Beyond Heavy Rain: David Cage on Interactive Narrative 2

Heavy Rain was a bigger success than David Cage anticipated. Where next? At GDC this year, the developer showed off a tech demo that may or may not represent the studio's next game. Entitled Kara, it tells the story of a sentient android caught in the process of manufacture -- and demanufacture -- as the operator of the plant realizes she's self-aware. It's gripping and emotional, and it's also very impressive from a technical standpoint.

Heavy Rain may have had plot holes and it certainly was hated as intensely by some as it was loved by others, but it also marked a meaningful step forward for interactive drama -- and was a surprising commercial and critical success.

In this interview, Cage reflects on what it is, precisely, that he wants to do with games. He looks back on what Heavy Rain meant -- both to him, the team at Quantic Dream, and its players -- and looks forward. He discusses both the technology the team is developing and also the creative mission which drives them.

You go right for the emotional punch right away, very strongly. Is that the goal of your studio?

David Cage: Oh, yeah. I want to create many cool experiences for a mature audience. That's my entire thing: I want to make games for a mature audience, and I think you need to go for stories, characters, and emotion. That's what talks to everyone. There are so many games out there where you shoot, or you run, or you jump. The industry doesn't need one more. So, yeah, try to create something emotional.

We saw that with Heavy Rain, no doubt. I remember, with Fahrenheit (Indigo Prophecy), you had put in some supernatural elements, and you said that was something you thought you had to do because of it being a video game; to cater to the audience expectations. You shied away from that with Heavy Rain. How confident are you now to step away from any of the conventions or the expectations of the game audience, now that you've found success with Heavy Rain?

DC: Heavy Rain gave me a lot of confidence in that field, because I realized the audience is comfortable with that. In Heavy Rain, the heroes didn't have a gun. They didn't need to shoot anyone. There was no monster, no supernatural power. And that was fine. It was not an issue for anyone.

I think what really matters is to create characters that the audience can resonate with. As long as you have characters that you like, and when you feel they are part of yourself in there, you get interested in what is happening to them, and then it resonates with you. That's the most important thing. And you don't need to shoot or kill anyone. There are other ways of interacting that are just as interesting.

It's my understanding that Heavy Rain performed better than Sony's expectations, commercially.

DC: I think it performed better than anyone's expectations, including ours, to be honest.

What does that tell you?

DC: [Sigh] Again, it gave me really a lot of confidence in exploring this new direction in showing there's enough room in the industry for different types of experiences. Not everybody has to do the same thing. The audience is more open than we think, in general. And, yes, they want first person shooters. And, yes, they want action/adventure, and they want all the genres, and that's fine.

But I think if you come with something that is really sincere, that you truly believe in, there's a place for you. It was the case on Heavy Rain. That is the case for anyone who is interested in trying something different. I think we are an industry in desperate need of innovation, so we need more new ideas and more silly concepts. Trying new things is very important.

What have you learned as a studio from the success of Heavy Rain and from the way people reacted to it -- good, bad, and indifferent?

DC: We suddenly learned a lot. One thing we discovered that was a bit of a surprise... We were surprised by the success. We were also surprised sometimes by some of the reactions that were sometimes very negative.

There were people becoming very defensive about what Heavy Rain was. Because it was not a game about challenge, it was not a game about shooting, then for some people was not a game at all. As if destroying zombies or killing monsters was the ultimate definition of what video games should be. And I think no, that's just a part of what games can be, but it's certainly not games in general.

Games is a really wide genre where you can do very different things. You can do puzzle things, or you can do Call of Duty, or you can do Heavy Rain. You can do many different things. There should be a place for all. The market wants that to happen, and people want that to happen. But it was surprising to see how aggressive some people can become because they felt that were touching their holy grail. I don't see any reason in that.

What about the positive reactions, though? Were they more than you expected?

DC: Oh, yeah. Oh, yeah. I was really amazed. Recently, there was a survey in the industry about people's favorite game on the PlayStation 3 cycle. And they asked this question to major game designers in the game industry. And Heavy Rain was mentioned very often. That's an honor, just to see talented people really enjoyed this game and thought it would have an impact on their work. So, yeah, it's really an honor.

Something I really liked about Heavy Rain -- and I wrote an editorial about this -- is that because the abstract controls perform every action, whether it's in combat, or tucking in your son, or whatever, using the same control method, everything has equal weight. And I'm very curious if that was your intent with that.

