Technology, Design: Rage

Everyone has their eyes on id Software whenever it makes new product announcements -- not least because these often coincide with new iterations of the company's game engine, id Tech. The same holds true of Rage, which is due to be released in 2011.

The title expands its gameplay palette from the shooting that the team has previously concentrated on by expanding the gameworld into a sprawling post-apocalyptic world. The technology, too, changes the fundamental way the company approaches its games, by introducing its megatexture techology which allows for massive background textures to match the game's environments.

Here, Matt Hooper, id's design director, discusses both the gameplay and technology changes that have resulted in Rage, and how the very roles and team composition were changed by the company's drive to make a game unlike, on so many levels, what it had tried in the past.

How has your role as design director changed as id has finally been back in full production on a large-scale internal title?

Matt Hooper: That's kind of a tough question, because it's evolved so much over the however many years we've been working on the project. After Doom 3, we had about 15 core developers, and now we have something closer to 60.

So, not only my role but the roles of a lot of people have changed over time. In the past, I worked very close with John [Carmack] on the technology, then we cracked that and I moved forward from more of a design perspective. But since I have some programming and technical abilities, it was nice working with John in the early days and pushing the direction of the technology.

Now, I spend most of my time working with our group of talented designers on the moment-to-moment interactions, including the different weapons and making sure they feel right. It's really more pure design now.

As you suggested, Rage has been in development for quite a while. Can you pin that down?

MH: In total, it's been probably about six years. Again, the company was smaller. We worked on other projects, some external projects and things like that. As far as actual production, after the initial technology and pre-production, it's been three years.

The scale of this game, the sheer volume of assets, is very much a departure for id, and that's reflected in the fact that the team size is now four times what it was after Doom 3. Has that changed the company philosophically, how you think about development?

MH: You know, the funny thing [is that] this is no different from id in the past. We were the first people to push dynamic lighting, and bump maps on characters and worlds. Even Quake 1 was one of the first truly 3D environments. And even with Doom, we've always done this -- disruptive technology. It's technology that comes in and completely changes the way you build and make games.

Some of the core aspects of game design are still there, but every new iteration of the [id] tech, especially when it's so fundamentally different, turns everything on its head, [in terms of] the way the artists have to work. This idea of being able to go back into the environments and add that last layer of detail with stamping – "unique-ing" all of the different texture space -- there's a whole new group of artists who do that.

We found that some people are better at that. Some people who might not fit as an artist on Doom 3 now have a tremendous worth on fine-tuning different environments. It's turned us on our head. We've had to completely re-figure out how to build games with the new tech. But that's what pre-production is for. Now we kind of know what we're doing, and we have a method for getting it done.

Do you think those new roles, for example the detail-level artist you mentioned, will be codified as long term development roles?

MH: I wonder. I'm not an artist, but from the outside, I see more traditional art. In the old days, you just make a whole character and animate it on 64 by 64 pixels, and that took a certain type of art. Then we moved to bump maps, where you need to make the high-poly models. The low-poly guys maybe couldn't do that, so you opened up to new artists. Now we have animations that look like movies, so we have a lot of movie guys coming in.

I think as the technology evolves, we're noticing there are groups of artists who can make tangible contributions with everything, from more concept art... You just have more screen resolution. A lot of traditional art is making its way into the game, and it's becoming slightly less technical. It does require a new group of artists who maybe don't have the technical ability, to bring their craft.

Speaking of traditional art, I was actually surprised when I saw the game how much really expressive hand-keyed animation there was, as opposed to the more sterile mo-cap kind of stuff. That's unusual for id as well.

MH: The credit goes to our group of animators, a really talented group. As we were iterating through and trying different things, we kept pushing diversity. We want each of the characters to look unique. The different characters look so wild and out there that it almost demands that over-the-top animation treatment. We really pick and choose the opportunities. Some are a little more -- I don't want to say cartoony, but they're a little more expressive.

The [animators] look at the character and they look at the emotion we said we wanted. When we first saw it, we had the same [reaction]. It's like, "Is this right? It's different from other games, but it's cool, and we like it." It makes the characters stand out. It didn't start out as "This is what we want." We just worked through it, and it made sense.

