ver since our first interview with Gabe Newell at the 2006 GDC, we’ve been trying to get details about Team Fortress 2. Originally announced and shown at E3 back in 1999, many people thought that the game was dead and would never see the light of day. After countless delays Team Fortress 2 seemingly fell off the map. That was until last Summer when Valve showed up at an EA Gamers' Day to debut the first ever trailer for Team Fortress 2. Many, many things have changed with the game since it's original announcement, and from the looks of it the team is going back to the core of what makes multiplayer shooters fun.

In part one of our Team Fortress feature we find out exactly what happened to the game from Robin Walker, co-creator of the original Team Fortress and current designer and engineer at Valve, Charlie Brown, Engineer and Project Lead at Valve, and Doug Lombardi Marketing Director at Valve.

Game Informer: Last time when I interviewed you around the time of Episode One, you were one of the main people behind Episode One. Now are you working on Episode Two as well, or are you just focused on TF2?

Robin Walker: I’m 100% on TF2.

GI: Is it good going back to your roots?

Walker: Oh, it’s really fun actually. I think we were making a joke one day when I was writing some code that I was like, “You know, I wrote this function ten years ago on TFC!” (laughs) Anyway, it’s been a lot of fun. It’s been a lot of fun working with Charlie and a bunch of other people who have some interesting fresh takes on TF.

Robin Walker

GI: Also when I interviewed you last time, I asked you about Team Fortress and you were really vague about everything. Now looking back on everything, that’s really funny…sort of. (laughs)

Walker: That was the point where you thought it was this realistic military game.

GI: Absolutely! (laughs) Now, how long has this version of TF2 been in development?

Charlie Brown: Eighteen months

Walker: About a year and a half.

GI: Originally, when was Team Fortress 2 supposed to ship?

Doug Lombardi: Well it morphed a couple times, because you had TF. TFC was originally going to be Half-Life: Team Fortress. It was going to be sold as an add-on pack in the summer of ’99. Then the decision was made to do Team Fortress Classic instead and give it away for free to Half-Life owners and then re-release Half-Life pas Half-Life Game of the Year Edition in summer of '99, which was the first time someone released a Game of the Year edition. We included it with Half-Life going forward instead of selling it as an add-on pack. So then Opposing Force was added in as the add-on pack strategy around that same time–the spring of ‘99 and it shipped in the winter.

So summer of ‘99 we showed TF2 after we shipped TFC. I want to say we called it a next year release. I honestly don’t remember saying a date.

The original Team Fortress 2

Walker: If you want to do some real spelunking, you can go to Quake World TF and it tells you when TF2 was supposed to ship before we even came to Valve. We had a fully playable version of TF2 that was the military version on Quake 2 that we play-tested before we came here.

GI: So when it was officially announced and shown at E3 ’99, it was heavily modern war based way before the whole World War II thing became a major facet of gaming. You even said back then it would be more of a first-person strategy game. Was the main reason to switch to this new style because the whole World War II thing is basically oversaturated?

Walker: Actually, not at all. The real reason is that we didn’t stop for a lot of that time. We were building things that were known as TF2 internally. We ended up building probably three to four different games. We didn’t like many of them. The real changes, for example, we were doing the commander mode stuff that we did in the first version, and it just wasn’t fun. It was one of those great things where in theory it sounds fantastic but when it comes down to individual experiences of players like–the guy being told what to do by the commander, or the commander himself, we just never found something we felt was that fun.

We like to look at what we have now as sort of this skimming of three or four games that we call TF2. Like the individual pieces of them that work best. The medic as he stands now, we had is medi-gun done in a way that worked done in a previous version, and when we play-tested it that was the one that they said, “This is really fun, but this bit is not so fun.” So we got the skimming of the cream off each of these different previous versions and brought them forward. It wasn’t a reaction of sort of external, or what anyone else was doing but it was entirely a reaction of our internal play-testing. Our core production philosophy is about lots of play-testing and reacting to it. It doesn’t matter how theoretically great we think something we come up with is. If it doesn’t play well, then it just has to go. That was sort of the reality of all of the earlier stuff.

Lombardi: The military theme also I think detracted it from the uniqueness of the different roles.

Walker: In the real world, if you need to kill someone you shoot them with a bullet, you don’t get something that has some travel time, or anything because it’s always going to be worse. And bullets arguably are the least interesting weapon you can have because it gives your opponent no time to dodge, no time to react in any way. You can’t anticipate anything. From your perspective, all you have to do is put this dot on a guy and press fire you’re going to hit him. You don’t have to lead him, you don’t have to look at how far away he is versus how fast he’s moving. All the interesting stuff you want to do with weapons, the real world doesn’t have any of them because the real world isn’t about interesting decisions—it’s about killing people real efficiently. In the realism, we felt like we were constantly straight-jacketed into, “Here’s a real interesting weapon idea. Oh. That makes no damn sense whatsoever.”

Brown: What’s a real-world equivalent? Ahhhh, there is none. But it’s still really fun to play.

Walker: The medi-gun for example. The medi-gun is this thing that it turned a couple of our best FPS players into these guys who all they want to do now is play medics. These are guys who are fantastic Counter-Strike players, and so on, and all they want to do is medic-up people. They run around healing the crap out of people and being the guys who win games. They connect up with one other really good FPS player. A top medic with a top Heavy is an incredibly nasty combination. So we have this great weapon that we’re really happy with, but trying to figure out how we can put that into a realistic—there’s no way in heck.

I think that the incredible stylization for a bunch of different reasons, like feeding the gameplay, and so on. Also it was just a reaction to every time we tried to do something interesting in the real world it was just broke down. You couldn’t find a metaphor for it that worked, that made sense.