DC: Oh yeah, definitely. I was always more interested in what happens in your mind than what happens with your thumbs, to be honest with you. But having said that, we always felt that having some sense of mimicry in the controls would be something very powerful, because the goal with Heavy Rain was to put you in the shoes of your characters, make you feel what they feel, and make them play their daily lives -- so that when something important happens to them, you're on board, you're with them, you're part of them. And I think the interface played really an important role in this mechanism, to work together.

Why go into games, and not film, if your primary goal is to tell stories?

DC: My primary goal is not exactly to tell stories. My primary goal is to make you feel something. And I think the most exciting thing to do it right now is to do it through interactivity.

So I'm not a frustrated movie director who cannot make movies so he makes games. That's really not how I work. I'm in this industry because this is really what I want to do. I think there are so many things to invent based on interactivity. And you can tell stories. There's nothing wrong. It's like you're doing a sin if you try to tell a story in a game.

I don't like game mechanics as such. I don't think that's the only way to create interactivity. There are other ways. Destroying or killing is not the ultimate definition of interactivity. Changing something in your environment, talking to someone, feeling something, making a meaningful decision -- this is interactivity, too. And it's just as meaningful and interesting. So, again, there are different ways of doing things. We just try to explore different directions.

But you do have a focus on a cinematic style of storytelling.

DC: Yeah.

Why is that?

DC: Because it's the most efficient way, I found, to trigger your emotions. There are other people in the industry triggering emotions not using storytelling, and I'm really impressed with what they do, and how they do it. Somebody like [Ico creator Fumito] Ueda-san, for example, doesn't use story the way I do it. He uses a sense of poetry in a totally different approach. Or the guys doing Limbo. There are many different people doing different things. Or Flower, for example. They created emotions maybe not using storytelling, and that's fine.

But I like storytelling since I was a kid, and I always liked stories. I thought this is a very universal thing. You can tell a story that means something to a Japanese guy, or an American woman, or a German. No matter where you come from or how old you are, you can understand a story and feel something watching Kara and sharing how she feels. That's my thing. It's something I believe in.

Prior to the interview, you said the way you handled Heavy Rain's performance capture, I think you said you retained 50 to 60 percent of the performance.

DC: Yeah, that's my take.

So, why do you feel that? What was lost?

DC: The body language. Because it was not shot at the same time as the voice. The actors just try to mimic on top of the audio. So, that's fine. It works. I mean, we did it, and we're happy with the result. But at the same time, all the subtext is lost. All the things that you say through your body, not your words, are lost. And now, thanks to this performance capture technology, we can get this back. We can get some nuances, some subtlety, in the performance, in the subtext, in the body language, that we never saw in a video game before. Now I have the confidence that we can get that.

With your new engine that you discussed, was it primarily constructed for performance capture, or is that only one facet of what you tried to achieve with that technology?

DC: There was one thing that I forgot to mention, that this demo is actually one year old. We did that a year ago. So, what you saw is the engine where it was 12 months ago. So, it's now 50 percent of the features we currently have. It's the first thing we shot in performance capture. We improved the technology. We improved the rendering. Everything looks so much better now, a year later.

But what it brings... We wanted to develop our own technology, rather than using existing technology from the shelf, just because we have very specific needs in the types of games we make, especially regarding virtual actors. We want very realistic tears, skin -- skin shaders, in general.

We have a specific approach to lighting. Lighting is so important when you start working in this area, with very close shots of the characters. You want perfect shadows and stuff. And you want to use the light to tell something. It's not just, "Oh, I need a light." You want the light to tell you something about the character and the moment and the emotion. So, we developed specific tools to create specific lighting for close shots, to use performance capture, to use facial animation, all these things.

We keep on improving again and again. Most of the engines that are on the shelves right now are focused on huge environments where you can shoot and do different things. We really focused on the cinematography of the engine, how we can create real cameras, having a weight, and having real lenses, and all this stuff.

Yeah, that was going to be my next question, honestly. Cinematography is an integral part of how movies tell stories, but cinematography tools in games haven't necessarily been as advanced. So it sounds like that's what you're really focusing on.

DC: Yeah. We learned from cinema for certain rules, but at the same time, we don't make cinema. When you play Heavy Rain, you don't watch Heavy Rain. You play Heavy Rain. You're in control pretty much the entire time. What it means is you can be inspired by the rules and the grammar of cinema, but you need to invent a different one, taking interactivity into account. So, you really need to invent a new grammar visually, and in narrative storytelling. And you cannot just totally copy what movies do, because we don't do movies.

You must have done a postmortem on Heavy Rain at the end of the project and really identified things where you felt you succeeded and failed. I was wondering if you'd share some of the things that you feel strongest about.

