Built in IA in a one-on-one match on a grid of 1000 tiles. A single round 100 game took about 1.5 hours to complete a Core i7, so all this time, the simulation is not trivial. But in general, the algorithm performed quite well, being able to win in this short time of about 17 percent of the time (from left to play a game to completion, the Monte Carlo search won just under half the time).
However, the authors questioned whether the algorithm could reach better decisions more consistent if she had access to the owner's manual, which contains various bits of advice on the strengths and weaknesses of different units, as well as some general advice on how to build an empire (stick first cities near a river, for example). So they decided to make their program to RTFM.