4 g5 39.fxe5 Bxe5 40.Ne4 Ra6 41.d6+ Kd8 42.Bd4 Bxd4 43.Rxd4 f5



to where mark made the move 44. Rf1 which came out to a dead draw.

I commented that 44. Nf6 Rxd6 45. Re8+ Kc7 46. Rd5 Rxd5 47. Nxd5+ Kb7 48. Re7+ Kc6 49. Rxf7 Kxd5 50. Rxf5+ Kc6 51. Rxg5 h4
would have likely led to a white win.
The line is nearly forced and any deviation from it is transposed back into it.
I have spent the last three days using my old analysis, rooting out from move 51, and triple checking everything up to it.

With a little help from FinalGen (an endgame tablebase generator with the ability to start from one specific point),
and the added benifit of Rybka 4.1, Houdini, and Critter I have found that there is no way to escape from a mate
in under 40 moves.

44. Nf6 Rxd6 45. Rd5 Rxd5 46.Re8+ Kc7 47. Nxd5+ Kb7 48. Re7+ Kc6 49. Rxf7 Kxd5 50. Rxf5+ Kd6 51. Rxg5 h4
52. h3 Kc6 53. Kc3 Rh8 54. Rg4 Rh5 55. Kd4 Kd6 56. Ke4 Rh8 57. Rg6+ Kc7 58.Ke5 Rh5+ 59. Kf6 Kc6 60. Rg5 Rh8
61. Kg7 Re8 62. Rg4 Re7+ 63. Kg6 Re8 64.Rxh4



The game would continue up to here for FinalGen to find the final mate.