This position is under analysis now.
45 hours?
seems to be so.
in the position I presented above, in 40 minutes only 33 plies depth is calculated.
Yeah, and I'm thinking that a minimum of depth 60 is needed to get the solution?
@tn85N, hardware specs?
Yeah, and I'm thinking that a minimum of depth 60 is needed to get the solution?
@tn85N, hardware specs?
45 hours with MultiPV = 20. Hardware is i7-980x (6 cores). Approx 7-8M nodes/sec.
Reported ply is not always equal to the deepest search depth reached, so probably reported 60 ply is not needed.
Stockfish 8 (without EGTB).
black: 8p & K & B(Q)
33 +99.96 1055G 45h 1.d4 d5 2.Qd3 Kd8 3.Qxh7 Bd7 4.Qxg7 f6 5.Qg8+ Be8 6.Qxd5+ Bd7 7.Qxb7 e6 8.Qb8+ Ke7 9.Bh6 Bc6 10.Qxc7+ Bd7 11.Qb8 Be8 12.Qxa7+ Bd7 13.Qc5+ Kd8 14.Qa5+ Kc8 15.Qa8+ Kc7 16.Bf4+ e5 17.dxe5 Bb5 18.exf6+ Kb6 19.Be3+ Kc7 20.Qa7+ Kd6 21.Qb6+ Bc6 22.Bf4+ Kd5 23.Qc7 Ke6 24.Qxc6+ Kf7 25.Qd5+ Kg6 26.Qg8+ Kxf6 27.Bd6 Kf5 28.Qf7+ Ke4 29.Qf3+ Kd4 30.e3# (in 30 moves)
I think this may be the reason we need Komodo or Houdini. at the time I reached the 13th move, I thought there must be better moves than 13 Qc5 and the other engine approved it. let us compare it:
Not necessarily. Stockfish 8 currently at ply 33 (...). As it will go deeper both the moves of 1-12 and above 12- will probably change.
I update the above results continuously.