Carmack on id Mobile, Crap Games, and Why iPod Sucks

Overview

Date: Nov 15, 2007
Original URL: http://www.gamedaily.com/games/orcs-and-elves/mobile/game-features/carmack-on-id-mobile-crap-games-and-why-ipod-sucks/5241/71158/
Synopsis: GameDaily interviews John Carmack

GameDaily BIZ spoke to John Carmack at length about the new mobile studio, how id plans to attack the mobile space, Apple's "horrible decisions," overcoming the challenge of developing in Brew and Java for many handsets and more.

Questions

James Brightman: What was the impetus for launching a mobile studio? What's the great appeal of mobile gaming to id?

John Carmack: It's been an interesting journey to get here in that we didn't start off with some grand strategic planning where we decided this was going to be really important. It all started when I got a cell phone and thought the games on it were really bad and we could do a lot better. That led to Doom RPG, which was a big success. Then we followed up with Orcs & Elves and now Orcs & Elves II coming out... By testing the waters like this we've found out it's a viable business right now, we're doing good business, the products are successful and profitable, we've got a plan laid out for a bunch of new things we want to be doing, but I think there's really some significant potential.

The worst, most pessimistic case is that we could carry on as we're doing right now, doing successful games and having fun with all of this, but I think over the course of the next five years or so there's the possibility that the mobile gaming market could wind up becoming significantly larger than some of the traditional markets. There are 10 to 20 times as many mobile phones as there are personal computers and 100 times as many as there are console platforms. As these things evolve into devices more like the media powerhouses we're used to developing for, there's a possibility we could have a breakout success there. I think if we go into this thing thinking we'll do our best, we'll be successful, then we may be positioned for something that could be a really significant tipping point some time in the next five years or so. Mobile gaming really could take off.

James Brightman: Everyone keeps talking about the huge potential of mobile gaming but the percentage of wireless subscribers actually paying to download games remains quite low. How do you see that changing and how does id plan to help push the sector forward?

John Carmack: There are a lot of reasons why mobile gaming is the way it is right now, and my initial assessment is probably shared by a lot of people. Most of the games are just crap; there are a lot of really bad things and shareware type games that people would give away for free on the PC. There aren't nearly as many cases where you take a professional team of developers and try to do something really good. [That said,] you are also significantly hamstrung by the platform itself... the download limits, restrictions in the APIs, huge variability across the hardware platforms and the carriers. It's an ugly market from a developer's standpoint.

Consoles are clean and wonderful. You produce one title and it goes through one publisher and you sell millions of units if you've got a good title. In the mobile market, it's ugly like a retail business. Domestically, you've got a handful of carriers and that's reasonably clean, but then when you go into Europe or Asia, where half or more of your market is, you're dealing with dozens and dozens of different [carriers] and it's a model that's not really comfortable for a lot of software developers. The tiny customization that you have to do means that the sort of hardcore technical programming winds up being not as relevant because you could do something that's spectacularly clever on one platform and have it be something that doesn't even work on a whole of other ones.

You have to approach it in a very different way, but I do think it's clear that the games are getting a lot better. In the last couple years I'd like to think that we had something to do with that with Doom RPG and Orcs & Elves being critically acclaimed titles that sold really well. I know for a fact that we were instrumental in having Sprint raise their over-the-air download limit and to allow us to do a high-end Java version to make it look somewhat competitive with the Brew version. Initiatives and little things like this will allow games to improve a lot more over the coming years to the point where they won't necessarily be an embarrassment to look at. You'll be able to have games on the cell phone that look like games on other portable platforms like the DS and PSP. There's no doubt that right now there are cell phone handsets that have all the hardware power necessary to be significantly better than the DS, but you don't see that in the games themselves for all of these non-technical reasons.

James Brightman: Considering it's been reported that your interest in Linux has wavered, would games be coded in Java or something else?

John Carmack: Right now we develop four primary versions of a title on mobile: a low-end Java version, which is limited to like 350k download size; a high-end Java version, which adds some extra features and takes it up to around 2 megs; a low-end Brew; and a high-end Brew. There's a vast difference in what we can do in the games between Qualcomm's Brew and the Java platforms. You're much closer to the hardware and you can actually write typed assembly code if you need to, so the best looking games right now are definitely Brew games. If people can afford to develop strictly for that they'll make much better games.

Unfortunately it's only about a third of the domestic market and a tiny fraction of the European and Asian markets. So the industry is definitely not united and [mobile] is not a great developer's platform because of this huge amount of diversity. If we were left to our own and the only idea was to make the best possible mobile game, we would develop for some high-end Brew platform and we could make a spectacular, stellar but it would only be seen by a relatively small fraction of the mobile playing market. It's one of those tradeoffs. We're obviously not going for everybody or else we would be developing 64k card games - these relatively modest games that can run on anything - which a lot of people think is the key to success in the mobile market. But we're happier developing more sophisticated games that are better games but can only hit a somewhat smaller fraction of the market. If we establish ourselves here I think we'll be well positioned as the hardware market sort of catches up with where we're going.