DC: [Big sigh] Well, there were different things. With the team, we were pretty much unhappy with everything. We thought we could have done a better job in all areas, and have better rendering, and better visuals, and better gameplay, and better everything.

So, yeah, this is definitely things we took into account designing the new technology, wanting to work with performance capture. And we wanted a better blend of storytelling and interactivity. We thought that sometimes in Heavy Rain there were moments where the balance wasn't exactly right. We are working on new ways of merging this in a more natural and fluid way.

There were so many things, after Heavy Rain, that we learned. We felt on a marketing point of view, I would say that some people maybe didn't give Heavy Rain a chance, just because they felt it wasn't a game for them. Maybe some people thought it was just a game where you would just press buttons, and some kind of interactive movie thing, which Heavy Rain absolutely was not.

We felt we lost some people because we couldn't convince them to give the game a chance. So, this is also something we took into account in trying to convince more people that they should give the game a chance.

On the flipside, what were you maybe the proudest of in terms of your accomplishments with Heavy Rain?

DC: I'm really proud of Heavy Rain. Not in an arrogant way, like, "Look at how good we are"; we're just proud of having made it. I mean, to meet people every day, telling us about their experience playing Heavy Rain. And many come to me and tell me about the scene where they need to cut the finger, and how the wife was sitting on the couch with them and saying "Do it" or "Don't do it" or whatever, and how it generated conversations and dialogues within the couple. Many people played with their wives.

I'm really proud of the fact that for the people who really enjoyed the experience, it seems to be a part of the culture now. It's something that they really lived, not just a game they played and they closed the box and that's it, "Forget about it." It's something that they keep talking about and that really left an imprint in their mind. Yeah. I'm incredibly proud of that.

Do you see yourself as an auteur?

DC: It really depends on what you call an auteur, because it has some positives and negatives to it. If you mean do I consider myself doing art? Honestly, certainly not. I don't think I'm doing art. I'm just doing it by passion, and I'm doing what I believe in.

It's more about crafting something, and building something all together for two or three years with a team. That's really what we do. And if something of what we create today, people still talk about it 50 years from now, then we'll say, "Okay, it was art." But that's really not something I have in mind every morning. Honestly, I don't care.

Now, I think I'm an auteur in that sense that I spend a year writing this stuff. It's one year of my life doing this from morning to night, non-stop, for a year. And I put a lot of myself. I'm not talking about me -- I'm talking about what I feel, what I think. Heavy Rain was really about me becoming a father, and all the fears that go with it. Yeah, all the fear and all the promise and all the things... In that sense, yeah, I think I'm an auteur, in a way.

Now, I know that a lot of time, obviously, during production of games, things change. And a lot of developers like to be really reactive to that and go with what works. But when you're working with a heavily written piece, it may not have that flexibility. So, can you talk about that process?

DC: My games are really written. I spend a year writing them, and very little can change. So, what I try to do when I write is identify areas where I know there is space for changes. Usually it's about interface. It's about gameplay. It's these kind of things that, yeah, you cannot plan it on paper, but you know that you'll need to do it on-screen and see if it works. And if it doesn't, you need to have a contingency plan.

But regarding the story itself, the production pipeline is so heavy -- because you need to build all these assets and shoot all these animations -- that you cannot really change your mind in the middle. So, I try to identify these places for potential changes, but if I've done a wrong decision regarding the story, there's usually little you can do along the line, just because the process is so heavy that you cannot really change.

But there are many things that changed during development. All of the 3D interface in Heavy Rain, for example. It was an idea that came in the middle of development. We were going for 2D interfaces in the lower corner.

Similar to Fahrenheit.

DC: Yeah. We said, "Well, wait a minute. We need to find..." We had this idea in Fahrenheit, of having them in 3D in the environment, but we couldn't find a way to make it. I just wanted to try again. And we tried again, and we found new solutions. We decided to implement it.

So, these are the kind of things that can happen during development, and you need to leave space for new ideas during development. Otherwise you are just implementing, and you lose track of what you try to achieve. When you know there is space for discoveries, and this is still a living being in a way. Your project becomes like a living being. You need to be careful that it breathes, that it's alive all the time.

So much is said about films as a narrative medium coming together in the editing room. If you watch making-of documentaries -- I'm particularly thinking of The Shining, and seeing Kubrick at a table typing up pages during filming. Those are two examples of where films are not set so far in advance. Have you given that any thought in terms of flexibility? I understand pipelines and production processes make these things difficult.

