So, uh -- this Portal thing is kind of a big deal, I hear. I guess that explains the huge, huge goddamn crowd outside the Portal post-mortem at the final day of the Game Developers Conference, everyone giddy like kids on the first day of summer vacation.
Portal project lead Kim Swift and writer Erik Wolpaw were on hand to dig through the development of one of the year's greatest games. It's not unexpected that there was a lot of hero worship in the room, and high expectations as to new insights into the development of the title. Hit the jump for the run-down.
Holy god, talk about packed -- the line for this goddamn lecture stretched outside the hall, out to the main portion of Moscone West and snaking from one end of the hall to the other. Kim Swift and Erik Wolpaw were on hand to talk about the development of Portal, and it was the hottest ticket on GDC's final day. Thankfully, your favorite Linde knows how to queue. Or hold still. I'm good at both.
Swift and Wolpaw kicked off the talk discussing the logistics of Portal's development. The team was small, and comprised of less than ten at any given point in development. While the game has been extremely successful, Swift and Wolpaw acknowledged that it was unexpected given the game's unorthodox nature, particularly in its emphasis of narrative as a driving force in gameplay.
"The story wouldn't make much of a novel," Swift said "And the gameplay on its own would be dry."
Thus, the gang developed a theory. Games tell two stories: "story" story and "gameplay" story. The closer that you put them together, the more satisfying the game will be. Wolpaw cited Clive Barker's Undying as an example of a game whose story managed to undermine the gameplay.
The team established some fundamental design goals. Story should never intrude on gameplay; less is more; trim narrative as ruthlessly as possible. "Portal was going to be a lean storytelling machine even if it killed us," Wolpaw said. In terms of practical application, Swift and Wolpaw insisted that playtesting was the most important thing they ever did in developing Portal -- sitting down and watching people play the game. Afterwards, the developers would ask the playtesters to read back the story, beat by beat. If they missed anything, they cut more of the story, which they insisted made it stronger every time.
One example was the opening chamber, the glass prison in which Chell is confined at the beginning of the game. Originally it was a dark, dank room and featured a force-field prison instead of glass walls. But players had no idea why they couldn't simply leave, got frustrated, and weren't sure what to do. Adopting the glass walls actually informed artistic direction, leading to the anaesthetic, clinical chambers we've come to love in Portal.
The use of the environment to convey story information came about as a solution to another story-related issue: getting simple exposition across to the player. "We had ideas for embedding email, and using email to talk about implementing email, until we realized that reading email was fundamentally not much fun," Wolpaw said.
The answer was creative reflection of story in the surrounding environment, including the wall scribblings in the behind-the-scenes areas throughout Portal's chambers. Swift wanted to use an absent character, "the Rat Man", to communicate these ideas and instill a sense of dread and panic; the creeping realization that something is very, very wrong. So the use of the wall-scribblings was implemented to get those bite-sized bits of story across in an unobtrusive way.
Swift said of the Rat Man's creation, "If I was trapped inside an institution run by an insidious AI, what would I do to pass the time?"
The developers continued on to discuss the infamous Weighted Companion Cube's genesis and incorporation into the level as a simple solution to the problem that playtesters simply wouldn't carry the cube through a marathon cube puzzle, which required that the player haul it with them from start to finish. Finding a way to remind the players what to do and attach them to the object was crucial. Curiously enough, Wolpaw remarked that the idea struck him after reading declassified government manuals on interrogation techniques.
"Isolation leads interrogation subjects to start becoming attached to inanimate objects. Something clicked -- you're sort of alone for this whole time, and maybe if GlaDOS needles you long enough, something will happen."
The climactic battle with GlaDOS, they noted, had gone through several revisions before finally being settled as what made it into the final game. "The biggest problem was the ending battle," Swift said. "What in the world is a boss battle going to look like in a weird environmental puzzle game? A boss monster for a puzzle game was a really, really devious puzzle."
The Portal team's first attempt was an action-based encounter with the infamous AI, with James Bond-esque lasers that had to be redirected via portals at the boss. A second revision pushed the action even further, taking place in the large turret encounter room near the end of the game, and focusing on rocket redirection for combat. But the high intensity of the combat overwhelmed GlaDOS's dialogue.
Another attempt included a chase sequence, but playtesters tended to walk around confused instead of moving in the direction the developers wanted them to go. "This failed in every way possible," Wolpaw said. "Bad pacing, players didn't know what was going on, every solution we came up with was either too expensive to do or made the gameplay worse. What we learned from this was that the more complex the boss battle was, the slower paced the action."
But playtesters were immensely satisfied with the fire pit, which gave important clues to developers: it was easy. It's the easiest puzzle in the game, the developers say, but the feedback was such that it was the most climactic and dramatic moment that playtesters had yet experienced. So what made it climactic? "It had time pressure," said Swift. "It makes people think that something is a lot more complicated than it actually is."
The team adapted these lessons to the final revision of the battle, and created a finale based upon a simple, time-constrained puzzle that was relatively easy -- the shipped ending. Simplifying the puzzle, adding that time pressure and drama, keeping the focus on the narrative and providing an easy solution allowed players to get absorbed in the moment, they said.
