[Naughty Dog calls Uncharted their 'biggest and most complex' challenge yet, and shares successes and stumbles in this fascinating postmortem, originally printed in Game Developer magazine.] 

At the end of 2004, some of the developers at Naughty Dog (Jak and Daxter, Crash Bandicoot), began work on the studio's first-ever game for the PlayStation 3.

Cryptically assigned the working title Big, the project would prove to be Naughty Dog's biggest and most complex game yet. This postmortem discusses some of the things that we Dogs struggled with, some of our successes and failures, and what we plan to do differently next time.

What Went Right

1. Strong up-front design -- that we can ditch when we need to.

As we began to build our new PS3 technology, we also started to formulate a design and direction for what would eventually become Uncharted: Drake's Fortune. Once we'd agreed on the game's general scope, we defined the tone -- a light, cheerful, humorous one, similar to that of our pulp adventure inspirations -- and what the core mechanics for the game would be. However, one of our strengths at Naughty Dog is prototyping early and knowing when it's time to deviate from our design documents.

A great example of this is our aiming mechanic. We initially thought that Uncharted would use an automatic lock-on aiming mechanic for its gunplay. We spent several months trying different lock-on target selection mechanics, control schemes, and camera combinations; we even implemented several types of mini-games for executing special kills while using the lock-on.

Despite all of the planning and production time that we'd invested in the mechanic, the gameplay never gelled for us and we eventually decided to try a completely different approach: free, manual aiming.

Even with the free aiming in a rough form, the game instantly became more visceral and fun. Although this new direction meant months of new work, we'd made our decision just early enough to be practical -- we had enough time to integrate and polish the new mechanic, and it was definitely the right choice to make for our game.

We use this kind of process a lot at Naughty Dog -- having a well-defined idea on paper is great, but the best way to firm up a game design is by trial and error, and you have to know when an idea just isn't working and have the courage to ditch in it favor of a better plan.

2. Focused gameplay.

Except for a few vehicle-based sections, Uncharted's gameplay is tightly focused on a few core mechanics. This was quite a difference from the design approach of the Jak and Daxter series, where much of the fun was derived from the sheer variety of gameplay in the missions.

This focused approach, along with the realistic world we created for Uncharted, made game design on the project quite challenging. We couldn't just come up with some wacky idea and give it its own separate mission in order to make a section of the game more interesting.

This resulted in what we feel is a much more elegant design overall. We were forced to think about the game as a whole, and to make sure that our core mechanics were truly exceptional.

Additionally, because we created things more systemically -- like the player's mechanics and the AI-polishing the game in the final months of production became a little easier. For example, whenever we refined an aspect of the player's move set, almost every level became more fun because they almost all relied on that core move set.

The actors did a blocking walk-through before hitting the mocap stage.

3. Great performances through subtlety.

We're happy to say that we've received a lot of praise for Uncharted's characters, and the performances we captured and created for them.

However, we think that the success of the story, the tone of the game, and the characters comes from a combination of elements that can be united under a banner of subtlety.

Because this was our first experience using motion capture, we had an open mind about how to do things, and our approach ended up owing more to the way a movie or theater production is staged than the way most people in the games industry run their mocap shoots.

For example, we chose to use the same actors for both the physical and vocal performances in our cut scenes, which is unusual for games.

We did a lot of table-reads and blocking walk-throughs with the actors, so that the scenes were well thought-out when the time came to do the shoot.

This also let our actors improvise with the motion-capture director and our game director, and many of our favorite moments in the game are a result of our actors' creativity.

Our choice to use the same actors to mocap both voice and physical performances paid off. We believed that on the PlayStation 3, the characters were going to be of high enough fidelity that the player would notice the nuances that a trained actor would bring to a mocapped performance.