Yesterday we played in a five-player game of 7 Ages. We have not played this game a lot because it is long (and has lots of fiddly bits), but I enjoy it when we do play.
7 Ages isn't exactly a civilization-building game (like Advanced Civilization) or a world-conquest game (like Age of Renaissance, Diplomacy, and many others). It's a little closer to History of the World, maybe -- instead of playing one civilization through the entire game you will, in theory, play several different civilizations, discarding and replacing them as they start to decline. (But no, not really like Vinci either, though maybe a little.)
That's the theory, but yesterday we only saw a few replacements, with people mostly playing their starting civilizations all the way through. In a game with an uneven distribution of possible civilizations I expect that more, but in a five-player game each person is guaranteed access to three at a time and that can't be taken away from you. I can't speak for anybody else, but what deterred me was the combination of my current civilizations doing OK score-wise and the cost -- in time and in conquering a new board position -- of starting a new one. Perhaps I do not play aggressively enough.
An aspect of the game that makes it fun, though also harder to track, is that each civilization has different victory conditions. I watched another civilization sprawl out next to mine and thought the player was setting his sights on my territories -- but his victory conditions weren't helped by conquering me, so he didn't. (And mine were only tangentially helped by conquering him, and I had other options, so I didn't.) I was playing the Assyrians (points for most land units on the board, among things), the Siamese (points for money, dominating southeast Asia, and others), and some guys in India who got points for India and world domination. ("Domination" means having the most territories.) That last seemed doomed until late in the game -- civilizations in China and elsewhere in Asia had gotten very big and I couldn't compete -- but then I realized that "world" could be anywhere, so my Indians started building boats to colonize Australia. If the game had gone two more turns we would have seen Rajeesh Columbus sail east hoping to discover Europe. :-) (There were unclaimed spaces in the Americas.)
The game ended up being very close, with the top two scores at 103 and 100, another in the 90s, one in the 80s, and one trailing behind. Everyone seemed to have fun and even the combat was good-natured, such as when someone started the Free States in the midst of a Chinese civilization and started what would be skirmishing between the two for the rest of the game.
We started at 1:00 at the beginning of age 2 (so most players were in age 1 after offsets), and -- after deciding to play a certain number of turns -- finished at around 10:00 with the lead civilization in the middle of age 5, about half the civilizations stuck in a dark age late in age 2, and everybody else spread out between. We are definitely playing more efficiently than when we first learned the game, without feeling rushed. I still think playing through all seven ages would be a weekend project at least; I expect that the game gets more complex as better military units afford better mobility.
There was a definite end-game effect as, on the last turn, everybody brought out the things they'd been holding in reserve. Perhaps a better way to limit the game is to set a limit (time, age, turn count) after which you start rolling a die to see if the game ends, increasing the probability each turn. That way you don't know when the game ends until it does; you don't get the last-minute hostile events and attacks and stuff, but you get people playing as if they expected another turn. I don't know which way is better.