Battlefield 3's executive producer, on balancing realism in Frostbite 2

[jkipp82]

It's been over five years since Battlefield 2. Although DICE has had concepts for a sequel in mind since then, the real work could only begin once the team had the tool Battlefield 3 demanded: a new engine. Why? "The short answer is this: No other engine can build a Battlefield game."

"If you don't have the right tech, you can't build the right experience," DICE executive producer Patrick Bach told us. Unlike other engines, which are meant to be repurposed and packaged for third-party licensors, Frostbite's reason for existence is singular: "The Frostbite engine is solely created to make Battlefield games."

Battlefield 3 is unnervingly beautiful. The GDC presentation is one of those rare moments where you see a new graphical benchmark being defined. Bach claims they are the "first ones in the next generation." Seeing the game running on a souped-up PC, it's hard to disagree.


There's no doubt the PC version of Battlefield 3 will impress, but how is it possible to bring a "next-gen" engine, like Frostbite 2, to the current slate of home consoles? "We're streaming all the data. There's not a single frame where nothing streams. From textures to objects to weapons to sounds and animations ... it scales down pretty well. All of the core systems will use streaming, so it's more of a question of how often you have to stream, and how big are those packages. The engine is built with that taken into account." DICE hopes that the streaming tech will enable the console versions to keep the mood and feel of the original intact. Still, not seeing a working console build did little to quell my doubts.

Given the focus on visual realism, I wondered if there would be an increased emphasis on realism through the gameplay. The brief demo of single player showed a campaign with far more gravitas than a Bad Company game. The game's reference films -- Hurt Locker, Generation Kill, Black Hawk Down -- show a more "adult" approach to storytelling. And, EA's other FPS franchise, Medal of Honor (which featured multiplayer developed by DICE), was marketed as a more realistic alternative to other popular shooters. "Authenticity is very important for all Battlefield games ... but it's always fun first. If it's not fun, then there's no point."

If realism isn't the end goal for Battlefield, then why bother with making such a bleeding edge engine, then? Because the visuals are part of the fun, Bach explains. "We see realism as a big part of the entertainment in a modern shooter. People know what guns and gear look like. It needs to sound real. It needs to look real. In general, we start with realism, and then we balance it to make it fun."


Although the engine keeps track of individual bullet balistics, and realistically recreates the penetration and destruction of walls, Battlefield is far from a "simulation." In fact, Bach calls out gamers that are singularly focused on realism in shooters. "It's bullshit. There are no simulations ... there are so many things that are not only visual. There's weight, wind ... you can't simulate that in a game. There's always a filter ... there is no absolute arcade, no absolute simulation."

Some are surprised that Battlefield 3 is even bothering with a single player campaign. The franchise has built its loyal fanbase thanks to its multiplayer offerings. Is it even necessary to bother with a campaign? Bach unsurprisingly stands by his product. "We got better and better at making single player thanks to the Bad Company series," he says. "[But] even if you don't like single player, you know what? Multiplayer is bigger than Battlefield 2 anyways. People shouldn't think we're spending energy on single player when we could have spent it on multiplayer, because we have a full-fledged team working on multiplayer."

Given the six year gap between Battlefield 2 and 3, I asked if Battlefield 4 would require a hypothethical new engine beyond this one. It's unlikely, as Bach seems to think the team has matured to more regularly work on the core franchise. "I think we learned by experimenting with these other things (Bad Company, Battlefield 1943, etc.) that the core is quite powerful. We should probably spend more energy on the core than on these experimental things."

But don't think this is the end of the Bad Company franchise. (Or Mirror's Edge!) Building a brand-new engine is expensive, and DICE will want to get the most out of it before moving on. "I wouldn't be surprised we saw four, five games on this engine."

[Resdayn]

I think this will be the years best game, Those guys just really know how to make a good game.