Jim Munroe interviews Marc Laidlaw

Game Designers' Conference in San Francisco, March 10, 2005
Transcribed by Phuong Nguyen

Munroe: I'm so glad that you've been getting so much deserved kudos for what you've been doing, and I've told you on many occasions that I think it's fantastic, but... I've been having a crisis of faith as to whether narrative is desirable in video games at all. I mean as much as you have this terrific praise at the moment, have you had moments or phases where you sort of question the use of narrative in video games?

Laidlaw: I don't question what I'm doing. There are so many different kinds of games. I think it'd certainly be wrong to say that most games need it, or even most [ games need it]. The kind of thing that's called narrative is the game that unwinds sort of like a story, and an experience that has a culmulative effect that feels like a narrative to them but really isn't offered that way. There are games that do that, I certainly don't require it of most of the games that I play. Plenty of those games that have sort of story element to them are really badly done--just cheesy or tacked on. But as long it doesn't interfere with how fun the game itself is, I usually just kind of chuckle and enjoy it. I've been playing Resident Evil 4, and you know, it has these characters and they've got their whole mythology, and since because it's the first time I've played through, I'm not really familiar with all of them. But it comes to a scene and I just kind of laugh, because it's so over-the-top. It's translated from the Japanese and not especially well, and the performances are kind of wooden, but it really does not interfere with my enjoyment of the game. There's a whole category of especially Japanese games that's like watching Italian Hercules movies. That's part of the fun, I think.

There are some people who are doing great translations, and it's totally seamless, and you don't really notice. But those aren't necessarily even a story, it's just good interface--good character one-liners and things like that. It's interesting to see so many people take different approaches to narrative, the Half-Life style is only one. It's a narrow set of rules that's been really productive for us--it's great to have this really hard thing we have to do all the time since we get really creative within these limitations. But it's not right for most other kinds of games. We saw this right after Half-Life came out, there was a low-level experimenting to try to do this kind of stuff in the same way, and then it stopped fairly quickly. People realized how hard it is, and unless you're designing your entire game around it, it's kind of unrewarding.

Munroe: The cut scenes in Half-Life 2 aren't conventional—the storyline is progressed but your perspective stays the same and it feels very integrated with the rest of the game.

Laidlaw: We [originally] thought we were not going to do cut scenes [conventionally]. At some point, we were going to do it, or at least turn the camera on the character, and then we realized no, we don't have to do it. It's not like we had a rigid philosophy, it's just a combination of what we wanted to try to do and what we were able to do ending up in a place where we felt pretty good with our limitations. We try to maximize the whole game that way. And you know, we've gotten some criticism that the game is too linear, but in the service of wanting to make sure everybody has the best possible experience they can have. There's still room to push on that. To add more things we haven't figured out how to do yet.

Munroe: In terms of the criticism, I'm actually kind of getting sick of the praise of the freeform sandbox model in comparison to the linear rail model: the rail model bad, sandbox good kind of dichotomy.

Laidlaw: It's pointless. Like you see here [at the Game Developers Conference], there's so many people with different approaches to games, and the market for games and the kind of games different people want to play. It's hugely varied, it's really hard to say, you'd never say one kind of game is right and one kind of players' love for certain kind of game is better than another's. I think it's an exciting feeling because it's so varied—some people are interested in doing some kind of purely textural game play and others are into a totally non-verbal experience. I kind of like when I first saw the totally non-verbal Quake kind of thing I thought there's no reason you couldn't do this sans story, and that was to me a really interesting thing to pursue--but at the same time, that's just totally out of my background. I could see enough of how the tools work, and I knew enough about writing a story that I could see how those went together, but anybody with a slightly different background is going to see another potential there. I don't know, I was just lucky to run into the set of people who were interested in doing this thing pretty much the same way that I was. And flexible enough to experiment to find out what worked and what didn't.

