Engines or Assets

A strange thought occurred to me a few days ago – is it feasible to license gameworlds?  Probably one of the most time-consuming and expensive parts of games creation is the extensive art design, asset building and debugging of virtual worlds.  This is particularly true of 3D games, but is also a factor in 2D games.  What if there were companies whose sole output is original area/world designs for use with other people’s game ideas? 

Rockstar could license Liberty City, for example, to third-party developers, why don’t they do it?
The most obvious benefit would be ease of development for smaller companies and lower costs.  While certain engines come with Software Development Kits and pre-built assets, there is still a great deal of work to do putting all of that together.  If a company could use a pre-built city or area, they would be free to concentrate on what makes their game unique.  It would be possible for game companies to produce AAA looking titles without the same cost. 
Its not all about the little guy, larger asset creation companies could re-coup development costs by licensing their worlds.  Movies, TV shows and the Music industry have been doing something similar for years.  Movies are sold again on DVD and Blu-ray, TV Shows are syndicated in different territories across the world and can be co-funded by several companies and music is licensed effectively on radio, TV, film and in games.  While individual games can be licensed for use in TV and Films, usually as a minor plot point or backdrop item, and companies are keen to license game engines, I don’t think that world assets have ever been exploited in this way.  If you could sign up several companies to co-fund a world for all of your games, or agree to use the assets you’re developing before starting work, then you could increase your budget and produce more impressive assets as a result.
Maybe there is a niche market for specialist asset houses, like movie prop companies, that focus solely on creating and maintaining interesting and unique worlds for licensing?  A talented bunch of artists could create worlds unlike any we’ve seen, free from the constraints of writer and game designer pressure.
There are a few small issues with this naive utopic vision.
Game companies are fiercely protective of their intellectual property, so are unlikely to license AAA content.  If the licensed content was used to create a superior game this would be doubly embarrassing.  The most likely source for game words would seem to be third parties.  The problem with this is that it is very niche.  If you develop a hi-fidelity city, like Detroit in Deus Ex Human Revolution, the cost would be in the millions, to make it cost-effective it would need to be licensed many times, or else the producer wouldn’t make any money.  How many games developers are looking for, and able to afford the licensing costs of, a futuristic cyberpunk city?  The cost of the license would be less than the cost of development, but you wouldn’t own the assets.
Unless game companies agree on a defacto standard for modelling, physics engines, game engines and sound design then there will always be different choices for each: 3DS Max or Lightwave, Source or Unreal, Havok or Euphoria?  Dependent on the engine there are going to be limitations to what types of games are more naturally inclined to each.  The problem is that the game company doesn’t get to choose what systems are used; they have to go with whatever the asset creator used.  This is great if you’re making a FPS in New York and someone already made an Unreal version of the city, but if you’re making an isometric crime simulator, it could be less than useful.
If you don’t own the world, then you can’t change it.  If your game is set in London and you want to move the location of Big Ben to be slightly closer to St Paul’s Cathedral, because it would take too long to get between the areas otherwise, well, tough, you’re stuck with what you’ve got.  There are ways around this, but depending on the design choices you’ve made in the game, an obvious cut to a different location might be jarring and take the player out of the game.
Game companies don’t generally use shared assets internally.  What I mean by this is that if a company, like Take-Two Interactive, has a Baseball Bat model for their Baseball games, developed by Visual Concepts, they don’t use that baseball bat model in GTA4, developed by Rockstar North.  I realise that Take-two is a publisher, not a game developer, but if they can’t get their in-house teams working from a shared asset pool to reduce costs, then I think its a stretch for companies in competition to work from a shared asset pool.
While digital assets don’t degrade, they don’t age very well either.  When graphics capabilities increase, the relative worth of the assets decreases as it becomes obvious the models don’t look as real as they could, and it becomes less and less desirable to use outdated content.  You can update the content, but that means updating all of it, otherwise you’re left with mismatched items that have the potential to be seen side by side.
If these could be overcome you’ve still got a few more esoteric problems to contend with.
Everything would look like Unreal Tournament or GTA4.  If you think that all games look alike now, with brown-grey space marines running around brown-grey war-torn environments, then its going to be even worse when you realise that you’ve just run down the same corridor for the millionth time in two completely separate games.  A related issue is that if the Gears of War assets were used to create another run and gun cover shooter what would be the point?  If game companies can see that certain games are selling well and that you can buy the assets of the games off the shelf, you can be sure that they’ll attempt to cash in on the craze and produce an identical game.
There are other things we could potentially lose if people have a lower-cost option, as some of the best games have been defined by their limitations.  If you take away the need to create interesting low-cost graphics, you are less likely to get games like Darwinia or Space Giraffe that had to creatively overcome the lack of budget with interesting gameplay and graphics.

So, I think we’ve established it’s an interesting idea that’s not going anywhere, anytime soon, right?

EDIT: I spoke to soon, after a little digging: Battlefield 3 shares technology with EA Sports titles