The writing and design side also seems very unlike most id games. There are a lot of interlocking world systems – quest NPCs with dialogue, more connected locations, things like that. How did you determine the upper complexity threshold there? You've stressed this is still an action game.

MH: Well, to be honest, we probably went too far. We probably have too many systems. That's why the game has taken as long as it has. But, in the end, it's going to be well worth it. We really didn't know. We just keep pushing each of the systems as far as we can in the time we have.

For the writing, Tim [Willits] does most of the top-level writing and the specific missions. But again, we're just looking for opportunities as we move through the design: "Hey, we need to add a sniper mission on the top of this mountain, because that's a great place for the player to have fun." We'll write some new dialogue and expand it.

But we're really conscious to keep the design and the scope of the different systems still on the action-shooter side. Those were conscious decisions. We don't want you to have to read tons of different dialogue. We'll put it in the inventory so you can read it. But we do want the story to be rich. That's why we went with this casual kind of storytelling.

When [NPC] Carlson in [a demonstration of the game] says that somebody attacked the wells, we give what we feel is just the right amount of information. You don't have to sit there for half an hour hearing this long dialogue, but you do understand what this means to the region and why you have to go down there. In the well, you overhear the guys talking about how to poison the water. They're ruthless. It gives the player context. We hope that compels the player to live through the story.

I noticed that despite the fact that the world is very decrepit and gritty, the guns don't necessarily reflect that. They have a very crisp, id-like feel and efficiency. Was that a decision you ended up with for the sake of gameplay perhaps over world immersiveness?

MH: Yeah. That is something we've always done, so we feel pretty strong about it. The control and the feedback is really what it's all about in the end. But we tried tons of different things. The Settler machine gun, for instance, has a larger spread. It's looser than the Authority machine gun you get later.

But we look for opportunities. Because it's so loose, you can get an attachment that stabilizes it. We go all over the place with weapons. But there are a number of things you would consider traditionally "id," and then extra things specific to Rage that you've never seen before.

How integral are systems like a world economy? Are you designing them with the intention to be integral to the gameplay experience, or is that a bit more peripheral for players who want more than just the action game?

MH: I think it's all integral. I feel confident that it doesn't disrupt the FPS. Yes, you do go to town. Yes, you interact with NPCs. Yes, there are opportunities to play mini-games and pick up side missions. But really, just go and talk to the merchant in Wellspring. It's a super cool 3D interface for buying, selling, and manipulating your loadout. What we found in all of the other id games was that about halfway through, you had a billion shotgun shells and none of the thing you want.

So you get to make the choices. All of the weapons are much useful and have nuances. Maybe you want to go stealth, and there's an opportunity to take somebody out with the crossbow. There are a lot of things that at first you consider to be part of an RPG or an adventure game, but they're just things we borrowed from different genres to enhance the first-person shooter.

These days within the action and RPG spaces it seems like it's all becoming so cross-pollinated that it's increasingly difficult to draw genre lines.

MH: Yeah. I would agree.

The reason I pressed that line of questioning, though, is because when you were demonstrating the game, you kept repeating, "But it's still an action game!"

MH: Yeah. And it's hard, because you want to show the things that are different, but sometimes when you do that in a demo, people just latch onto the things that are different. It's still an action shooter. That's why we make sure to reinforce that. At the same time, we want to show the wasteland, we want to show the vehicles, and we want to show you walking through town because it's a different story delivery mechanic, and we're proud of it. We think it works.

But showing a game is a difficult to task -- how far do you go without losing the focus of, "You know, we still have these crafted experiences you go through."

You described how the world is populated with these varieties of bandit clans, each of whom have their own geographic areas of specific art direction and tone and gameplay. How much did that result from an original top-down design versus emerging out of the exploration of the various group identities?

MH: We did start with basic guidelines for the different areas. To be honest, our top-level design initially wasn't, "Hey, let's make each of these as diverse as possible." You're always trying to do that, but what we started to notice, and a lot of this comes from animation and design, that we would find things we liked and then focus on them.

