English
Gamereactor
articles
Portal 2

Valve on Portal 2 (part 1)

Joshua Weier is the project lead for Portal 2 and we got to have a nice, long chat with him about how one of this year's most anticipated games came to be.

Subscribe to our newsletter here!

* Required field
HQ

Right after finishing Portal 2, I sat down with Valve's Joshua Weier to talk about the game and get details on its development. We also talked about the story, going into the background of Aperture Science, so spoiler sensitive people beware. You have been warned. If you're still here, read on...

Who are you, and what do you do?

My name is Joshua Weier, and I'm the project lead for Portal 2.

And what does that mean?

This is an ad:

Hmm, what doesn't that mean... Well, when we started out, I was doing programming and design with the team, and kinda just organizing the team, and as we grew and grew, doing interviews like this, marketing stuff, management, just kinda whatever needed to be done.

Portal 2

When did Portal 2 development start?

We started a little bit after The Orange Box had come out. We took a little bit of a break, and then we basically started right up.

This is an ad:

When you started, did you have a specific list of things you wanted to accomplish with the game, or?

I think in some ways we started with, I guess a list of both. We knew that we didn't want to just make the game harder. I think a lot of puzzle games do that, where it's like "okay, let's just make everything twice as hard", and, you know, our market will shrink for the people who can actually do that. And we wanted to do quite the opposite, we really felt that Portal 1 had this broad appeal to it, that there was some things in Portal that made it a little difficult. With Steam we can actually kinda see people's progress through the game, just statistically, and so we can kinda watch them go along and hit these cliffs where a certain puzzle was too hard.

And almost in every case that wasn't because of a mental challenge, but because of a physical challenge that someone who didn't play games enough couldn't get their head around enough. We really were kinda sad about that, because the ending of the game was really great, and a lot of people never got there because of that. So between not wanting to make it more difficult, we also wanted to say "hey, let's back off those really difficult maneuvers". But at the same time we wanted to make the game with a broader appeal to everybody. So a lot of the mechanics that you play with, like the aerial fate plates and the excursion funnels, were based around the idea that, you know, the first Portal had this really neat thing where some of it was mental and puzzles, and some of it was flying around the world and doing these really cool acrobatic things. And so a lot of the mechanics feed into letting more people have that really dynamic, interesting interaction without all the twitching. But if you are really good at that, you can use those mechanics to do some interesting things, too. So it was really just a way for us to give more people that really interesting experience that the first game had.

Portal 2

And I think of course that we wanted to make it a full game, so we had to find ways to make the pacing make sense, so that you aren't playing test chambers for ten hours straight with no break in between, until you want to jump out a window because you're tired of it. I think all of those things were kinda where we started from.

What was the biggest single challenge in making the game?

I think one of the hardest things, or at least the thing that I think we were the most nervous about or key on, was to make sure that we didn't ruin what Portal was in making Portal 2. I think, it was interesting to us just how personally fans took Portal 1. You know, people love Half-Life, they love Left 4 Dead, but with Portal 1, with the cake and the companion cube, it was really a different feel for us. So every decision we made, we wanted to make sure that we weren't, you know, somehow just stepping on that or losing that feel, or making it that Big Game that kinda crushes what it was. And that came in a lot of ways. We really liked the mechanics of Portal because of that one simple mechanic that kinda had all these interesting things that came out of it. So that was key to us, we wanted to make sure that kept that. We didn't want to add, like, three different weapons and make all these different buttons and just clutter it up.

And I think from a story standpoint, we wanted to make sure that the humor was the same, but at the same time we wanted to expand it. So with things like Wheatley or Cave Johnson or the defective turrets, all those things kinda gave us new voices for the writers to add comedy. So I think the biggest challenge was to make sure that the tone was still there, that it was the same world.

Portal 2

And we really accomplished that by just playtesting. You know, we bring in people from outside the company who are big Portal fans, and let them play it and make sure they had a big smile at the end of the day and said like, "yes, this is Portal 2!" And if there was anything they didn't like, we'd really dig into it and make sure we understood it and tweaked it, just to make sure we weren't wandering of the path.

Speaking of playtesting, Valve has a reputation for being really, really thorough with their playtesting. How much playtesting do you do with a game like this?

We playtested pretty much from day one. I mean, as soon as we had enough to build a puzzle and put people through, we would sit people down. And it starts within the company, so we grab somebody who is on a different team and sit them down and just let them play it and talk to them about it. Pretty much for the whole course of the game, we playtested every week. So we would work the whole week and then Friday we would test everything we had, talk about it, and then start rolling in all the details, you know, the day after.

