April 13, 2011
By Jimmy Thang
While the Bad Company games have kept the Battlefield name relevant the last few years, hardcore PC gamers have been clamouring for a true Battlefield game for over half a decade. With highly-anticipated shooter Battlefield 3, Swedish developer DICE is promising to deliver the biggest, best, and most beautiful game in the series yet.
We spoke to Battlefield 3 Executive Producer Patrick Bach, a man who firmly believes consoles are holding PC gaming back. Like PC gaming enthusiast, he is also dismayed by the fact that many game developers merely add anti-aliasing options to the PC version of games before calling it a day. Eager to buck this trend, Bach talks about how DICE will create one of the most visually-breathtaking games you've ever seen by using the PC as Battlefield 3's lead development platform. In this interview, Bach also touches upon amazing achievements with their new Frostbite 2 engine and touches upon the possibility of one day achieving photorealistic graphics.
GeForce.com: In regards to the title, "Battlefield 3," is there a reason you wanted to bring the mainline back, as opposed to creating another Bad Company game?
Patrick Bach: I think we've been all waiting for Battlefield 3, including people at DICE. We've been wanting to build this game for quite some time, but you know, the technology wasn't ready for it so we just had to wait. So we did other games in the meantime, which I think was a good thing, because you know, games are not about technology foremost. It's not what it's about, but if you don’t have the appropriate technology, you might not reach your vision and looking at Bad Company 2, for instance, we used a souped-up version of the Frostbite one engine and we actually achieved a lot more compared to Bad Company one. I think that’s the big benefit with having a good core engine, you can actually do iterations and make it slightly better, but then you come to a point where you just have to say, “Okay, if we want to make a big step forward, what do we need to do?” And you end up saying, “Okay, we need to rewrite, we need to redo it from scratch.” I think you can achieve a lot by changing art direction, story, core game design, and balancing stuff like that, but technology is a big part of gaming. It's based on super advanced technology and we just need to be aware of that and utilize whatever things we can to make our vision come true. Because, you know, gaming today is probably the most advanced piece of technology combined with the most talented people in any business. It's both a very creative and very technical business at the same time.
Do you feel the consoles are holding PC games back?
Yes, absolutely. That's the biggest problem we have today. Most games are actually still based on the same core idea that the consoles are your focus, the superior platform or something. I don’t know why. That was the truth 5 years ago, but the world has moved on. PCs are way more powerful than the consoles today and there are actually almost zero games out there that actually use the benefits of this. So for our target of what we want to hit, we are now using the more powerful platform to try and prove what we see gaming being in the future rather than using the lowest common denominator, instead of developing it for the consoles and then just adding higher resolution textures and anti-aliasing for the PC version. We're do it the other way around, we start with the highest-end technology that we can come up with and then scale it back to the consoles.
Can you talk about the benefits of that? That probably makes the PC version look better but does it then hinder the consoles in any way?
This is interesting. In theory you could argue that you're building it for a more powerful platform and that it will look crappy on consoles. That's not the case because when you build the target high, you can then pick and choose from the target and ask what actually creates this picture and then pick the best things from that and turn that into your console solution. The other thing that is very interesting is that since you're building the engine based on the knowledge that you will release it on the consoles, you actually streamline the whole technology to be able to scale back to the weaker platforms. So in the end, by, you know, painting this high-end target, you actually set a new bar and then when you scale it back, I think people will be surprised to see how good it looks on the consoles. We can't show it right now, because we're aiming to use the PC to set the bar, but it's actually helping us make a better console game.
Can you go into detail as to why DICE decided to go with a new engine altogether? How was the old engine holding you back?
We couldn't build the game if we didn't have the new engine.
That's a bold statement.
It's true. There's no looking back. Historically, we've always been more or less, you could say… we've been forced to build our own engines. There's no engine you could buy today that could build a Battlefield game to the quality that we're building Battlefield games. It's actually quite a complicated process to build a Battlefield game cause you are more or less picking the hardest thing to build, big open landscapes with quite high detail when you zoom in on it, it's an infantry ground focused game, but there are also air vehicles and they all need to work together and then you have all the physics involved with all the trajectories and bullets in the air and that has to be synced over a network and then all the players need to see what you're seeing. There's a lot of stuff going on in the background as well. We are extremely focused on making it look great.
For those unfamiliar with the Battlefield franchise or just the Bad Company series, can you explain what all the buzz behind Battlefield 3 is about?
It is the successor of course to Battlefield 2. It's not supposed to be connected to the Bad Company series. It's actually been more than five years since Battlefield 2, which, of course, has made us think a lot about how this next big Battlefield game should be. One of the conclusions that we made quite early was that if you really want to move gaming forward, you need to go back to the original idea. Look at the core of what you want to achieve rather than to just do an iteration on our existing technology for instance. We actually designed the game based on what we wanted to see in the future rather than what can merely be built today. And then we realized we had a lot of problems with the technology we had so we went back to the drawing board and just redid the whole engine based on our needs, rather than doing it the other way around. I think that's one of the big reasons why it's been taking so long for us to release anything from Battlefield 3 because the technology wasn't done. And also the fact that if you look back 5 years when the consoles, the high definition consoles, were released, they were actually better or good as high-end PCs back then.
