That's totally true, but that's also the hardest market to break in to. How many Sims clones are out there? The MMORPG market is over saturated for the most part, making it a risky venture, and games like Civilization can't be cheap to make. Sure, a ton of clones of these games get released, but very very rarely with any money behind them and they are almost never successful. Because in order to compete with The Sims you have to be as good as the Sims. Not only that, but you have to offer everything it offers, more, and make enough of a name for yourself in order to make someone stop playing something they were compulsively enjoying and give your game a try.
So, if you're a game company and you want to pump some money into a PC game, you probably want to take as little risk on it as possible and hope for a high return. How do you do that? You appeal to the type of person who buys 2-3 games a month, the type of person with quite a bit of disposable income, and the type of person who's prone to playing PC games.
Generally, this is the type of person who has a somewhat new system because they just bought it or they're upgrading every 12-18 months. What does this type of person like to play? FPS, action games, and RTS games. How do you compete with the other ones out there? Well, it's a pretty generic formula for the most part, but what has always worked in the past was to make the game look good and people will buy it. So you either develop your own engine or lease one that's already out there, push it a little past it's limit or just sloppily code it, throw in a sewer level, an escort level, some light squad-based gameplay, some lighting effects, a rechargeable flashlight, some middleware. You go down to the movie store and look at old 80's action movies. Find an obscure one with a title you like and call it that.
Then when you're done and you realize your game is pretty generic and kind of runs like ass, you go back, throw some bigger textures in there, pump up the triangles, and add some pointless and inefficient post processing effects. You go into an interview and claim that the game is made for the next generation of video cards, and that the ones from a year and a half ago won't even really run it at any really playable level, and just watch the hype build.
Next thing you know they're using it for benchmarking, and every answer to a "why does this run so slow on my current card?" topic on forums is answered with "SLI NOOB! 100% more money for ~20% more performance is where it's at!". So you can sit back as Ram prices go down and people buy new cards and think to yourself how even though you didn't make any sales records with the last one, you at least broke even. And now you know a bit more, you're a little more experienced and this time when you release your requirements before the demo is out you'll be sure you're the first on the block to have 3-way sli and 3 GB ram in the recommended section.
No, it's not the easiest way to make the next big phenomenon, but it's the most market tested way to make a return on an investment.
