the engine is played with the numpad. it is very self explaining.
- first the egine ask my to play or type 0 to let it move.
- I play on square 5
- My move is on the board
- The engine answers
- The engine give back his evaluation
- 0 = draw, 1000 = X is winning, -1000 = O is winning
- the evaluation is changes that the ai wins against random ai
- in this example 742 means that O has 74.2% change of wining against random moves
good luck
If you type 100 you enter the menu.
- option 1: is to start a new game
- option 2: is to change the AI to the AIMulti.cs class, this is the ai of v1.1. (I don't clear the transpo table between moves so i would think of this as a cheat)
- option 2: can also be to change it back
- option 3: is to set the AI to a random Move maker
- option 3: can also be to change it back
- option 4: benches singlethreated ai with a blank board
- option 5: benches multithreated ai with a blank board
- option 6: benches the transpo ai with a blank board and after 2 moves.
- option 7: benches multithreated transpo ai
- option 8: runs a full benchmarking suite with:
- multi to warm up the cores
- all the ai from a startup config and a blank board
- all the ai with a second move
- the ai's (running) from startup
- option 9: allows you to let the ai play vs a random move generator and see the results