Interview: John Carmack and Tom Mustaine on Doom, iPhone Desires, and the Future of id Mobile

If John Carmack is id Software's mad scientist, then Doom: Resurrection is his latest experiment. Developed largely by id/Ritual veteran Tom Mustaine and his team at Escalation Studios, Resurrection is a rail shooter take on Doom 3, priced at the premium rate of $10.

With the launch of the game today, I sat down with Carmack and Mustaine to talk about the risky project. Also covered are Carmack's criticism of the iPhone's current shortcomings, his take on how the ZeniMax acquisition will affect his time to develop pet projects, and what's coming next from id Mobile.

Tom Mustaine: Did you get a chance to install and play the game at all?

Shack: I did play it a little bit, yeah.

Tom Mustaine: Cool, so you can see that we've kind of taken a different approach to traditional shooter, Doom stuff. How far did you get?

Shack: I want to say about 40 minutes? I played mostly on a subway--not the best tilt-control environment.

Tom Mustaine: [laughs]

Shack: The controls are interesting though. It's a clever scheme. How much work went into figuring that out?

Tom Mustaine: It's interesting. We actually started developing the game about eight-nine months ago. We knew we were going to go for a crafted, pull-you-through-the-world experience, and that's what best fit the devices.

Carmack on the 3GS: The combination of seeing people download 700mb files of Myst, and the new capabilities--I could do some mind-blowingly cool stuff on there.
Initially we kind of looked at the Apple devices and said the best controls to do with this would have been touch screen. And we had a solution for it where you could touch on the screen and shoot the monsters in a very similar way you'd expect with most of the other stuff that's come out on the device. And basically we got about halfway through the project and we just weren't having any fun. Visually it looked great, technically we had solved all the goals we wanted to solve, but it just wasn't any fun at all tapping the monsters. Basically, tapping on the screen, you're covering the monster, you don't see the blood, you kind of miss all those core elements.

So we made a pretty big change. At the halfway point of the project we were kind of considering, this could get canceled if it doesn't get fun. And so we moved it over to the accelerometer controls, and it was kind of surprising, because I think we've played almost every other game called a shooter on the iPhone, and nobody's really landed on this control mechanism before. And that was kind of our "ah-ha" moment.

The goal was to capture the visceral Doom experience down to the iPhone, so we've got shooting, dodging, covering and reloading, all the things you expect in a traditional shooter. And then you get the aiming mechanic in there, and you have that core microsecond compulsion where you're taking aim, you don't know if you're going to get killed by this guy, and it actually really kind of gelled together and became the right ingredient for this title.

John Carmack: From the beginning we were looking at this as an experiment. We obviously know we can do a great job with traditional FPS on a traditional platform, but when you're looking at something that's really different, I was looking at this like my original foray into mobile with Doom RPG. It's a different platform, and we can't do the same type of game that we've been doing, so what do we do from scratch here?

And the fact that we were fully prepared--literally, it did reach a point where I thought I was going to cancel the project, because it just wasn't coming out that fun. And even though we had pumped a not-inconsiderable amount of money into it, we weren't going to release something that wasn't any good. We were expecting this to be our best, first foot forward on the iPhone platform.

In the end, things did sort of work out that way, in that I did the Wolfenstein 3D Classic sort of on a lark, and it turned out more fun than I expected, and we did turn that into a product ahead of our strategized pace. But it'll all turn out well in the end. I think that that turned out to be a pretty good first foot forward there, and this is sort of bringing around one of the big guns for us.

Shack: The graphics are also rather impressive for an iPhone game.

John Carmack: We are cutting a whole lot of corners. There are smoke and mirrors going on here. People look over their shoulders and go, "Holy cow, it's Doom 3 on the iPhone!" But it's really not. It's pretty much old-school game design where we are cheating so much, but in ways you can't really see.

It probably would be possible to do it on the 3GS, to do a more free-roaming game like this with similar visuals, but this is most of what can be accomplished on the first-generation phones. There are a lot of things that we hide very well in the current presentation, but there are no rockets in the game because there really isn't a collision system. The lighting is set up for only specific things happening. We made all the trades to get the type of experience we were aiming for here.

And yeah, it could definitely be faster. I did some performance analysis at the end and we looked at all of this and said well, this is about as good as we're going to get in the time that we've got available. If I wanted to go back and do everything from scratch right now we could do better, but that's true of every single title I've ever done in my life.

Tom Mustaine: Yeah, and basically we targeted--there's a significant speed difference between the first gen iPod, all the way up to the 3G phones, and then also the second gen iPod Touch is then also the fastest device up to the 3GS.

We targeted all of those devices, and you get a good framerate on all of them. And like John said, we pulled some good tricks, specifically there's actually a custom visibility solution in there based on some stuff that John's team did with Rage. So we actually have some really crafted, cool stuff in there to make sure it's fast on every device. _PAGE_BREAK_

Shack: It seems like a lot of you old-school developers are getting a kick out of designing those smoke and mirror tricks on the iPhone. Are the constraints making it more fun to work on this platform?

