Reflecting On Uncharted 2: How They Did It

By Kris Graft

[As Uncharted 2 tops U.S. console retail charts, Gamasutra sits down with Naughty Dog's lead designer on the project, Neil Druckmann, looking back over the sequel's creation, polishing, character honing, plotting, and more.]

Uncharted 2: Among Thieves has become the PlayStation 3's standout exclusive -- vastly outpacing the original in terms of hype, review scores, and positive buzz. The game has, in its first month of release, sold over a million copies.

The team at Naughty Dog has spent a great deal of effort making sure that the game is better than the first in the series, which debuted in 2007 to very positive reviews -- and that the lead character, Nathan Drake, is better-developed.

Here, lead designer and writer Neil Druckmann discusses the process of creating the game, and how the narrative is shaped and mood is created by using different techniques: what makes the game, work, then, as a linear story.

People like Uncharted 2 quite a bit. In general, how does that feel, after two years of working on a game, to have the kind of reception it's garnering right now?

Neil Druckmann: It's fucking great. It's pretty amazing. We were hoping that we would get some good reviews and do better than Uncharted 1, but this has been off the charts for us, no pun intended.

The original one had a good reception also. Can you talk a little bit about the lessons you learned from the development and design of Uncharted that you applied to Uncharted 2? Were there some things that you wanted to remedy from the first one?

ND: Absolutely. When we finished the first one, the first thing we did was have a meeting and say, "We have one game out now and we have a history going. What is Uncharted? What are the things that really define our world and our characters?"

We put a list together, and one of the things was, "Okay, it's got to have these really interesting characters and the pulp action feel. It's got to have this unique combination of traversal gunplay, which is really using our platform mechanics that we have experience on from previous games and combining it with the hardcore shooting mechanics of a third-person shooter."

Also we were keeping a light-hearted tone for the story. It was really important for us. You could say it's a darker story than the first game, but we still wanted to be light-hearted in tone. There's more at stake for Nathan Drake, but he's still quirky. He's still funny. He's still getting into intense situations the way he does. Those are all things that we wanted to keep from Uncharted.

For things we wanted to improve, we didn't have too many moments. We had a lot of good narrative moments set up, so we didn't have these big cinematic set pieces in the first game. We really wanted to push the technology forward that would allow us to do that, so one of the first things we did was create a tech that would allow us to have moving objects that let us keep Drake and all the allies and other NPCs on those moving objects with all of their mechanics and all of their move sets. We really didn't have that in the first game.

That let us do the train level in this game, which is amazing. It's not just a level where the train is moving on a straight line. What a lot of games do is that the train is stationary, and the environment is moving around you, which is why the train can only move straight. We have twists and turns. It will careen around corners.

It will have a collapsing building where, as the building is collapsing, you're still in control of Drake and are still shooting. What other companies have in cutscenes, we wanted to have in the game, to let you play those big cinematic blockbuster moments.

There's an incredible amount of polish that was done to this game. How did you guys approach that -- the idea of polishing this until it's so slick? How much time did you spend on doing that, and what was your philosophy?

ND: The way we work, we like to keep the game stable and playable. As soon as a level goes into the game, even if there's stuff not working, it has to be playable from beginning to end, so that the whole company can play it and give feedback. Because of that, we're constantly iterating on everything. The warzone is one of the first things that went into the game, and until the game shipped, we were still iterating on those sections. That's one thing.

We also gave ourselves a few more weeks for beta this time for polish, because we knew we were going to have more set pieces. We had more stuff in the game, but a little bit more time to polish it, so that was not too much of a gain in comparison to Uncharted 1.

The game director constantly pushed us that nothing could be mediocre. If it's not amazing, it's getting cut. We took a lot of stuff out so that we could focus on stuff that was working better and iterate more on the set pieces that you see in the game.


Did the actual structure of the team stay pretty much the same from Uncharted 1, or did the team grow? Did you apply different practices?

ND: We grew a little bit. The company is still pretty flat. You have different departments, and the leads of those departments, and the game director, creative director, and the art director. Then, above those, you have the co-president.