Even the incorporation of the year's runaway culturebomb "Still Alive" was created due to the lack of resources. Without the time and money to develop a massive ending loaded down with special effects, they wanted simply to leave the player with a sense of happiness. And what better way to do that than with a song? Y'know, especially that song.
The developers noted several times throughout the engagement that constraints played a huge role in the development of Portal. Though they weren't exactly broke, the limited amount of resources and the shot-in-the-dark development process actually helped them devise new, creative solutions to otherwise complex problems and create one of the finest games we've seen all year. In closing, Swift and Wolpaw suggested to the audience that developers should embrace restraints as fuel for creativity.
Necessity being the mother of invention, we need to take more developers of games and shaft them of money/resources? I think not, but props to them for doing well in their situation. Can't just throw money at a problem (lol madden)
Fascinating. I just love reading how developers eventually came to make the decisions they made. More please!
Good interview. Interesting to see what the thought processes were in what was probably the most original game of last year.
BRILLIANT!
Those crazy kids certainly seem to know what they are doing, but if anything kills a project it's bumping the team size up and having communication fall apart (Bungie said this was the main problem with Halo2). I hope they keep the development team small for Portal 2 or get some help from Valve managing such a sudden influx of new teams.
Nice article, im glad the devs had the sence and acumen to do as they did, their attention the the game/story balance really shines through. The song at the end is sheer serendipidy. The cherry on the cake. :)
Very good interview. It's nice to hear a developer say that their original ideas sometimes sucked. Usually developers let you think they've created the best game ever. OK, we know better but still.
Exellent Read.
"Story should never intrude on gameplay" if only they had learned that when they were making Half Life 2, so we didn't have to wait for 5 minutes in closed rooms/locations, waiting for some story dialogue to end.
The idea of moving story and gameplay more towards each other is nice, but most games with the best stories actually have the opposite. This kind of thinking is what leads Hollywood to ruin storylines in action movies and horrors, please don't do it to gaming too.
Finally, if you get attached to an inanimate object after 45 minutes of playing a game, something is seriously wrong with you. The companiion cube is not Wilson from Cast Away, even if the internet seems to have some really lonely souls that fell for it.
Good read though, nice to see how they could have fucked it up more :P
FTA: In closing, Swift and Wolpaw suggested to the audience that developers should embrace restraints as fuel for creativity.
Finally! Someone that gets it. Less is more. Less forces you to make more with less. Increasing budgets and resources doesn't make things better, it makes them more expensive. Older games aren't great because they're overly complex, they're great because they're simple and were made by using a lot of creativity to expand what little they had.
Nintendo were on the right track, but then they lost me when they stopped making games I like.
FTA: In closing, Swift and Wolpaw suggested to the audience that developers should embrace restraints as fuel for creativity.
Finally! Someone that gets it. Less is more. Less forces you to make more with less. Increasing budgets and resources doesn't make things better, it makes them more expensive. Older games aren't great because they're overly complex, they're great because they're simple and were made by using a lot of creativity to expand what little they had.
Nintendo were on the right track, but then they lost me when they stopped making games I like.
Whenever I hear a developer talk about alternate versions of boss fights that didn't work out, I wish they would include them on the disc as unlockable extras. It would be the same idea as alternate endings on a DVD. Sure they might not be the best way to end the game, but the fans would love to see this stuff.
Very, very interesting read. I love reading about how pieces of entertainment get changed over the course of their development, be it games, TV shows, films, or books. The creative process fascinates me.
I would have loved to see the version of the GLaDOS fight with redirecting lasers. Sounds pretty cool to me.
And while I love the combination of story and gameplay in Portal, and hope to see more games in this style, I hope it never becomes the "standard". I still want games like Zelda, Kingdom Hearts, Final Fantasy, etc. where the gameplay and story are two separate things.
"Good read though, nice to see how they could have fucked it up more :P"
*PUNCHES YOU IN THE DICK*
The Rat Man is a lie.
simplicity, genius, and well...just a really great game. thanks!
its a decent game but not as awesome as people are making it out to be.
great interview though.
You have a stupid opinion.
@generaldane Go die in a fire!
It is of my proffessional opinion, generaldane, that you may be afflicted with pre-portal poor taste syndrome. You may wish to see your local qualified physician, or just take everyone else's word for it.
It's not an interview. It was a lecture. With like 900 other people. :D
Constraints, I suppose, are also a great way to enforce the idea that gaming can be done on a budget. With development budgets skyrocketing out of control in the post-Halo cataclysm, I suppose it's good to be reminded that we can do this without becoming the movie industry.
I love how that team put Portal together. They really did something brilliant with it and I can't wait to see Portal 2. I'm sure they'll stay on the same brilliant development path.
Although brief, I think that Portal had one of the best told stories of any game last year. It was hilarious and non-intrusive and the vehicles used to convey the story were exceedingly clever. I still remember how excited I was when I found the first "Rat Man's" nest.
Professor Pew, I agree. Not only does the companion cube show up at the ending, but it's just a weighted storage cube with hearts attached to it. It was actually harder knowing you had to incinerate the curiosity core, it reminded me of GIR. RIP curiosity core. <_<'