She has her little psychotic break about two-thirds of the way through, and then you get that part of the game. At four hours you could support that story structure where GLaDOS talks and talks and talks. Or it's just not working, or the pacing needs to get changed. We probably did write 13,000 lines of dialogue. We write a crap load of dialogue and some of it just doesn't end up in. You get instantaneous feedback and for people that are still wrapping their heads around portals, you see the laser is going in here and it's coming out here. Lasers do the exact same thing, except they're much clearer. Remember the pulse beams from Portal 1? Changing those into lasers was such a win, because they're a lot easier to use, you can see them, they're always on, and you don't have a frustrating wait for the ball to travel across the screen. Sometimes things just need a little nudging, or it's things that annoy us. We start play testing really early on, so we can see what fails or doesn't fail. Sometimes we'll take an idea and end up shipping it somewhere else. We thought about what we could do story wise, and we came up with a whole bunch of ideas and tried some stuff out.Д«ut a lot of stuff gets melded into things we do ship, too. Once we had all these new puzzle elements that combined in these interesting ways, using the exact same design philosophy of, let's show you the neat things this can do, it ended up being about two-and-a-half times as long as Portal 1, just naturally, with no padding. There was a period of exploration where it was like, well what other puzzle elements, what things can work with the portals but not replace the central mechanic? We came up with a list of stuff that seemed to be really fun and then combinatorially interesting once they were all working together.ДЄt no point when we were working on Portal were we like, we're going to make a four hour game, so how do we get to four hours? It was, we're going to show you all the neat things you can do with this Portal gun, and we're going to try not repeat ourselves too much, and let's see how long that ends up being. We didn't want to add a new gun that, you have to switch weapons now, or add a bullet time button or something.

We liked the elegance of, you have a portal gun that shoots two portals. It's not rocket science, it's Aperture Science.

We knew we didn't want to change the central mechanic. Initially we were thinking about what puzzle elements could we add to the game that would expand the puzzle space and give us some new things to do with the portals. Because Portal was so short, people still wanted to do more with the Portal gun. We felt Aperture Science was a rich environment to go back to. Part of it is so many people love not this GLaDOS or not this Chell, but the Aperture Science world. We sat down and we were like, 'What can we do that will make this interesting?' We weren't banking on Portal 2 being the game that saves the company.

Portal was such a beautiful surprise, almost perfect in a self-contained way. How did Portal 2 come to be? What's changed? How does it all tie into Half-Life? And what's next? Read on to find out.

Here the childhood friends talk in-depth about the creation of their latest offering. Now, six-and-a-half years later, the writing duo have an impressive credits list under their belt: Half-Life 2: Episode 1, 2, The Orange Box, Left 4 Dead, Left 4 Dead 2 and, come April, Portal 2. When Gabe Newell asked Chet Faliszek and Erik Wolpaw of Old Man Murray fame to come work at Valve, they of course answered the call - the developer was about to ship Half-Life 2.

Portal 2 episode 1