John Carmack: What Went Wrong With "Rage"

I used to be known about the catch line, you know, it'll be done when it's done. You know, when we'll doom ship, when it's done. When we'll quakeship, when it's done. And it felt good saying that in terms of that was sort of the being rebellious about, we don't have any publisher that's going to force us to be out in time for their quarterly earnings. We're going to make sure we ship the game when it's actually done. But the aspects of seeing with a little bit more perspective now, it's like if you're talking slipping a quarter, slipping six months, yeah, great. That's definitely fine. But when you're talking about slipping years, you know, when years go by, the world changes around you in a way that being a kind of totalitarian about it will only ship when it's done. I largely recant from that now where with a little bit more perspective, time has a physicality that you may not appreciate. And I have the two big reads on that. I am seeing some of like Virgin Galactic, they're never going to make that money back. They're looking into satellite launch for things now. But even the last big game that I worked on it, which was Rage, we spent six years on that game. And we went into that. It was using flashy new technology, which there's some other life lessons about that. But we had like we had an E3 where we were game of show at E3. But we kept on, it didn't quite ship. And by the time it got out, the world had changed around us, you know, the technology decisions that were made for some earlier systems weren't necessarily the right thing for the very latest ones. We now had Call of Duty and Battlefield coming out as these jogger knots that we were competing with. And I look back as one of those real decisions, I think we should have done whatever it would have taken to ship that two years earlier, be less ambitious with some of the technologies and get it out earlier. And I can even make reasonable cases for going back to the earliest games like Quake, where Quake was the first really traumatic game to ship internally, where we're still only talking like two year developments. But at the time, it felt really long. And we had all sorts of internal strife for things, because we were trying to do so many things. It was, you know, six degree of freedom rendering, modding, I am internet based game servers, I, you know, three six off models. And it was a lot of stuff. And I later looked back and said, you know, we could have done half of those things in a super Doom and shipped it earlier and then done the other half even better on a game coming in later. And I still roll that over in my mind sometimes where I love Quake and I love Doom, I think all of those were Doom, I think was the optimal game to ship at the optimal time. Quake was challenging and painful enough that maybe we could have done some things slightly better there.