It really happened later in the project than you would expect. It was engineering, a MacGuyver-ing of different items. We had just a few concepts [at first], and then we'd say, "You know, this one is really cool," and we added this and added that. The next thing you know, we felt really strong about that particular system being something that enhanced the experience in a way that no other game has done.

The different bandit clans started just from the visual style of the game. We were pushing diversity through the visual aspects, and it compelled us on the gameplay side to match that.

When we saw the Ghost [clan], we said, "Those guys look tribal. Maybe they have gritty voices, almost brutal." But they were thin and agile, and the AI programmers introduced to us a system where we could completely the environment with opportunities for the AI to dynamically grab a rafter to flip around, or do a running wall kick off.

Now the AI for that clan can move all around that way. That was from taking advantage of opportunities starting with the visuals. That's what nice about having the time to iterate through the design.

Speaking of influences, I had heard a big influence on Carmack for the environment tech came from satellite photography?

MH: Yeah. He was taking satellite photographs, almost like how Google Earth works now, which wasn't even there at the time. For the idea of streaming in textures, he used that as a proving ground. He got a bunch of different satellite images, and as you moved around, it would stream in what it needed. Now, what with Google Earth, you think, "Well, yeah, I know exactly what that is." But at the time, that didn't exist.

That then inspired the design side for Tim, who set out the initial top level design. He said, "Okay, we need a world that's big, and this is what the tech does. Let's open up the environments. We've never done that before. It will fit with the new property." That was really a back and forth.

Can you bring to mind any particularly interesting or fruitful learning experiences you've had going into an open world? They're increasingly common these days but still ambitious endeavors.

MH: The important thing to note is that's why we always say it's open but directed. We still have a directed path. We decided to have those [open-world] opportunities but to still let the player know where he should be going in the main story arc.

We don't want the player to feel like he's lost. We were really conscious about how we opened the world up to give the player as much as we think he can handle. As the game starts out you're working with local settlers and a couple bandit clans. Maybe you see your first mutant in a very localized area, but if you blow up a bandit blockade, that opens up another section. We try to let the player learn.

We found that, at least in our case, we shouldn't overwhelm the player with, "Here's the whole world." That works in some games, but it's not what our games are. Ours is more of a slow unfolding of the world until eventually you have an entire section of the wasteland unveiled, and then there's another section to be unveiled. We learned that feels better for this type of game.

How much has the design changed over six years?

MH: A ton. Everything. Part of it was technology; part of it was what we wanted to do on the design side. We used to have six little chunks of wasteland with discrete opportunities. Then we went with two giant but more integrated chunks. We've been back and forth, and we moved environments around. We found that with some of the more opportunistic things for the player, like little modular sewers, he can go in and achieve something even without a specifically-crafted job that he has to do.

We keep finding those opportunities. I mentioned the sniper mission [on the mountain]. That's something that we're putting into the game just because the world looks so beautiful up high, and there's a sniper rifle. It just fit, so we've added those things in. It's constantly changing.

Given that level of design fluidity, do specific artists and designers end up getting attached to particular parts of the world or factions or what have you, and then just go all out with it?

MH: We try to take advantage of that. That's classic management, right? If somebody has a lot of passion, they're going to probably do a better job. We're always looking for somebody to say, "Guys, this really fits here," and we'll let him go with it.

Sometimes we'll switch, though. We might take something from a designer who did the initial outlay and give it to another designer who may be able to breathe fresh new life and put new ideas into it. We're constantly doing that kind of thing.

Do you find you frequently have to proactively rein that kind of exploration in?

MH: Yeah, we always have to do that. That's not just so we can ship the game. You could go on forever with, say, ammo types, right? But if they're not meaningful, then you're just bombarding the player with not-meaningful choices, and that's terrible. As a player, I hate that. The same goes for getting lost and not knowing where you're supposed to go, or even being able to find the right kind of information. You can go too far with that. Having less really badass stuff is always better than having a whole lot of mediocre stuff.