Increasingly, developers can import a model of a real-world city into their games, rather than designing it themselves. But what does that option really bring to the table, and isn't it a little creatively stifling?
A city is so much more than a tangle of streets and buildings. They are places of life and of systems. High-street crowds, transport networks, economics, waterworks and local weather all shape a city, making it what it is; an intricate artificial organism that lives and breathes as you pass through it.
That notion has long fascinated game developers and their players. Back in 1989 the original SimCity charmed, not through the types of neighborhood and roads it offered, but because it provided a chance to play with the systems that can make an urban sprawl behave like an animal. Years later, Grand Theft Auto’s first move into 3D with GTAIII provided another taste of what it was to play with a ‘living’ metropolis. Liberty City felt alive. Gangs would gravitate around certain areas, NPC behavior would change through the day, and trains would roll on above the roads, hinting at the private lives of local residents.
Those spaces, however, were not fused with reality. They were attempts at capturing the life of a city, but could never be like the real thing. Areas shaped and defined by the way people move though and use them over centuries.
What if, then, studios could build around real spaces, blending the realism of mapping data with the creative work of games design? A new generation of tools that give game makers the keys to real cities are emerging. And they are doing so in a way that is pragmatic and, potentially, revolutionary.
PUTTING DATA ON THE MAP
The team at Mantle Technologies is one of the outfits helping shape this trend. The developers have long been fascinated by location-based gaming. CEO and co-founder Dean Gifford was particularly keen on geo-location releases like 2011’s quietly seminal Shadow Cities. Eventually he and his colleagues got around to shaping their own, Fractured Skyline, with the core idea of crafting an experience that pulled data from the player’s real- world location, reimagining the environment around them as a science fiction setting.
"We started solving the problem of how to create 3D content that happens to be geo-locational, anywhere the player is in the world," Gifford says of a jounrey that would ultimately lead Mantle to offering a tool for other developers.
“We got a working prototype that effectively pulled mapping information in,” Gifford continues. “We did pixel testing that could identify where we can and should put game objects. Through that process we ended up with something that could take the GPS location of the player, pull mapping content down, generate a small gaming plate, and then put a whole bunch of missions and content on top of that.”
That is the basis of many of these types of approaches. Instead of having to toil to build fictional worlds, developers can drape their content over the framework of a real, organically evolved city. Pulling existing data that captures the layout and systems of many locations.
As human beings, we live in them and have this indelible understanding of how they work
Dean Gifford, Mantle
The gains for developers – and those in many other sectors, from architecture to retail – are manifold.
“The organic nature of cities and streets is really powerful for games developers,” Gifford says. “City Engine does a great job of procedurally generating blocks and city shapes, which is great, but there is nothing like reality. The number one takeaway for us from all this is that city streets are an organic natural noise that makes sense to us. As human beings we live in them and we have this indelible understanding of how they work. When you place content against those guidelines, you get this natural noise and it gives you a shape that works for your users.”
“People are familiar with these areas,” confirms Kelly Malone, vice president of product management and business development at Taqtile, another outfit working with real city data.
Taqtile is not a game developer, Malone is keen to point out. Its team builds applications of many kinds, having made a decision some 20 months ago to focus new attention on VR and AR. But Taqtile has acquired a game studio, Kihon Games, to help it better understand how game technologies are increasingly used to serve many different sectors currently embracing what VR and AR have to offer.
Taqtile’s apps have historically been about places and events. Now, in building HoloMaps, the team is working on a project for HoloLens that lets users stand over, interact with and in some cases upload 3D models of real cities, inspired by the ‘projected strategic maps’ seen in films like The Hunger Games and Prometheus.
“Cities can be iconic and add a sense of reality to a game when it takes place somewhere familiar,” Malone says. “The second thing is that, from a developer’s perspective, building content is time consuming and expensive. If you can access content that has already been built and is available, then that can really cut down on development time, letting you focus on things that are more important. In our case, instead of worrying about the geometry of buildings, we can now focus on the more important data overlays and the context we want to put around those buildings and places. That’s true whether using existing real-world locations, or if they’re dynamically generating those locations with tools like Esri’s City Engine.”
Procedural efforts like City Engine offer a generous slab of that potential, but where the opportunity is most pronounced is employing real urban spaces. It’s something a developer can do themselves, building around and employing offerings like Mapzen and Mapbox, which systems like Mantle make use of. That is a wildly complex approach though. With a tailored effort like Mantle developers can pull in and implement data in a matter or seconds or a few minutes. Which brings up an important point – prototyping.
THE UNCANNY CITY
Perhaps the most advantageous benefit of employing real-world city data to build game environments at speed is the opportunity to iterate and prototype.