And I think that was interesting about Portal 1, because a lot of companies will sort of build a design document and then go and make it, and for us, it would always start from these little, you know, germs of an idea, and from that we would build bunch of stuff. We'd find something new and embrace that, go over there. It's an interesting way to build a game, and I think and important way to build the game, because it let us make sure that everything we did made sense, that we didn't over-commit to anything, and could kinda see where the world went in a way. And I think that was a really fun way to build it, too.

With all the new gameplay mechanics that are in there, how many of them were there from the beginning, and how many were added over time? Did you cut any along the way?

Yeah, there were a few that we cut. One of the criteria that we used for them, like I said, we wanted to keep things simple. So, we wanted it to be about your portals, which you already knew or would learn about, so they all had to work with that. Like the light bridges were an interesting way to do that, because it could project through your portals. When you think about it, it's simple, but when you play the game, they're like, you know, you have to think about it a little, and then... I think, the other important thing about those mechanics is that we wanted them to surprise you, so you have that bridge, and when you turn it this way suddenly it's a wall, and oh, that's an unexpected thing. And it's still simple, like you understand it, but it would just put spins on it.

Portal 2

And building on that, we wanted the mechanics to add into each other, we didn't want this one mechanic that never interacted with the other ones. We felt like that just wasn't very interesting. So like, with the excursion funnels, when the gel drops into them it kinda coalesces into this blob, and it's just interesting, like "oh, I wouldn't have thought that could do that, it makes sense but it's neat". So that was a big, key part of the mechanics, making sure that they all fit together and that we could keep adding them in, and it all made sense and was a good experience for the player. Most of the things that we cut or got rid of were things that didn't play that way, that just felt isolated and not very deep. That was important too, we wanted things to feel like you were learning them and then learning a bit more and then a bit more, so by the end you felt like you had learnt all these really great things and felt smart.

The paint idea came from an outside game, some guys you hired. Did any of the other gameplay stuff get in that way, or was it in-house ideas, or?

Yeah, everything else was in-house ideas, and a lot of them are based around on us sitting down and like saying, well, we really want to give the player a way to move around quickly the way they did when they were flinging, but in a more interesting way, so like the aerial fate plates came from that. And then we'd sit down and say, okay, what are the core things that this lets us do and then make puzzles around each one of them and then build it that way. So that's kinda how we approached everything.

Portal 2

Let's talk story! Portal 2 goes really into the background of Aperture. How much of that was there by the end of Portal? Did you come up with it for this game, or was the universe already created, or?

Some things were there by the end of Portal 1, like Cave Johnson and kinda the history of Aperture was very loosely there, so we knew kinda where it was located, and we had an idea for the timeline, that Cave is the founder of it. But other things were just sort of as we went along, things that needed to support the world. We knew where we wanted to take you, so we would kinda shape it that way. Or we wanted to show you, like, one part of the game is going from the 50's to the 70's to the 80's and showing you that visual progression, so a lot of that was fleshed in as we sat down with the artists and decided what we wanted to show you, and then the writers could support that with their writing. So it was a very organic process, but we had loose guidelines that we knew the fans knew, so we didn't wanna deviate from that.

When I was playing through those abandoned, old parts of Aperture, it sort of reminded me of parts of Half-Life 1, where you go to those backwater, closed down parts of Black Mesa. Is that intentional?

I don't know if it was intentional, I know... I wasn't at the company when, I came on the company for Half-Life 2, but for me, I loved those parts of Half-Life 1. I think, you know, if I was emanating any sort of idea, then I think it came from that, because I did love that from the first Half-Life, this feeling of there's this place, and it has all this character to it and this history, and you're just kinda getting a sense for what it is now. And a lot of people who did work on Half-Life 1 worked on this, so I think there probably was a bit of that going on. It was kinda a fun nod to the first Half-Life I think, but again, it's in the same world, so it's kinda this interesting thing...

Portal 2

And still, there's this kind of disconnect them. Like Black Mesa is mentioned in one of Cave's off-hand remarks, and the only time Aperture's been mentioned in Half-Life is at the end of Episode Two, I think. Will there ever be any more cross-over or links between that will be more elaborate?

I think it's really fun for our fans to connect our worlds that way. Some of them are a little harder to connect, like it would be difficult to connect the Team Fortress world. But for sure I think the nods in EP2 and even in Portal 1, where GLaDOS makes references to the compound and there are the powerpoint slides... So I think that always, when our worlds connect in that way, we try to keep that alive and keep making them more interesting. So I'm sure we'd love to keep doing those things.

Did you drop any characters along the way?

Hmm, did we drop any characters... No, I don't think so. Early on we kinda had different ideas for the personality cores, and so Wheatley was just one of them. And as we went along, we just kinda simplified it and just made him the character. But a lot of them came at the end of the game and made a reappearance there, so we didn't really drop them, we just shuffled them around.

