Fast win in round 2 by Protej, who showed better branching factor, depth and thinking time despite being 32 bit and with a much lower NPS.
[pgn][Event "47th Amateur D6"]
[Site "ChessGUI4"]
[Date "2014.06.11"]
[Round "2.5"]
[White "FireFly 2.7.0 64-bit"]
[Black "Protej 0.5.8b"]
[Result "0-1"]
[ECO "D10"]
[PlyCount "47"]
[EventDate "2014.??.??"]
[TimeControl "40/1500:40/1500:40/1500"]
{i5 Quad} 1. d4 d5 2. c4 c6 3. Nc3 Nf6 4. e3 e6 5. Nf3 Nbd7 6. Bd3 dxc4 7. Bxc4 b5
8. Bd3 Bb7 9. O-O b4 10. Ne4 Nxe4 11. Bxe4 Bd6 12. Bd2 O-O 13. Rc1 Rc8
14. Qa4 f5 {!? N} 15. Bd3 c5 16. Ba6 {??} (16. Bc4 {??} Bxf3 17. Bxe6+ Kh8 {-+})
(16. Be2 Bc6 {=}) 16... Bxf3 {-+} 17. gxf3 Bxh2+ 18. Kg2 Qg5+ 19. Kh1 Rcd8 {?!}
20. Bb5 Rf6 21. Qa5 Bf4 22. Qxd8+ Nf8 23. Qxf6 Qh5+ 24. Qh4 {black resigns} 0-1
[/pgn]
After 15...c5 the position is critical: only 16.Be2, and possibly 16.Qd1, maintains the equality. My point is: most engines find 16.Be2 around depth 13/15. Ours reached, in the given time and hardware conditions, depth 11-14, so spotting or not this weakness was a question... of luck: just a little faster time control and Protej could have missed it, and just a little slower one and Firefly could have spotted it. Do you agree? Obviously this assumption may be verified with testing; this is only... human reasoning. And yes, you can say that, basically, it works this way for any move of any match between any two engines
FEN: 2rq1rk1/pb1n2pp/3bp3/2p2p2/Qp1P4/3BPN2/PP1B1PPP/2R2RK1 w - - 0 16
Alex
