[Gamasutra presents here the unabridged version of the Portal 2 interview with Valve project manager Erik Johnson, which originally appeared in the October 2010 issue of Game Developer Magazine.]
Portal was one of those rare game experiences that took audiences by surprise. The simplicity of the interface -- shooting portals onto various surfaces to traverse through levels -- was striking. Layered onto that was the excellent story, mostly driven by the player’s love-hate relationship with GLaDOS, the touchy artificial intelligence that managed the facility in which the game took place.
Portal was a mere three hours in duration, and many felt it was a perfectly defined experience. How does one take that and extrapolate it into a sequel?
Valve absorbed many of the team members for the DigiPen student game Tag: The Power of Paint, using their ideas of painting different effects onto the game world in order to increase scope without over increasing complexity.
But could they really add complexity of design without detracting from the simplicity that made the original game unique? Project manager Erik Johnson tells us it’s all about understanding what will be meaningful to players.
There seems to be a lot of added complexity to the design in general. What was the thinking behind that?
Erik Johnson: There are two parts to that. One is that in the trailer, the one we showed at Gamescom and the one at E3, implies a lot more complexity than exists for a player that plays through the game. We are showing you towards the end of being trained on a particular element.
We thought Portal 1 was the right length for the number of things in the game, the pacing was good. We want to have that same kind of pacing in Portal 2, so the game ends up being a lot longer, but still brings people up on new things at the right speed.
That was the sense that I was getting. It was looking like a longer game in general, because more is happening there. And the personality spheres seem like they have the potential to put more story in there than before. Is that their function?
EJ: Yeah, definitely. Portal 1 was strictly your relationship with GLaDOS. It was just you and GLaDOS. She starts off as just a voice, and is the tormentor of your existence. The personality spheres are definitely this new character in the game -- different than a character in Half-Life 2 that is all about expression, and how they look -- more or less a story-delivery type of character. The reaction people have had to the personality spheres has been a lot stronger than we've expected, so that's good.
I presume that when those personality spheres existed in the final boss battle of Portal, there wasn't an intention of making them part of Portal 2, right?
One thing Valve doees well is taking elements from previous games and fleshing them out in a sequel or related product. It always feels well-integrated -- how do you go back and retrofit in a way that makes sense in the universe?
EJ: First, the reason why we do it -- especially in Portal 1, there are a bunch of fans of the product already that have very strong opinions about the kind of game it should be. In general, we try to be the servants of their opinion. We try to build the kind of game that they want, and one of the ways to engage with customers is to point out knowledge that they have, like "Hey, we know that you guys played Portal 1," to give them something to hold on to.
And you definitely want to approach all pieces of the game, from story, the way it looks, to how the characters act, in terms of this coherent universe that you can fit things into. A good application of that thinking would make it so that if somebody came up with a given idea, it would be really easy to determine if it would or wouldn't fit. I think Portal has done well in that regard because it is relatively simple in terms of that. And a lot of it is long-term community development, to do those sorts of things.
Portal was a game that many people felt didn't really need a sequel, despite how much fans loved the universe and would appreciate a return. How do you balance that expectation?
EJ: The first thing we did was judge fans by their reaction, and partly because Portal 1 was a much smaller game, they were saying they wanted more -- but what you are saying is really accurate, too.
For many people, it was this perfect experience. It was the game that, far and away, more people finished than any game we've made -- we can see in Steam if the game gets finished, and it was huge in that respect.
We looked back to find the core things players liked about Portal. We felt it was the story and the tone, the type of story it was, and the delivery mechanism of the story. We felt like for a lot of people, their reaction was surprise about the gameplay. Portals are obviously a huge part of the gameplay itself, but the thing they wanted was to be surprised -- like the type of game, what they thought when they played the game.
And music was the third thing. Still Alive is something that people have a really strong connection to. We looked at those three things and said, "Okay, inside of the Portal world, what can we do on those axes to create something that people want.