Gabe Newell

When the head of the dev studio behind the finest FPS experience ever speaks, we listen

Gabe Newell, founder and managing director of Valve Software, sits atop the hill as one of videogaming's most popular and successful developers, and one that's responsible for what we regard as the finest first-person shooter series ever - Half-Life. At the recent Games Convention in Leipzig, we crept up on Gabe and bombarded him with a barrage of questions covering Half-Life 2, Steam, next-generation consoles and beyond. Read on to find out what he had to say...

When stacking Episode One, Episode Two and Three up against each other, do each have a different gameplay style? In Episode Two, for example, you have the open environments...

Gabe Newell: They are bringing new things forward on the technology side, the game design side and on the story side. Episode One was very much about Alyx - about having this companion in the world, and how she made the gameplay better for you. And she could be a proxy for emotional states, like she would say: "You should be scared now". It gave us a very useful tool on the story-telling side.

Episode Two is different: we're using that similar approach with the Vortigaunts this time, instead of Alyx. We didn't want to do Episode One again, because you can always go and play Episode One again.

And the G-Man plays a bigger role in Episode Two, doesn't he?

Gabe Newell: Yes. The overall progression of the three episodes is that the G-Man is losing control of you. In Half-Life he made you; in Half-Life 2, he used you for his own purposes, which are still mysterious to you; and now he's starting to lose control. At the same point that he's losing control, you've come to the attention of other forces - the Combine hierarchy is now saying, "We thought that Earth was under our thumb, and this one person is being such a pain in the ass for us. So we're going to have to pay more attention to this person." So, although you have more freedom from the G-Man, other forces will come to bear on you.

So you know what is going to happen in episode three?

Gabe Newell: Yes, we know what happens after this episode; there's a fairly large chronology. Things tend to vary more in the specific details of the minor characters, so as we're working through the issues in an episode, we discover stuff about the characters regarding what is and isn't working. For example, Dog is getting much more of a role than we had originally envisioned for him.

To what extent are you reacting to people saying whether they liked this vehicle and so on?

Gabe Newell: We have this overall plan where we sit down and say: "Where can we get the most bang for our buck - here are the technologies we need to develop within that road-map." Then we order it based on feasibility and risk. You want to have some things that are risky and some things that are predictable in each of the chunks. Then we map that against how the story needs to move forward. We still have lots of flexibility in the production - if, say, we need more exploration, more story-telling and less action. One of the things we did was the re-use of areas in Episode One: some people liked it and others didn't. So we've gone back and re-edited Episode One to reduce that.

What can you tell us about Portal?

Gabe Newell: Portal is sort of at that beginning stage, where it's showing a lot of potential and we're working through all the technology issues, such as: "Oh, how do sounds work through portals and so on?" We're used to using DSPs as a method of determining sound attenuation, but that doesn't work when you're putting holes in everything. And we're looking at gaming issues: we want to make sure we can teach people how to use these things - it's such a change from how people are used to thinking of the world working. You know, a cliff is no longer a barrier - it's kinetic energy that you haven't used yet. Is it something that people can pick up on? It's great for us to get that out there earlier and learn as much as possible from gamers about how it's going to be received.

  1 2 3 4