The thing that's unique about Naughty Dog is that everybody's working on the game. Even the co-presidents are implementing stuff in the game. Christophe Balestra, the co-president, and Evan Wells do a lot of game design. As a lead, even, I'm in charge of levels. I'm doing level design and I'm scripting.

We did get two guys off of the design department that are dedicated scripters and have really freed us up to do a lot more of these scripted cinematic sequences. So that was a change.

We grew some from Uncharted 1 to Uncharted 2. I'm not sure by how many. Maybe by 20 or 30 people. Some of those were contract workers and are not here anymore.

Another thing that changed is that there are no producers. Everybody is working on the game, and the people making the decisions about what needs to be cut and what goes in and out of the game are the people implementing the stuff. We find that really helps us make the best decisions about the development of the title.

As I'm playing this game, it's a very focused game, and a linear game. Game reviewers will sometimes knock a game for being linear. But when a really good linear game like Call of Duty 4 or Uncharted 2 comes out, they love it. Is it necessary to have a very linear, focused game, in order to tell a good story?

ND: I think it would be pretty damn hard to do if it wasn't so linear. I wouldn't say it's impossible, but I think back to all of my favorite story-based games, and they're all very linear in nature. We set out to make a linear story. We're not letting you affect the story in any way. Drake is always going to make the same decisions. What we try to do is mirror those emotions Drake is feeling through gameplay.

[*Mild spoilers*] For example, in one part, Drake is fighting with Chloe about leaving Jeff [Elena's camera man] behind after he was shot. That whole sequence is meant to show the frustration that Chloe is having and the desperation that Drake is feeling, and to give you both sides of the argument so that at the end, it doesn't feel like you betrayed her.

It feels more like, "Well, she kind of has a point, but Drake also has a point. He's trying to save this guy." The gameplay is helping the narrative, not letting you make your own narrative.

A lot of games don't do that. Uncharted tried to express themes through the gameplay. Is that something you tried to do throughout the whole game?

ND: Yeah. We really tried to. It's no coincidence that when you're running off with Jeff, the whole atmosphere of the level changes. It's raining, and your whole gameplay is constrained to make you feel that frustration. We did that throughout the game.

The whole part with the village... there's a reason why we did a slow pace where you get to know the villagers. You see the village in this peaceful state, and you meet a guy who didn't speak any English. It was for a reason, because we really wanted you to start trusting this guy and get to know this guy, so that when things flip, you have an emotional reaction to it because of what you've played.

I got to that village on the mountain, and you can't run during that part, you can only walk around at a slow, deliberate pace. I did manage to jump off the mountain, though. I just wanted to see if you could do that. I was able to jump over the fence and die.

ND: (laughs) Did you shake anybody's hand or pet the yaks?

I did pet the yak. Drake said something snarky to it. You also have your guide, Tenzen, telling Drake to follow, and I realized, "Wait a minute, I can move around and explore this area." I kicked a soccer ball with some kids. Little touches like that, like how you can pet the yak or kick the ball...

ND: To go back to you talking about polish, that's one of the areas that we've had for a while. Initially, you were just following Tenzen. We watched focus testers play the game, and we had punching in there at the time. They were running around and throwing punches, and going up to people and yaks and trying to punch them. We were like, "They're trying to do something interactive. Why don't we do something more contextual and meaningful for this situation?"

So now if you throw a punch and nobody's around, Drake will try to throw the punch but his side is hurting. He was shot, so he caresses that wound. And when he goes up to a guy and tries to punch him, he'll shake his hand or say hello. When he goes up to the kids, he'll play with them. The mechanic is still there, but it's replaced with something more contextual that helps the story.

You guys are constraining the player, and there are a lot of developers who are all about letting the player do whatever they want. Obviously, that's not necessarily what you guys wanted with Uncharted.

ND: That could work in something like GTA, where you're a criminal and a murderer and it's okay for you to run people over and shoot them. But that wouldn't be true to Drake's character. We have to contextually change what your mechanics are to make you feel like that character. You're not playing yourself. You're playing Nathan Drake, and everything you do has to be appropriate to Nathan Drake's character.