Munroe: I'm amazed by the economy of it, in the sense that when I first played the game, it gives you the impression of these endless places and ways that you can go. Like when you walk by stencil graffiti on the wall you don't always stop to check it out, but it really adds to the texture of City 17. But what I found was that, I played it through the beginning part twice, and I did stop and check out those things but I actually had got pretty much the full experience the first time. There wasn't all these touches that I had missed. It just gave the appearance of that, of a richness.

Laidlaw: I mean that's definitely something that we desired to create--that feeling of the world being bigger than you could possibly experience. Some of the scenes it's definitely true, depending on where you look, you're always going to miss something else going on. That's to give some more replay value and also because that's also realistic. I can't see everybody in the room at the same time and what they're doing, but if I could rewind the scene right now, and look at someone else, I should be able to watch anybody's face in the room and see their expression change. The scene unfolds the same way each time, the non-regularity is in the player's experience, where they put the camera and what they choose to do in that space.

Munroe: Frankly, it cheered me up to see it wasn't endlessly detailed. I don't know if you know Chris Ware, the cartoonist. I actually found his work enormously depressing at first, because all I could think of was this poor little geek working alone every Saturday night doing nothing but detailing all these exquisite and painful pieces. When I found out, well, you know, he actually does have a wife, a home life, etc., it made me feel better about it. I had the same sort of feeling where I found the limit to the Half-Life 2 world. It's about versimilitude, the appearance of reality and freedom, not the actuality of these things.

Laidlaw: Also, there's diminishing returns [on endless detail]. We can do branching scenes in a game, but that ends up being work someone is going to have to do--that's going to dilute or take away from something else. The animators do this all the time, everyone does this up or down the line: what is the most important scene and what do you want from it? Who is the most important character in this part of the scene, and what is the most important gesture that you are going to give them? Then the emphasis will eventually go to those identifying those things as really important. I work really closely with the animators to make sure I'm not writing scenes that are going to be boring to animate. That there's always a point to them, and my work is not really done when I do my draft of how I see a scene, because that will always fire some sparks in Bill Fletcher's head for something he wants to do with the characters. So he'll want the character to step back, think about something, and then be like, "I've got it!" So okay, I'll write that in the scene and work it through like that. To see that character smacking her head and realizing something--that could be the whole reason for the scene. In the same way we set about designing an ambush with some monsters, we're going to design a scene where we want a specific emotional impact.

For instance, the scene where you first get to Eli's lab, we wanted you to feel like you were watching a family dynamic with this daughter and stepmother kind of energy going on--we want to you to discount what some of the players are saying about the others. Like Alyx doesn't like [Dr. Judith] Mossman, 'cause there's a stepmother dynamic, so any suspicion she might have of Judith's treachery you discount. So, this is how we design these things. It's the challenge of the animation to bring those out because no character is going to sit down and reveal these things but their gestures... that's another thing. I'm not a big fan of too much dialogue--it needs to be just enough. But we tend to overwrite and record a lot of extra stuff that we don't use, and then it's kind of like scaffolding. Because as soon as you have communicated this to the animators, they're able to express a lot of it non-verbally, as soon as it's expressed non-verbally, we have the confidence to cut down the scene down further and just communicate more visually. And that's always a goal, it's a visual medium.

But I'm happy working in a medium that is so visually oriented because it gets me away from the thing that I'm always doing. That's been a good thing for me in Half-Life 2, getting a better handle on animation and acting. It's turned me onto acting. We basically created radio plays that worked in a vacuum--once they were entertaining on their own right and were dramatic, then we could start to add stuff to them, and get them in the game. It was like, OK, now they are worth animating, because it's already a strong scene with no visuals, so let's bring the next level into that.

Munroe: Yeah, starting with audio makes a lot of sense, really. Especially since you give people the freedom to look away from the characters who are speaking, if the audio is communicative they won't necessarily miss something big.

