QuakeCon 2007: John Carmack Talks Rage, id Tech 5 And More

Overview

Date: Aug 03, 2007
Original URL: http://www.gameinformer.com/News/Story/200708/N07.0803.1731.12214.htm
Synopsis: Game Informer interviews John Carmack at QuakeCon 2007

QuakeCon's big event is Friday night - John Carmack's keynote speech. We had the opportunity to catch up with things since our CES interview and get the important details about id Tech 5. Will it work with DX10? Wii? What is Rage? What's going on with Doom 4? What the heck is Quake Zero? Carmack spills the beans on the next Quake Arena project and a whole lot more in this extensive QuakeCon interview!

Questions

Billy Berghammer: When we interviewed you at CES this year, and QuakeCon two years ago you weren't too thrilled with developing for multicore systems. Obviously now, that's the case across all platforms. Were you kind of at the point with id Tech 5 where, you said, "We give in?"

John Carmack: You have to take advantage of what's on the table. Although it's interesting that almost all of the PS3 launch titles hardly used any Cells at all. We hired one of the best PS3 guys around who did the Edge Acceleration technology for Sony - he's on our team now so we've got some of the best PS3 experience here. In fact when we were doing all of the tech demos, we'd bring in the developers and they'd walk over and say, "it's running on the PS3!" (laughs) They'd sit there and stare at it for a while.

There's no doubt that with all of the platforms that we have running here PS3 is the most challenging to develop on. That's what I've been saying from the beginning. It's not that it was a boneheaded decision because they're a lot closer the fact that they can run like this [points to the 4 different gaming stations running Rage] - they're a lot closer than they've ever been before. It's a hell of a lot better than PS2 versus Xbox. But given the choice, we'd rather develop on the Xbox 360. The PS3 still does have in theory more power that could be extracted but it's not smart. We don't feel it's smart to head down that rat hole. In fact, the biggest thing we worry about right now is memory. Microsoft extracts 32 megs for their system stuff and Sony takes 96. That's a big deal because the PS3 is already partitioned memory where the 360 is 512 megs of unified and on the PS3 is 256 of video, 256 of memory minus 96 for their system...stuff. Stuff is not the first thing that came to my mind there. (laughs)

The PS3 is not the favorite platform but it's going to run the game just as good. To some degree there's going to be some lowest common denominator effect because we're going to be testing these every day on all of the platforms, and it's going to be "Dammit it's out of memory on the PS3 again, go crunch some things down" That's probably going to be the sore spot for all of this but because we're continuous builds on all of these we're going to be fighting these battles as we go rather than build these things out and go, "Oh my God we're so far away from running on there." Which is the situation where Enemy Territory is suffering with at a degree right now, and a lot of other people have that.

Billy Berghammer: Will this engine support any DX10 features?

John Carmack: No, not currently. We're not expecting to. We're not sure if we're going to be a Vista title or not. There will be some support benefits by being Vista only. It depends when we get the game done what the adoption has been. But it's a OpenGL title on the PC and Mac right now, obviously D3D on the 360, and the PS3 it's kind of an in between where it's Open GLES but we do a lot of direct command buffer writing there. If necessary we can move the PC version over to DX10, but there's not much strong pull for us to do that. All of the toolset is in OpenGL, I wouldn't want to convert everything over.

Billy Berghammer: You didn't seem to hot on DX10 or Vista at CES.

John Carmack: Microsoft has done a great job with all this stuff. I mean, I honestly think that DX9 with how it's implemented on the 360 is a clearer and more open API than OpenGL is. It doesn't hide the state.

That's sort of the Microsoft way. They start off with a piece of crap, and then over a number of versions taking a lot of people with them over the painful route they eventually get to something that's better than what they are competing against. It's a valid strategic direction. I think they've come out at the end with a good platform and a good product. Some of the DX10 stuff I don't think there's going to be huge draws for the features there, but a lot of what they've done with the structuring of the API I think are still positive things to do there. I think they have a good team with solid engineering there.