Are there going to be any new destructible enhancements to Battlefield 3? Bad Company did a really good job with that.
Destruction was a bold move when we did it for the first Bad Company. We actually did the same back then, we asked ourselves, “what will be the new cool thing in 5 years that everyone will have?” And this is back in 2004 and still no one is really building destruction which is a bit weird in a way because you would have thought that the world would have moved on when it came to static worlds. There's a lot of work involved in it so the first Frostbite engine focused on making destruction just possible and it's not all about eye candy, of course, it's actually a strategic element of the Bad Company games so that using the destruction as a strategy, something that you need to adapt to, as it actually adds a lot of variety to the Battlefield core, and now we're taking that back into the core of Battlefield series. Battlefield 2 didn't have any destruction so Battlefield 3 will be the first core title with destruction and of course we're doing some bold moves by not only being in the more urban environments, where we've been before, but we’re going into cities now, so you need to have even more destruction where you can actually affect objects - the buildings around you, to make it feel like a physical world.
Regarding why destructibility hasn't been fully embraced in games, it probably has to do with the fact that, unlike Frostbite, most engines are not capable of it and for developers, it's just cheaper to use a prebuilt engine.
To that end, do you believe that every triple AAA game should have its own proprietary engine? Furthermore, what are the pros and cons of designing your own engine?
I think cost is the biggest drawback, of course. We've been working with the Frostbite 2 engine for more than three years. There are a lot of people in the team only working on the engine. It takes a lot of muscle from the company you're working for to put the money into it and kind of wait for the kind of payback later. It's not a safe bet to build your own engine, there's a lot of engines out there that have been failing and then you've been spending millions of dollars on something that actually ended up being worse than that and more expensive than an engine you could have bought.
First of all, you need to have very talented people building the engine so that you know that it's actually better than what you could have gone out and shopped for. And also, you've got to make sure that it pays back in the end, Battlefield is such a big brand now, so we've been building Battlefield games based on the Frostbite one engine for quite some time so we made the investment back. And also, you can actually see the benefit of the engine in the games like our destruction, for instance. We don’t see that in any other games because they don’t have that as a core element of their engine, we have it as a core element because that's what we designed it for.
Are you guys going to license the engine out?
No. We are owned by EA and we don't want anyone else to get the benefit of using our engine because we are big enough to make money on our games based on the engine, so we don't have to sell the engine itself.
Was there much hesitation with designing an engine for the first one?
Oh yes! Oh yes! There was a lot of debate! It takes a long time until you... when you go from having nothing to having something where you can actually see the benefit. That takes a lot of energy, a lot of time, and you'll also learn a lot over the course of time, so you can't really go back and just change stuff, you have to kind of keep going. And then you have to ask yourself, “When do you start to build the actual game? How long can you wait?” That's also the challenge when you don't have anything at all. So I think the Frostbite 2 engine is more or less based on the learnings from the Frostbite one engine and we rewrote more or less all of it based on those learnings. Like now, we know what we shouldn't do, we know what we should do, and we know what we should do better, and I think, for instance, rendering is a big part of that. The whole rendering model is completely rewritten and is actually interesting when you look at Battlefield 3. The screenshots or the imagery we have out right now, you can actually see that it looks quite different than any other game out there because it's such a unique rendering engine and that of course helps us to build a better game.
For me, when I first saw the gameplay trailer, it was sort of reminiscent of that first Killzone 2 trailer.
Oh, you mean the one that was fake?" (laughs) That's good.
So what exactly is this engine allowing you to do with Battlefield 3 that you couldn't otherwise do with Frostbite one?
First of all, as mentioned, I think the rendering is completely new, we're using a completely different rendering model. Deferred lighting, we're using dynamic radiosity, combined with all the particle systems being rendered in the same world, looking completely different than what we've seen before in our engine, at least. That combined with, for instance, our animation system that will be working with a central EA tech service, where they created this animation system called "ANT," primarily for EA sports titles. So you can see it being used in FIFA, for instance, and the characters in FIFA look amazing, probably the best looking sports game out there, but taking that into first person experience was quite cumbersome, it took quite a lot of energy and time, quite a lot of collaboration to get that going, but if you look at the result today in the images you've seen from Battlefield 3, we're taking this to a completely new level. It looks completely different than other first-person shooters, just the fluidity of animations at any time is unseen in other games, I would argue. For us, that's a huge step forward.
