The third game in the Master of Orion series was published by Infogrames and developed in-house by Quicksilver, and was released in February of 2003 (after about three months of delays). Before its release, this game was extremely hyped by many members of the gaming press and community, but this is nothing new. It is, like the previous games, a turn-based strategy game, set on a galactic scale.

The first Master of Orion featured an extremely simple interface, and very simple rules. These simple rules, when put together, made a pretty complex game. Each star in the galaxy could have, at most, one planet (thus making "star" and "planet" synonymous). Planets could only do a few things: improve themselves, make ships, build missile turrets, research things, and so forth. The things it could do were divided into five categories, and you could set a slider for each category to divide that planet's efforts. Though the game's graphics weren't pretty even in its day (not to mention now), this interface was reasonably elegant and easy to use. And, just to drive the point home, it was simple.

Master of Orion 2 was a bit more complex. The whole philosophy of using sliders for everything, as the first game did, was pretty much abandoned. The designers imported much of the design (most of the bits that differ from the first one) from the game Civilization 2. The diplomacy screens, for example, seem almost identical to it, and, more to the point, the planetary management stuff seems very similar as well. The concept of simply dividing a planet's efforts between five discrete categories was abandoned in favor of a system where you build buildings, which each affect the planet in some way. So, rather than adjusting the planet's ecology slider to make it terraform itself, you'd have it build the "terraforming" improvement, and maybe allocate more workers to construction to speed it up.

The result of all this is insane micromanagement. Where in MOO1 you could set a planet's sliders to a reasonable level and then just leave it alone for a while, in MOO2 you need to pay pretty close attention to what each of your planets are doing at any given time. The interface, while prefectly reasonable for what it's trying to do, is fundamentally more complex than the MOO1 interface: you have to dive two levels in to change what a planet is doing, while the MOO1 options are right there on the main galaxy screen.

So MOO1 and MOO2 are very different (and each very good) games. Obviously, there are other differences between them as well. However, this difference in the complexity of planetary management, and of the interface needed to work with it, is the main one, and usually goes right to the root of whether one prefers one game over the other.

This third game attempts to combine the best aspects of MOO1 and MOO2 into a game that fans of either game can enjoy. They do this through the fairly simple method of first making the game even more absurdly complex than MOO2, and then making a very sophisticated AI which can handle it all for you. You, the player, can then give the AI an overview of how you want to run your empire, or you can go in and micromanage to your heart's content. Sounds good, right?

The trouble is that they didn't pull this off. A player seeking to play it like it's MOO1 becomes enmeshed in the hellish system of telling the AI what to do. The easy way is to give very basic overarching commands, but this is hardly satisfying and, furthermore, boring. The hard way is to define different strategies for each of several types of planets. There are two problems with this: first, exactly how the system works is pretty opaque (the documentation for the game is very poor); and second, this process of using this system begins to border on the level of micromanagement seen in MOO2 that a MOO1 player would probably be seeking to avoid.

Someone seeking to play the game like it's MOO2 would realise that this would swiftly become an excercise in fighting your own empire's AI more than the other players. If you leave a planet's AI on, you will constantly have to change what your planets are doing if it decides it would rather do something else. If you decide to turn it off, first it's not immediately obious what things aren't now being controlled, and second it becomes very much more difficult to control the planet, as things really are more complex than MOO2.

A player new to the series would be frustrated by the fact that everything seems abstracted out a level. Rather than controlling things, you say how you want them controlled. Because the AI is the one actually playing the game, and in control of all the low-level functions, all the player is encouraged to do is tell the AI what sort of things it should be doing (Should it make lots of ships at the expense of production in other areas? Should it oppress the people more in order to put a stop to enemy spies?) and to control the movement of the fleets. Designing new ships (a mainstay of the series since the first one) can be doled out to the AI, but one can usually make a better ship than it can. Unfortunately, the interface is bulky and fairly unworkable (I actually think the decision to remove the six-ship-type limit was a very bad one in the interests of keeping ship management sane).

To put this all another way, you can control your empire as a whole easily with a series of (what are essentially) slider bars, but actually trying to change what each individual planet is doing is extremely annoying. Relying on the AI to make the choices that you would make for you is just not something I trust it enough to do. Actually controlling your emipire exactly how you want it controlled is basically impossible.

This game is a well-intentioned, heavily thought-out game packed with features, whose interface is a horrible, horrible kludge of options packed together in such a way that most things are two or more levels back from the main galaxy screen, making everything annoying as hell to deal with. On top of the fact that there's a lot to deal with, you'll find that most of your time is spent battling the interface rather than playing the game; that so much is automated seems to accent this problem (the little that is left isn't especially quick to do).

The one thing, the final thing that kicks the game in the ass in terms of it being worth the price of entry, is the documentation. All of this stuff about the game being so absurdly complicated and opaque wouldn't be so bad if they bothered to tell you about it. Instead, the documentation simply goes over roughly what the function of a button is, rather than really describing how the functions they represent work. Also not in evidence are tables, of the sort which break down the differences between the races or list the items in the tech tree. The result is that the game seems even more opaque and abstract, which really isn't a good thing. The only way around this is to blow $20 on the strategy guide, which is completely and utterly unacceptable: you should not have to buy an extra book to play a game in this day and age. Of course, online guides probably exist, but the developer should not have to rely on random internet users to create their documentation for them. It should be in the box.

Highly anticipated, grand in scope, a horrible interface, and basically no documentation. It can be enjoyed, and I admit to blowing a few hours to it, but you need to be exceedingly patient and willing to try and learn it. If you feel that you shouldn't have to spend a special effort above and beyond reading the manual and the in-game tutorials, this game is not for you.