Billy Berghammer: You've got all the platforms except one. Could someone theoretically use a light version of id Tech 5 to develop a Wii title?

John Carmack: We could port the Megatexture stuff over, we could port most of the engine over there, but you wouldn't be able to use the same set of original content. There's a bigger gap there. We don't have any major intention to port this entire technology platform to the Wii. What I am kind of hoping for is if Orcs and Elves is big on the DS I want to do Orcs and Elves for the Wii. You know wave the wand in the air.(laughs) My whole reverse engineering for the platform. I still have high hopes for that. Start on mobile, go to the DS, go to the Wii. Then maybe wind up on the 360, PC.

I'm thrilled that Nintendo has had great success with the Wii because while Nintendo has never been my favorite company - relationships between id and Nintendo - we're not a good match in general. We match better with Microsoft with how they position the 360. But one of my tenants that's near and dear to my heart is the significant improvements that are going to be done are with IO devices and the Wii I think is a really good demonstration of that. They've taken something that's much less powerful but it has an innovative IO device. But the fact that they did something different it added a lot of value.

Billy Berghammer: Do you play [Wii] a lot?

John Carmack: Not a lot. I've got a Wii. My almost three-year-old boy, I got that out so we could wave things around there. I play a little bit but I don't play a huge amount of video games in general. No time.

Billy Berghammer: So why did you choose a Mac show to first debut id Tech 5 for the first time, and then show it all off at QuakeCon. I know you showed it to a few people at E3, but not the masses.

John Carmack: The Mac thing was just this weird happenstance. Macs have been sort of infiltrating our office. They've been sprouting up a lot and people are getting Macbooks and they're generally pretty well regarded. Robert Duffy who's the programming lead has four Mac's or something. He was like, "I'm going to port Rage over to the Mac." Okay, that's a good thing. We learn something every time we move over to a new platform. It was always a good idea there. As it was going over we had a meeting with Apple and they said, "What about doing some of this for Steve's keynote." I've had ups and downs with the whole relationship with Apple and Steve, but we thought it was probably a decent thing to do because it was going to be the first crunch point for the team. It's a good team building exercise. The time where you get everyone there working at 2am trying to get things done. In the large scheme Mac Worldwide Developers Conference isn't the hugely important thing to do. It was a good point to see what we could do, because we knew at that time we were already aiming to be showing at E3. So then we were like, "Can we limit features a bit and show something a month before E3?" We figured it was a reasonable thing to do and in hindsight it was a worthwhile technology path to take.

The E3 stuff was all developer only. We still hadn't announced Rage. I think we were actually waiting to see if we were trademark cleared at E3. Trademarking game names is a far bigger deal than most people give it credit. It's amazingly frustrating. You can have a page of names and then have them come back and say none of those are okay.

The old story was that Quake II which has nothing to do with Quake I was going to be a completely different name, but we got fed up with getting the names rejected so we said, "Dammit, it's Quake II. We own that trademark."

Billy Berghammer: What can you say about Rage? When everyone saw the shakey-cam footage people were wondering if you were making a racing game....

John Carmack: It's about 50/50 racing and first person action. It's interesting. If we go back post Doom 3, we started making a completely different game. It's internal name was The Darkness, which it's obvious that someone else has taken since then (laughs) but it was going to be another spooky dark survival horror thing set on an island. We spent a lot of time going on that. But we reached this point where, "Do we really need to do another dark, spooky game?" id always gets slammed for the game is always too dark. Maybe we should do another game that's a little brighter. We can branch out a bit more. Maybe it would be more fun to run over people with pick up trucks and ATVs. We rebooted our entire development project. We were reasonably far into production on The Darkness title and we just said we wanted to go in a completely different direction. We wound up with this post-apocalyptic Road Warrior type thing.

I think there's going to be some neat stuff in it. We've got the whole outdoor wasteland - big areas, going between lots of different areas. We're doing some of the sandbox play there. I like racing games. Mario Kart is the last game I played reasonably. That type of thing is fun.