Laidlaw: Yeah, we wanted it to sink in even if you were across the room. What we found is that we have the full range of players from people who wanted to really know all the characters and the scenes and who position themselves so that they could be the ideal director, and then we have the player who's all "Where's my gun? Where's my gun?" run and jump and knock things and throw stuff at the characters. You'd think that it would be the norm with first person shooters but it's not, it's actually a small percentage of people that's like that. After play-testing we asked both of those kinds of players not just how did they get through the level, but what kind of stuff they did they pick up about the characters. Even the people who appeared to pay no attention to them still picked up on fairly subtle dynamics. Maybe that's how they go through life, absorbing stuff while they're running around the room. And if you slowed down, and read headlines, and stopped and watched the monitors, and listened to every word that's spoken, you'd get more--but if you don't do that, you'll never get lost.

Munroe: So can you sort of break down the work flow of the day?

Laidlaw: Different days are different, but the process... a bunch of people will get together, programmers, level designers, animators, and artists and talk about an area that we want to build. One thing we like to do is come up with a character that suits an area and design the area so that it kind of explores the character. [Also it could be] I know an actor with a great voice, if they're interested in doing this part we could create a part around them, and we could create a part in the game for this character. And we can design exactly what we're going to do with this character in the game. We'll start to get a little more detail about the progression of the story, and what's the background of the characters. We talk about character profiles, motivations, try to flesh out the psychology of the characters, and then eventually, I'll do a script and some of the scenes and get together with animators and talk about it. At that point, we all start to talk about the things they really want to do with their character, in terms of animation, gestures, things that express your character and then I'll go back and rewrite the script that will try to bring out some of that stuff. So, eventually we will go into the voice studio and do a voice session with the actor. And we'll get a bunch of extra stuff, they're never just reading a linear script, they do a lot of alternates: let's try this line; you're doing this line really close up; now you're twenty feet away; you're angry; you're scared. We'll take that stuff and pretty much take it back to the lab, and these are our pieces for building the scene. And then in the process of that, we'll usually find little weird bits and pieces in the outtakes and the alternates that will inspire one of the animators to be like, "This is totally not obvious, but I got a picture and you could totally do something with this."

Munroe: Are you thinking about something in particular there?

Laidlaw: Yeah: in Eli's lab, you've just met Eli and Alyx comes in and he ribs her a bit, he's kind of teasing you, and he goes "awwwwyyyyiii!” Well that's just the sound [voice actor] Robert [Guillaume] made. When Bill Fletcher and I were going through the audio stuff, we just heard this sound, and we were like "oh, we gotta use that sound." Because Bill instantly saw something to do with it, and so he took it away and fed it into the scene, it wasn't supposed to be there, but as soon as we heard it, it had to be there--it was just such as interesting sound.

Munroe: Would it be possible for you personally to do kind of a literary videogame?

Laidlaw: I'd have to find some angle in it that would interest me as a writer. A lot of stuff works in games because it hasn't been done before in a game, you know, it's been done to death in every other medium. One of the things Half-Life is really good at is recycling cliches, and kind of standing them on their head, whatever, putting them in a blender. In the first game it was the cliché of the transdimensional teleporter, this one has the cliche of the Orwellian future. We're always on the lookout for that, you know what are the science fiction cliches of this genre, they're good because everybody recognizes them and you don't have to explain them before you turn them on their head.

Munroe: it's not a satirical thing, even...

Laidlaw: Right, it's saying everybody knows what this is, we're all aware of this, we've seen this done to death. But you've never done this before and we're going to make you feel like you're doing it: we're finally getting to play the action hero in one of these science fiction movies. I think we've gone a little beyond that. We're not satisfied with recreating something that would have been in a movie, we want to find the stuff that would never be in a movie. That sounds like the Katamari Damacy designer this morning, he was saying "We want to do something that you can only do in a video game." I think that's really important for me, and what attracted me to the industry--that you're pioneering, in a sense.

Tags: audio , Chris Ware , conventions and festivals , design , designers , Half-Life , Half-Life 2 , Hercules , interviews , Italy , Japan , Marc Laidlaw , narrative , Quake , rail , realism , Resident Evil , San Francisco , sandbox , translation , versimilitude