As the position 1.e4 Nh6 2.d4 g6 3.h4 d5!? 4.exd5 doesn't seem to have been played in master tournaments, an engine match has been set up between two chess engines, under these conditions:
Strelka 5.1 This UCI engine, by Jury Osipov, was suspected to be a Rybka 1.0 clone an it was banned from engine tournaments, although later some experts studied its code and they found that this was not the case, if my memory supports me correctly. This is a very strong engine indeed.
GUI: Fritz 9
Hardware: ACER Netbook, CPU N270, 1,60HHz, 1GB RAM;
No permanent brain;
Engines hashtable size: 156 MB;
Time control: 10 minutes per game plus 5 seconds per move for each player.
After each engine move you will fine a few info. For example, if you take Black's 4th move 4...Qxd5 in the first game, you find "1.03 15 33" this means that according to the engine moving, (the evaluation is expressed in pawn units always from the point of view of White) that the program displaying a value of 1.03, it means that it considers the white position to be better by the equivalent of 1.03 pawns. Also the search depth "15 33" tells you how many moves deep the search has progressed. The value is in “ply” or half-moves. The first number gives you the “brute force” depth, the second is the depth to which certain critical lines have been investigated. So the display of “depth = 15 33 ” means that the program has looked at every continuation to a depth of 15 ply (seven moves and a half), while some promising or dangerous continuations are being examined up to a depth of 33 ply.
This is not to be considered a serious match for the following reasons:
A) The hardware used is a slow one;
B) I adjudicated the first 2 games in a won position;
C) For unknown reasons, the last game "stopped" in the middle game position as you will see.
Please feel free to draw your own conclusions on this little experiment. On my part, I just conclude saying that it so much fun to play an early Nh6 move :-))