Then you've got the run and gun internal shooter stuff, id's always done well, and we're going to do a good job on that. Then you have the sort of RPG-ish elements, pimping out your ride, getting money to buy accessories and building it up. It is a different style of game which is risky especially when you're talking a $20 million budget, and the safe this for us to do would be to run right into Doom 4. But we made the conscious decision that we want to broaden id a little bit. We've got Wolfenstein, Doom, and Quake. We're trying to bring Orcs and Elves up from the low end, and we're going to try to bring Rage in from the high end to broaden our whole portfolio here. There's no doubt it's a gamble. It's a big deal. There's an element of stress there that we don't have all of the advantages that we previously had going into a title, there. But we think it's going to be pretty cool.

There will be a Doom 4, we don't have it scheduled or a team assigned to it, but there will be a Doom 4. There's going to be a Quake Arena sequel. There's a Wolfenstein thing in production. We're following along with all those. This game doesn't have to be Doom. It's going to be something different.

Billy Berghammer: Now that you're creating something totally different, do you want to make another Quake or Doom game in house, or are you interested in letting your friends - Raven, Splash Damage - work on those games and then you move forward with the engine situation, and creating new IP.

John Carmack: The big changes at id is that we are going to be starting to staff up for a second team.internally. This was mostly driven by publishers where the publishers would come to us and say, "We want to do a Quake Arena sequel, or Doom 4. We'll be happy to work with other teams like the normal id relationship." It's important to realize with a pure technology license with Prey, we kind of tossed the stuff over and "Good luck, we wish you the best, don't bug us too much." With Enemy Territory we have half our company working on that project right now, even though technically it's Splash Damage. There's always a lot of id involvement on an id titled game.

So with the publishers, if we do it in house, they said they'll give us a bunch more money. This came from several publishers on there. Really significant amounts of delta's there. So we are beginning to staff up internally. We've got a plan to do this.

We have a kernel team and we've seeded it with people from our main team and moved them over, and the plan is right now to do a bridge project that we're calling Quake Zero. This is experimental and this is one of my pet ideas. We're taking the Quake Arena code base data sets and repackaging it for incremental web downloads and make it a free game and we're going to try a sponsored advertising supported free version of the game. I want to try this. Where the website is the hub for everything, most of the menus are gutted out of the game. The game is something that launches instantly. You set up everything on the web page, all of your community stuff there. You get sponsors and advertisers on there to see if we can support this. We have a unique opportunity for this because we own all the stuff on there - it's still a good game. We can do this with the kernel of the first six developers on here.

The goal is here to make - after Quake Zero - then we'll probably take all of the lessons we learn from that interaction and make an id Tech 5 Quake Arena sequel with all the bells and whistles. All of the modern stuff on there. That kind of boot strapping process will be the best path. I expect we'll wind up with 40 something employees. We're still tiny as far as the big studios go. I want to keep us under fifty.

Billy Berghammer: I haven't seen you this excited since you talked about your portable mobile project. What's it like to say again, "we're totally doing something different - here it is!"

John Carmack: It's an exciting time right now at the company. It really is. We've got a great team here. It was a great week last week. We've been going through the features, and we made the video and asked one of the artist, "How's it going?" and he goes, "it's going awesome!" And we have that feeling across a lot of the company where we see how the product is going to turn out and we think it's going to be something that's going to go out there and make a dent and have some impact for us.

There's a lot of trials and tribulations across the development process. There's going to be highs and lows across all of it. I think we're accomplishing what we set out to do. That's really the things that drive me. Here's the problem, here's the goal, here's the tools at our disposal, lets craft something that's going to get us there, then evaluate how we're doing as were making our way there. I think right now we're doing a pretty good job. There's a lot more left to do and a lot more challenges. I think we'll continue to sit up at the top of the important and relevant companies that are doing the technology for gaming and then the games themselves.