So it was always this bigger cast than Portal 1, but still this really tight little cast?

Yeah, I think that sorta... If you talk to the writers, one of the rules they kinda set down was, when characters were talking, they were talking to you or about you, so when dialogue was going on you'd want to hear it. Rather than two people you overhear talking about something that doesn't necessarily relate to you, so you kind of feel like, oh, here I am. And I think, again, Half-Life 1 had a lot of that sort of thing where, you know, they're talking to you about this thing that's gonna happen. I think that gave it this really immediate sense that's important to the players and make you feel kinda pulled in.

Portal 2

In the first game, you never really saw the output of Aperture as an organisation, other than the portal gun, but I sort of got the feel from Portal 2 that you saw more Aperture products. How much of that is deliberate, and how much is just a function of having to introduce a gameplay mechanic?

I think a lot of it, it was deliberate at first to put it across that way, because that's one of the fun things about Aperture, that they're not just doing science, but doing this kind of crazy, not quite... I don't know if you saw it, but we have this old video where we show this turret that's being used as a babysitter. So there's this weird sort of, you can almost see this corporation doing that, but of course we know that that's just such a bad idea. So I think we have a lot of fun with that, like if you watch the videos by the elevators, they often have this preposterous usage for everything.

And when you go further back, there's a poster advertising the propulsion gel as dietary supplement or something.

I think for us that was always fun, because it was sort of, Aperture was sometimes sorting things by making this incredible technology, and then using it in the most strange ways, or just brute forcing some other thing to try to make it work, so I think that was a really funny way for the writers and artists to play with that world and kinda set it apart.

Portal 2

The co-op part, when did that come about?

After we got most of the single player mechanics, that's really when we started it on earnest, because we wanted all those single player mechanics to also kinda work in co-op. But I think the idea for co-op had been around since we started, and it wasn't necessarily that we came out of Portal 1 ready to do that, or even that fans were asking for it. It was just that, everyone who'd talk to us would say, "hey, I had a lot of fun with this game, and my girlfriend would watch me play it from the couch", or for some people it was their parents or their kids, so everyone was kinda having this cooperative experience without the other controllers.

So we just felt like, man, we just have to capitalize on that, we have to put a controller in that other persons hand, so instead of them just pointing at the screen they can actually be part of it. So I think from the beginning that was sort of our goal, to make sure that we built that experience and people could say, "hey, remember that game we played, now you can play it with me". But I think, what was interesting is that we did because, it wasn't just what we expected, which was okay, two players are gonna play together.

Portal 2

But the amount of communication and how much more difficult we could make the co-op was really fascinating. There's this anecdote I've been using, where we'd watch people playtest, and maybe fifty people would playtest and they'd probably play it from front to back, and a lot of times those players would get stuck. And we'd come and talk to them, and we'd want to find out what they were thinking without of leading them, so we'd say "what are you trying to do?" and "what are you thinking?" And it seemed like nine times out of ten, they would start to tell us, and they'd say "well, I'm gonna do this, and then... oh!" and they'd figure it out just because they started talking and would form their thoughts.

And co-op just naturally does that because you try to tell your buddy what you want them to do, and half the time, just the talking is helping you solve puzzles. Whereas in singleplayer you get frustrated because you just don't have any ideas and then give up. When you have co-op, there's two things - one is like, you have this social pressure not to let your buddy down, so you feel really more invested in it. And the other thing is just having two heads, that really does work, so one person will say "oh, I got this idea", and that little brainstorm let's you progress, and then maybe at the next puzzle the other person has the brainstorm. I think that kind of social dynamic caught us by surprise, and I think was really interesting in moving forward and just worked really well for the game.

How much work went into the ping system?

I think basically the idea for the ping system was from day one, because the minute we sat, I mean basically we sat like this, with one person over there on his computer and the other person here, and we'd say "put a portal over there". And he'd go "where, where", and you'd be pointing on your screen. So we're like "okay, we have to put something that will kinda mark it". So it started as a very simple thing where we just showed a position. But we knew going forward that a lot of our customers wouldn't have necessarily voice communication, so the ping tool had to evolve so you could mark specific locations and tell people what you wanted them to do. But it was pretty much invaluable, being able to say "do this here" just short-circuits so much communication and makes life so much easier.

HQ

Related texts

0
Portal 2Score

Portal 2

REVIEW. Written by Rasmus Lund-Hansen (Gamereactor Denmark)

"Portal 2 managed to surpass my very high expectations of it and it's the perfect sequel to the classic Portal. It does everything right and doesn't take a single wrong step."



Loading next content