Another question I had is about creating a strong lead character. It sounds like a lot of it is keeping him within certain boundaries of what he would and wouldn't do. What else is there to creating that strong personality?

ND: The thing with Drake is that he's a very strong character but a very grounded character. A lot of the stuff he does is what players would do themselves. And even the things he says. He has the same reactions as a lot of our players have. When he sees a building collapse, he's like, "Holy crap, we were just in that!" That's what the player is feeling. That helps us mirror at least some of the emotions that players are feeling.

Also, what makes a character more rounded is contradiction. Here you have a guy who is a criminal and hangs out with criminals, but has a conscience. That's what differentiates him from Chloe or Flynn, to some extent. That contradiction is what makes him interesting and gives him depth.

And surrounding him with interesting characters as satellites brings different characteristics of his personality out. Chloe is there for a reason. Elena is there for a reason. Flynn in there for a reason, and he has less altruistic emotions than Drake, whereas Elena pushes him the other way.

Talking about themes in the game again, I found myself with this weird feeling at one point while playing the game. You start out, and it's this really big quest, and you're trying to find the ultimate treasure. So initially, you're going after this big goal. After the train crash, Drake picks up the key to this treasure, and it dawned on me, "All he was doing is chasing this little trinket, and all this horrible stuff is happening around him." It seemed kind of petty to me.

ND: And he kind of realizes that too. That's why he's ready to quit. Once you meet back up with Elena, she convinces him to keep going. The train wreck part is his lowest point. He's shot, he's lost his friends, and then it's just a trinket -- a material possession. It has to become more to him.

And then the way that he picks it up in the cutscene after the train his body language says, "Well, here's this damn thing that really screwed things up for me right now."

ND: And the goal starts shifting from less altruistic to more so, as you reach the end of the game. Even the big part is about revenge. That changes toward the end of the game.

People have referred to him as a mix between Indiana Jones and Lara Croft. Do you think that's accurate?

ND: I'd like to think he's his own character. Especially in Uncharted 2, he gets out there on his own. It's definitely cool to be compared. We're all fans of Indiana Jones and the Lara Croft games, but by now, we think we've carved our own niche and our own gameplay enough to stand out.

There are the contrasting personalities of Chloe and Elena. Why did you guys decide to bring a second prominent female character into the mix? Was it to bring out a certain side of Drake?

ND: Yeah, we wanted to show a different side of Drake that you didn't necessarily see in the first game. In the first game, he starts out in these special circumstances where he's already out on an adventure and already on the island. You don't get to see him and what he does day-to-day.

With this game, we wanted to say, "What if you see the people he's usually surrounded with, like Elena, Chloe, and these other criminals?" And the first thing you do is go on this heist and rob a museum. It's like, "He's not the best guy. He's not the typical hero." These characters draw out Drake's traits that you didn't see in the first game. And then we bring Elena back into the story to say, "Okay, what is his true character? When shit hits the fan, what will this guy really do?"

When it comes to design, was adding multiplayer to the game a big challenge? It was initially conceived as a single-player game.

ND: It was a huge challenge. But it's something the whole team was really adamant about doing. When we finished the first game, we realized we had this unique traversal gunplay, and everybody came up to us, the lead designers, saying, "How cool would it be to have a multiplayer game like this where I can climb, jump, surprise people, and run and gun?" And at the beginning of Uncharted 2, we were like, "Let's go for it."

We hired designers just to work on the multiplayer part, and had separation so that the single-player wouldn't suffer. We had artists and programmers just for the multiplayer part. Every Monday, Wednesday, and Friday at 2:30, the whole company was playing multiplayer and giving feedback and iterating on stuff. And we ended up with a pretty cool multiplayer component.


What were some of the biggest issues that you found yourselves running into when creating the multiplayer?