John Carmack: Well I think that we do have the memory of when game design was a different thing. When it was this much faster turn-around, and you could try things.. a lot of us do have that sense of, while the games we make today are majestic and incredible and really unbelievable--if you showed somebody fifteen years ago what a modern title looked like, they just wouldn't believe you that that was what people were running in their living rooms or their offices.

So I'm not going to go on about the good old days or whatever. Things are spectacular today. But there are certain things that we have lost, as things are these huge productions with tens of millions dollar development budgets, and fifty or a hundred people working on a project. There are certain positive things that you get with a small team working for a short amount of time, on a project where any individual can still move the whole direction of the project.

The glass is half full whatever I'm looking at on there. I try to look at the upside of each platform, but there are certain aspects of working on the iPhone as a project that are positive, rewarding things as a developer that you don't get on the big top-end development.

Shack: What's your take on iPhone DLC? Is that something you're considering for Resurrection?

Tom Mustaine: Yeah, we are absolutely. It won't be in the shipping version, we actually kind of got sideswiped by the 3.0 launch happening so quickly. But we actually have a version of the game that does cooperative multiplayer, so you can actually do peer-to-peer multiplayer, where you see two persons on the screen at the time, and maybe we'll actually have more than that, up to four players.

And it's a blast. You can play through the levels, they scale up, and you can compete for score and play cooperatively. And then also this game really fits well with downloadable content, so we're definitely looking at those with a future 3.0 update.

John Carmack: Yeah, we still really don't even know what the plan is going to be for the 2.0 to 3.0 transition in terms of using new features. We haven't checked into how the App store handles things. Can somebody accidentally download something they don't want in an upgrade? How fast is the market going to migrate over?

And unfortunately Apple is unlikely to give us any statistics of what's going to be a guess about when you want to start doing something 3.0-only on there. I've got the same problem with Doom Classic, where right now it plays on your 2.0 with Wi-fi multiplayer, and eventually we expect to do a 3.0 Bluetooth multiplayer and downloadable content on there. I'm still leaning on doing a 2.0 version first and a 3.0 version later.

Shack: John, it was reported that you were looking to meet with Apple and offer some advice on how to improve gaming on the iPhone. To either of you, after shipping this product, what are some specific things that you would ask Apple to change or improve?

John Carmack: I've got a number of pretty specific stuff, and some of the best news about all of this that's happened recently is that [former Ensemble Studios developer] Graeme Devine, who worked on the Doom 3 project with me at id, as well as Quake III, is now in a significant position at Apple for gaming technology on the iPhone. This has made my life so much better on there, because Graeme doesn't need a translator. I can just talk to him in the most efficient possible way, he knows exactly what I'm saying, and he's just running around making some things happen, finding out why some things are broken.

One of the most obvious, painful things right now is something is really, really sick in the multitouch processing on the iPhone, where you can chew up literally 50% of the CPU by putting both thumbs on the screen and waving them around. Bad stuff happens there, and that's just really unacceptable. The memory management system right now is very, very problematic with the iPhone. The fact that there is no number that if you fit in it, you can be guaranteed that you won't be killed. I think that is long-term unacceptable. Apple will have to put in some at least discreet swapping, if not random paging. I mean, there are directions that I think they will wind up going.

And I don't question what they've done so far--it's hard to argue with the level of success that they've had. And there are a couple of things on the OpenGL extension side that I think they should do. I love working on the platform right now, but there are a few steps they have to take to allow it to live up to its potential as you would see on a DS or a PSP. _PAGE_BREAK_

Shack: What do you think about the hardware upgrade with the 3GS? Is that something you're concerned about if Apple continues to segment the market down the line?

John Carmack: There are a few aspects to that. One, the worry is--and it's still not clear that this will be a problem, [but] left to their own, OS and software companies, whether it's Apple or Microsoft, they tend to just put more and more features into each release. And for a given hardware platform, it just gets slower and slower with each new OS update. Apple has done this before on previous desktop platforms. Sometimes they get speed boosts, but large-scale, long-term things generally get worse as they pile more things on them.

We are trying very very hard to convince Apple that this not the direction they should be going on the iPhone, because we expect that every time somebody upgrades an iPhone to a 2G to a 3G to a 3GS, the old device becomes an iPod for somebody else, and we think that they stay in play there. And I am much more excited to see Apple have 50 million baseline spec systems out there than the latest and greatest hardware.