DC: I mean, the pipeline is so heavy that you cannot leave too much for improvisation as you go on, because you need to be focused one thing at a time. That's the way I like to work. When I'm on stage with the actors, I want to focus on their performance, not on my script, because otherwise you lose focus and you don't know what you're doing.

The games, at least the way we do them, they allow that, because you need a script, then you shoot, but you don't care about the cameras, just performance.

Then you can really focus on the cameras and change them as much as you want because it's real-time. And about editing and the rhythm. And then about gameplay. And you can all do this in sequence, rather than having to think about everything at the same time like they have to do sometimes in films.

But once they shoot, they shoot. They cannot change the camera afterwards. We can. So, I'd like to take this liberty and use it as much as I can.

When you work with the actors on stage, do you let them ad-lib? Do you have a back and forth with them?

DC: Oh yeah, yeah, yeah. It's really a collaboration work. The more it goes, the more we work with very talented actors. And they come with their views on the script. They read the script. They have their take on the story. They want to propose things about their characters, how he moves, how he shoots it, how he says that. That's the most exciting part in what we're doing. The thing I would hate the most is just doing actors who are machines, and just do whatever we tell them to do. That would be really boring.

Games usually can't have as much of that. It's more like little tweaks here and there. Obviously in film, you can let people go wild, if you so choose. We can think of examples of great filmmaking where people have entire ad-libbed scenes. The famous example is Orson Welles in The Third Man; he just improvised a whole scene. Games, usually, it's not like that at all.

DC: No, you want your lines to be short, to be effective. You want things to move quite fast. The difficulty we have is that I would love to have long dialogues, but actually people, when they have a controller in their hands, they don't want to listen to long and boring dialogue. They want things to go fast and they want to interact.

So, a part of our challenge is to have interactivity in the heart of the story. You play the story all the time. You don't watch it. When you have dialogue, you play the dialogues. There are stakes in these dialogues. You want to make sure that you're going to say what you want to say. You won't miss anything. All of this has to be a part of interactivity. It's not just about telling a story. It's about telling an interactive story. That makes a big difference.

I interviewed Ninja Theory's creative director, Tameem Antoniades. He was talking about for Enslaved, he and Alex Garland realized that the ending of the game they had shot didn't work in the story. As you say, you can't do reshoots. They did new voice recording just for that bit. They rewrote the script, and then cobbled together new performances out of pieces that they had. Have you faced a situation like that? Do you think that's a feasible sort of way to do things?

DC: Well, yeah. It's definitely feasible. I try to spend more time on the writing to avoid being in this situation. It's not a guarantee. You can still believe in what you're doing, spend a lot of time, read to other people, and still have something bad in the end -- but I don't like to change my plans because everything is really tied up.

And it's not just one story in our case. It's several stories with different variations, and different endings, and different stuff. So, if you really want to change something at the heart of the story, it suddenly becomes a huge change, because it changes all the options potentially. So we try to avoid this situation as much as possible.

Speaking of tying things up and having multiple endings, Heavy Rain was an incredibly complicated narrative with four main characters, a lot of contingent things that happened, complicated relationships. Why choose something so complex? Did it work?

DC: Well, when I did it, I didn't think it would be that complex, and I realized it [while] writing it. I think it worked for many people, from what I heard. Is it required? No. You can do much simpler stories and move people just the same way.

But I think my goal in Heavy Rain was to kill death, in a way -- was to kill Game Over. So, I started thinking, "How can I get rid of Game Over situations?" Because in the context of storytelling, having a part where you die and you need to play again -- the same bit of story, the scene -- doesn't make any sense. It becomes really boring.

So, I really had to find solutions to say, "How can I get rid of Game Over situations?" and the first answer was to say, "Hey, why not have four different characters, so if you lose one, two, three, or four, it just alters the story one way or another, but the story carries on." This is just what I tried to achieve with Heavy Rain. And yeah, the complexity came out of this, but I think people enjoyed it.

It's interesting to hear you say that, because that does imply that it was a design-led decision, initially -- a game design-led decision, rather than a narrative decision.

DC: But I don't make differences between those. Sometimes people ask me, "Do you design the story first or the game design first?" The right answer is you need to decide both at the same time. Otherwise it doesn't work. Otherwise you have this great story, and you try to mock some game design concepts on top of it, and it doesn't exactly match. Or the opposite way, you have these great mechanics and you try to put a story on top.

You need to think about everything at the same time so there is a kind of consistency, where the story supports the gameplay, and the gameplay supports the story. Both need to work together. So, yeah, it's a story-game design decision.

And the same thing for whatever you're working on now, I'm assuming?

DC: Yeah. Yeah. That's a way of thinking, with game design.