If you take an “adjourned” position and give it to Fritz15, Stockfish, Rybka or whatever, infinite analysis will look for a win, that is, best moves by either or both sides. However, for the chess engine, is that the same as looking for a forced draw by one side?
I’m 100% not trying to be flip, but it depends on the engine and the settings.
Alex Relyea
I think that’s true. How then does one set Fritz 15 to just look for a draw for one side? That’s really my specific question.
There is a difference in a “draw” or playing for a “drawish” position. Often when a program runs out of “best” moves for both sides it might repeat the position several times as it considers those moves to be the best available. If given access to an endgame database, the engine will seek out the matching endgame to the position it is playing. Programs seek out best moves regardless of whether the position is drawish or not. Humans often keep pressing in “drawish” positions, either hoping for mistakes or because it is not clear yet whether or not the endgame is a draw with best play. While there are many “book draws”, there are often enough practical chances to justify playing on until is very clear there is no play left in the position.
OK, so how can the parameters of the engine’s analysis be adjusted to account for the position being even enough that it isn’t worth a human playing on with the risk of blundering?