Now I am very excited about what I can do from a hardware and graphics standpoint with the 3GS. With vertex fragment shaders and OpenGL 2.0, I'm pretty convinced that I can actually run the MegaTexture id Tech 5 content creation pipeline on there. And I'm not sure what game I want to do that with yet, but the combination of seeing people download 700mb files of Myst on there, and the new capabilities, I could do some mind-blowingly cool stuff on there. But that's going to be a much, much smaller market. So there's no way I could justify doing something like Doom: Resurrection, where we take a significantly experienced dev team and have them work for a significant number of months to develop something. You're not going to target something like that exclusively for a brand new emerged platform there.

But I hope I can wind up doing something in the good old classic way of doing things, where I do something that's just amazingly cool to look at but maybe a little thin on the ground in terms of what the actual substance is there. But do that to kind of get things out and test it and see what's going on. Because our mainstream development is going to be focused on the original baseline platform. The 3GS will always run it faster and smoother, and maybe we can add a few simple things to add on there, but you just can't--especially scaling all the way from fixed functions to programmable shaders--it's just a completely different development strategy if you want to max it out.

There's always that tension between saying, "I know what I could do on this hardware given the time and development budget on here, but can we actually justify doing that?" So far the iPhone baseline stuff--Wolfenstein 3D Classic has sold really well. It completely justified even my relatively inexpensive time that I spent on that, and I think Doom Classic is going to do great. Doom: Resurrection is an experiment for us where it was fairly expensive to develop, it was a high-end development strategy on here, unlike most of the iPhone titles so far. The big titles have been ports, and the small titles are small titles.

So this is somewhat chancy ground on here, but if it does as well as Wolfenstein 3D we'll have come out okay. If it sells equal units at the higher price point on there, we'll have come out great, and we'll put twice the effort into the next from-scratch iPhone project. But we don't know. The biggest worry about the platform in general is that if everybody thinks all games are going to cost $1.99, you're never going to DS or PSP-like original titles coming on the iPhone because it just won't be justified.

Shack: That brings me to my next question, which is what is the price of Doom: Resurrection, and how did you arrive at that point?

John Carmack: It's coming out at $9.99. Doom Classic will probably be $5 or $6, we're not positive on that. And this is, you know.. the App store is great in that we get lots of data. We find out what happens with promotions and sales, and it does turn that market into more of a data- driven science rather than just what people think will work out. But there definitely is that worry, if people think that iPhone games just really have to be cheap. Obviously people pay $20, $30, $40 for a DS, PSP title on there coming out at the high end.

But we're still finding out how things are going on the iPhone. It is great that we've always got the possibility to do sales and promotions and things like that. That is one of the real benefits of the App store.

Shack: Will the ZeniMax acquisition affect id Mobile at all? Specifically, will your deal with ZeniMax limit the amount of time you have to work on mobile projects?

John Carmack: So my interaction--the whole mobile stuff starting with Doom RPG, Wolfenstein RPG--first of all it's really been cool to see the evolution of that. But my interaction on those would be, I would go and spend a few days writing some technology, a few days here updating it, a week writing the new 3D engine for the other stuff. Most of the work was done by Anna and her team at the id Mobile team.

The iPhone stuff, somewhat similarly, I started off doing proof-of-concept stuff there with stuff like Wolfenstein 3D Classic, and doing some various technical direction and various work on Doom: Resurrection. With Wolfenstein 3D, it was interesting, that was really the first thing where I did pretty much go in and do the whole project myself. The same with Doom Classic. And that was actually fun for me, to have that sense of total responsibility there, where it's not just, "Well, I'm doing my part, but there are fifteen other programmers here." I don't expect much of that to change.

I do hope that after Doom Classic, we'll be trying to put some other developers on the future classic titles. I don't think I'm going to personally wind up doing all of the work on Quake, Quake 2, Quake Arena on there, although I may need to go in and make sure specific things are done the way I want them to be. And we have so many things that we're excited about on the iPhone that--I like doing the actual raw programming work like we did on that classic games there, but I'm certainly not going to have the time to be that involved in all of them.

I certainly want to do this MegaTexture demo for the 3GS stuff, I work on that myself. But from a high level strategy sense on there, ZeniMax is behind the iPhone. My counterpart Todd Howard at Bethesda is a huge iPhone fan, and they're excited about doing some things. So I don't think anything likely is going to change. Obviously none of this stuff has been extremely well planned out, as witnessed with Wolfenstein 3D Classic and the "Oh, let's just go make a product out of this!" way that that happened. But I don't think anything's changing.

Tom Mustaine: Yeah, it's definitely turned our head. We're going to stay on the Wii as well--we're Xbox and PS3 developers as well--but right now, from our size, the iPhone is really kind of a sweet spot, and the Wii is kind of a sweet spot. We're still very excited about what we can do on the iPhone. We've been talking on-off with John and the team here about potentially doing other products together, so we'll see what happens. But it's a good place to be right now. Hopefully Resurrection comes out and everybody loves it, and we'll see what we can do next.

Shack: Thanks guys.

Hello, Meet Lola