“From a prototyping perspective, if you start with a blank slate, it can be very, very difficult to lock into something that feels good, or work as a space,” says Gifford. “Using mapping content, you get started off with something, and with this an ability to mix and match different aspects of city data. You get something that makes sense from a spatial perspective, and you can modify it until you have something that is more fitting to what you want for your game or other application.”
In that way, these kinds of technologies can be considered something of a location-scouting tool for games makers. If an environment has only taken seconds to build, then trying half a dozen game levels before sending all but one to the cutting room floor is much less expensive.
And, Gifford adds, working with real, identifiable city spaces can be far more creatively constructive in the prototyping stage than toiling away with grey-boxed sketches. Those environments of shaded geometry fail to inspire like a real metropolis, or are a little too uncanny.
However, one of the key challenges to building games from the map data of real cities is equally highlighted by this. If those grey boxes are brimming with detail imposed by reality, isn’t the developer losing something fundamental to their craft? Something like creative freedom?
Absolute creative freedom is rarely possible. With budgets and deadlines and technical restrictions, there’s always the argument that creative limitations can engender the best works of game design. Still, some developers may find concern in the idea of using, rather than creating, the setting of their games.
“You can still stylise an existing city using different shaders,” Malone offers in response. “If you look at how Hollywood represents cities, using these glowing green or blue wireframe models you can see a different form of focusing creativity energy.
“It can be a focus on reflecting a city, and what kind of mood you want to create, as opposed to concentrating on how you lay out streets and buildings, etc.”
These maps, in other words, offer a starting point, rather than a finished environment. It’s a point the team at Mantle agree with wholeheartedly.
“A city can provide a blank canvas to work with,” offers Gifford, before pondering the creation of the Mantle engine. “I wanted to have granular flexibility to be able to determine, from macro types of layers all the way down to micro decoration, what materials were used, what prefabs were placed, how often they are placed and so on.
“It’s a layered approach, and then within the layers there’s this granular styling you can do. It gives you a lot of freedom. You get to spin up a world, and all the time you save doing that, you get to pour into gameplay,” he asserts.
What is going to get really interesting is how we now integrate so interior views of the world.
Kelly malone, Taqtile
Many other challenges may not be immediately apparent to those using real-world city data for the first time.
“One of the things you have to consider is usage rights,” suggests Malone. “If you want to tap into a source like Bing’s 3D mapping data or Google Maps, you definitely need to check their usage rights, what you are allowed to commercialise through their API, and the copyright notices you need to display.”
Just because a technology is there, doesn’t always mean it provides the best option for a game developer.
“As beneficial as the detail of a city can be to a game designer, you have to be careful not to hang yourself with it,” Gifford concurs. “You have to be very careful that you’re pulling in all the content because you need it, as opposed to simply wanting to use something new, or feel that a city will look pretty as a backdrop.”
Over at Taqtile, there’s a similar sense that developers should be mindful of the technical reality of working in this way. “The different shaders you use can really impact performance,” Malone says. “You really need to think about what you want your game or your project to look like. That may determine what tools you use, and what approach you take. Whilst we really love the Hollywood-stylised views that you see in Iron Man and Prometheus, actually doing that would be quite taxing, and wouldn’t be suitable for a lot of experiences, because of the many draw calls required to display that.”
Despite those challenges, this approach to generating worlds has plenty of potential to thrive and become something more meaningful than a thought-provoking trend.
For one, considering that VR, AR and mixed reality all excel in offering users human-scale experiences that are inhabited, rather than simply watched on a flat screen, the potential of harnessing real-world spaces designed for humans to live within is vast.
That same knack for scale and immersion has also made the likes of VR a big hit in the world of architecture, where it is used for previsualisation, and even selling homes to prospective buyers. As
such, while developers learn the craft of designing spaces for humans to inhabit, just as an architect would, their contemporaries serving the construction industry are increasingly having to learn more from game design.
“What is going to get really interesting is looking at how we now integrate some interior views of the world,” Malone ponders. “I’m keen to see how this gets integrated into commercial applications, as well as game development. You look at Street View, for example; it’s still kind of compelling to go down and get that 360-degree view of a location.
“Now, through Google and Bing, you can even go into some buildings, and get the interior view. It’s already happening with businesses like restaurants, which now let you view their interiors.
“Imagine the next step, where that interior data begins to come in three- dimensionally. Think about some of the really compelling applications and users and games that could be built from that perspective. It’s really exciting.”
Having imported data providing not just a city’s outside spaces – but also the corridors and rooms that could make up game levels – perhaps raises more questions than it answers as a game development solution, but one thing is certain.
The world you and your players exist in today might already be you next great bit of game design. That fact alone has the potential to disrupt not only the foundations of your last grey-boxed prototype environment, but also the very foundation of the creative process.