ND: In multiplayer especially, the mechanics have to be super-tight. You won't have any patience if Drake is taking half a second longer on an animation to climb up, and if you get shot during that, you get frustrated. So we tightened up all of the controls. We tightened up the shooting mechanic. The camera is zoomed in closer than in Uncharted 1. Guns just feel better. They're more distinct from each other than in Uncharted 1.

It took us a while to realize that you need to know exactly where you are and be able to easily communicate it. So if we had a temple that had two sides of cascading stairs, we cracked the ceiling on one side and had a waterfall coming down in it.

This was so you can be like, "Okay, I'm on the wet side, the dry side, the high side, or the low side," and can easily communicate to your teammates where you are.

Lag turned out to be somewhat of an issue with the melee system. Melee in single-player can last a few seconds when you engage with an enemy, and it couldn't work with multiplayer, so we ended up tweaking the way melee worked for multiplayer.

And then once we had the competitive component, it was so much fun, and we started imagining what it would be like to play it in the single-player scenarios. That's when we added the co-op, and started building up these cooperative scenarios to play with.

Online multiplayer and co-op are something that are kind of expected now, particularly with shooters. Was that something you guys wanted to do because you wanted to see it in the game, and not because you felt pressure like, "Hey, we've got to have this bullet point"?

ND: Yeah, definitely not. If you're thinking that way, you end up with features that are tacked on just so you can get the bullet point. We have to really believe in something and see how it's going to be fun and how it fits in with the game for us to go through with it. Otherwise, we're not going to do it.

How did the writing work on the game?

ND: All of the writing for Uncharted 2 was done in-house. There are three main writers. Amy Hennig is our creative director and head writer. Then there's me and Josh [Scherr], who is the lead cinematic animator. We do a lot of the writing as well.

Recently, Electronic Arts hired a co-writer from the film Monster's Ball for Dante's Inferno. Do you think doing that is misguided, hiring from outside somebody who hasn't done video games?

ND: It doesn't have to be. It can be misguided if they don't understand the process that needs to happen. For us, even if we hired a writer from outside, they have to be in-house. They have to understand the game.

Things are changing and shifting so much, you have to constantly accommodate for it. If you're working with a writer you're only talking with once a month or once a week, I don't see how that's going to work. It will be very difficult to make it work.

The pacing of the game is like a page-turner, or a novel where every day you can't put it down, because you want to see what comes next. It's like you're teasing the gamer, saying, "Hey, just play through this next part and you're going to find out what happens next."

ND: Pacing is actually huge for us. We've had so many discussions about, "Okay, is this battle too long? Is this sequence too long? It's been a while since you've fought somebody. There's too many battles. Let's have a down moment for a little bit." We spend a lot of time, effort, and innovation on pacing.

What was your goal with the visuals in this one?

ND: Just to push the PlayStation to the limit, as far as we could. We re-wrote almost our entire graphics engine... the first game was entirely on the GPU, and in this one, we were able to spread it out.

I'm sure people have seen reports in the past where we said we were using 100 percent of the PlayStation 3, and we're not lying or exaggerating when we say that. What we mean is that none of the SPUs are ever idle. They're working 100 percent of the time. In the first Uncharted, they were idle about 70 percent of the time, because we were just on the GPU.

That let us put all of these systems in parallel and let us get better shadows, ambient occlusion, and much better depth of field effects, which gives us much better camera work than we had in Uncharted 1. We had more compression on our animation. It let us push all of these different things in different directions, and there's still room to optimize, so the future is looking even better.

It's interesting looking at Uncharted 2, and hearing you say that you guys, an in-house Sony company, weren't utilizing the PlayStation 3 closer to its fullest capabilities in the original Uncharted. I know that there are third parties -- you can see in the cross-platform games -- that aren't using the hardware to its fullest extent. I understand that there are technical and cost limitations, but to actually see the console firing on all cylinders is nice.

ND: And we have an in-house team that actually develops tools for other first parties and even for third parties. So hopefully, you're going to start seeing that shift, where more and more games that aren't even Sony first-party games get similar results to what we have, hopefully. I'd like to play more games that look like that.

Return to the full version of this article
Copyright © UBM Tech, All rights reserved