So we have the rendering, we have the animation, and we have the improved destruction that we've talked about. We're going into the cities, and then, of course, the audio is something that we've been really good at before, I would argue that we're the best in class but we still have improvements to do and we're taking that step now with Frostbite 2. So there's lot of bits and pieces from everywhere that adds to the whole package, it's not just one part. It's actually all the parts that make a better game. We also have of course, the benefit that the consumer might not see directly, but it's on the development side. We have faster iteration times. We have sub-level streaming, you can actually stream things in real time on any level, in and out of memory, it's super-fast, which gives us room to scale down high-end PCs to console.
So performance should be pretty good across the board?
Performance should be very manageable cause of this because we can choose where to add or remove stuff in run time, which is amazing.
In terms of the visual style, are you guys trying to go for realism? Is that the end goal? Because it does look very realistic.
Yea. The goal is to give you a realistic kind of rendition of the world so it feels like, “Okay, I understand this world. I see people, I see the world, I can recognize myself in it,” and then add all the gameplay elements of the core Battlefield experience into that and if you remember in 2005 when people played Battlefield 2, everyone said that was the most realistic-looking shooter at the time, but if you look at it today, compared to what we have, it looks like a cartoon. (laughs) It's way better today and I think we can of course, in the future, hope that we can take another step forward so games should not look like games, but games should look like what you want to create. If that’s reality or more stylistic cartoonish style or whatever, then we can't blame technology anymore, because then we'll have taken that step. We can then in theory create whatever we want. And I think to us, a very creative team, that's the big benefit. Then it becomes a discussion of not what can we do, but what do we want, which puts a lot of pressure in the creative part, of course. Back in the day you could blame technology, but now it's like, “Okay, we can do whatever we want, we don't have limitations."
Do you think we'll ever get to a point where it will become impossible to distinguish between real life and a videogame?
...hmmm... Yes? Maybe. But I still think there's a lot of…well, I can't see it being in the close future. Because there will always be people, it depends on who you ask, if you talk to the experts, the developers, they will see right through that. I remember when I saw Gran Turismo 2 for the first time, I was blown away. I saw it as photo-real. When I look at it today, it's like, “That’s not photorealistic. That's a game! It looks crappy.” (laughs) So I think it has to do with what you relate to, but you know, getting to reality is a big step, it's really hard to get there, but I think we're closing in, and at moments, in even our game sometimes, we actually have to ask each other, “Okay, is this part a rendered movie or the actual game?... Oh, it’s the actual game? Cool." So I think we can get there at times, but you know, creating a whole experience looking like real life, I think that’s really hard. There's so many small details.
Moving on, let's talk about the game's single player. That was never a big focus for the main Battlefield franchise; yet, from the trailers, it seems like this one's going to be pushing a campaign pretty hard.
I won't go into any details on the single-player. We're not releasing any info on that right now. In general, we want prove with the footage we've shown so far that we know what we're doing and we have a really strong idea on how we want the game to feel and look, and so I won't go into any details.
Will it have co-op?
We will have co-op in the game, but I won't go into details for what that means...
Interesting... *strokes chin*
Before we began the interview, we talked about how there was this perception that the Battlefield franchise has been "consolized" with Bad Company. It seems like DICE is really trying to get away from that perception.
How are you addressing that?
Well, first of all, we're using the PC as the lead platform. We’re setting the target with our high-end PCs, making sure that when we show footage now, it's on the PC. And we are a PC studio from the core and also looking at the audience today, the hardcore PC gamer is no different than the hardcore console gamer. That was not the case five years ago. Then, you had the more casual console audience and the more hardcore PC audience. I think that is very blurred today. People are extremely hardcore on consoles today. Extremely hardcore. So we don't have to dumb the game down in any way to make it fit on the console. It's still a very hardcore game, but then you still want the game to be accessible and some PC users might think that it’s a good thing that games are complicated. We do not agree. Battlefield 2 was a very accessible game. It's real easy to get into, but it's super deep and really hard to master and that’s what we wanted to achieve with battlefield 3.
In multiplayer, how many players will the PC version support?
And the console version?
So that’s one of the benefits of being a PC user right there.
That's one of the benefits of the PC. More memory, more bandwidth. The network bandwidth is higher on the PC. Consoles have a limitation, which is a problem for us, but we want to create the same experience on the consoles. 24 players is still more than almost anyone has on the consoles. In addition to that, we have vehicles and infantry on the same map.
From a gameplay perspective, what can we expect to be different about Battlefield 3?
It's going to be bigger. It has to be big, we're going to have jets!
Are they going to be hard to take down?
Our goal is to make a fun game first - we don’t put anything into the game that is not balanced. That’s not how we do things.
In real life, I wouldn't know how to shoot down a jet...
There will be plenty of ways to take down a jet. There will also be plenty of ways to defend yourself from getting shot down as well. It’s a part of the rock, paper, scissor way of thinking that we have. We don't necessarily try to see things from the shooters' perspective, but from the getting-shot-at perspective. We focus on how you can defend yourself given any weapon you might have, we then try to give you options and that's what creates the magic.
Patrick, thank you for your time.