James Brightman: id's announcement that Rage was coming to Macs was a huge deal. Since you seem to have a long personal interest in Macs, might future games from Fountainhead/id come to the iPhone or iPod?

John Carmack: We've certainly been looking at it but Steve Jobs and I have not been seeing really eye to eye on a lot of important issues. We were in a fairly heated argument at the last WWDC [Worldwide Developers Conf.] and we've had a few follow-ups. I have an iPhone right now and it's a platform I would enjoy developing for but Apple is not taking progressive steps in regards to [gaming]. Their strategy seems to be working just fine from a business standpoint, so I'm not going to second guess them and tell them they're being fools or idiots for not focusing on this.

The honest truth right now is that Apple's not exactly hugely supportive of this. When they finally allowed games to be put on the iPod... in many ways it's one of the worst environments to develop games for. You have to work on an emulator... just all these horrible decisions. I expressed my fears directly to Steve Jobs that some of these mistakes might be carried over to the iPhone, so they're at least aware of all of them, but they're not giving any spectacular signs that it's going to be a big deal for them in the next year.

James Brightman: There was an interesting mobile news announcement recently from Google, which launched the Android OS with multiple phone makers and even carrier networks. What type of impact, if any, does that announcement have on id's mobile business?

John Carmack: It might, but I've not looked at all the details of what they're doing. The most important thing for us is native code execution. Java was sold in many ways as a security architecture, and there's some truth to that at least on early cell phones... but for modern cell phones with virtual memory and memory management units, traditional OS methods really are appropriate. And that was my big argument with Steve Jobs about the security measures on there. If you've got a real OS and a real MMU on there, you can have secure native mode applications just like you do with any traditional Unix system. I certainly hope that becomes the direction things go in. Java still does have the benefit of being able to run across different CPU architectures but Qualcomm seems to be content with saying ARM is it. If we can get away with that and say ARM owns the mobile space, as a software target it would be much, much more desirable from a developer's standpoint.

James Brightman: In terms of your mobile games strategy and your portfolio, are you looking to create new IP, just continue with IP that's been working like Orcs & Elves and Doom, or both?

John Carmack: Orcs & Elves was actually the first new id Software IP since Quake. While we were working on Rage, our big high-end title, we were able to sneak Orcs & Elves after Rage was started and ship it well before Rage was going to ship. The fact that we could get that out and have it be successful I'm really pleased with. So we intend to pursue both; we have additional new IPs we're going to be specifically aiming at the mobile market, but we're probably going to be hopping back and forth between traditional id IPs and new IPs. The project under development right now is the Wolfenstein-themed project. We are probably going to try to do that simultaneously on mobile and DS, which would be a first for us. We're still sort of feeling out all these different strategies about where IPs start and where they go, whether it's simultaneous or subsequent development.

James Brightman: What's the plan for marketing and publishing id's mobile titles?

John Carmack: Currently we're working with EA, but one of the nice benefits of being id Software is that we can get away with signing only one-at-a-time publishing deals. We don't have any long-term deals tying us in. We're reasonably happy with the support from the Jamdat guys that are now a part of EA, but if another publisher wants to make us a really good offer we're certainly willing to listen.

Todd Hollenshead: For example, Orcs & Elves II is being published by EA but beyond that when we've talked about the future of what id Mobile is going to be bringing to the market, those titles are all completely unsigned.

James Brightman: You touched on this a bit before, but in terms of the power of the mobile handsets and how they could easily eclipse the DS, obviously not every handset is created equally. And as PC developers you are quite familiar with trying to create something that'll work across numerous specifications, so when you're optimizing these games for mobile how do you make it playable across handsets?

John Carmack: In that respect it is actually worse than the PC. The PC feels bad compared to the consoles - whether you're on a DS or 360, it's great because you have one target, but on the PC you might be scaling over a factor of four. On mobile we're scaling over a factor of at least 10, and that's pretty grim. So we make significant gameplay decisions that strategically at the very beginning have to affect all of that.

Katherine Anna Kang: So far I think the strategy that's worked for us is that we have four versions (high-end and low-end for Java and Brew), so we know that the lowest version of the game should never look or play worse than the lowest that we've developed and given to the publishers. So we have a very, very bottom and a very, very top, and after delivering those to the publishers they will take those versions and port it over to the hundreds of handsets that need to be taken care of.

James Brightman: I would assume that significantly raises the cost of development, right?

John Carmack: Oh yeah, it definitely does. It's hard to say what the breakpoint would be, but we do feel that it's fairly necessary for what we're trying to accomplish. We could do some profitable games if we were high-end Brew only, and they would be much flashier and more impressive, but we'd be serving a much smaller chunk of the market and we wouldn't be able to accrue the consumer goodwill and broad based IP market growth that we're trying to do. The thing about the cell phone market is that it's possible to go out there and sell millions of units. These are good numbers just to have people say, "This is a great game." Whether they played it on mobile or some high-end platform, if they think, "Orcs & Elves, that was a great game..." then we've built some long-term value that the company's going to be able to take advantage of for year's to come. And while we might still be able to make money doing a high-end only Brew game, and in some ways it would be a much more fun thing to develop, we would lose out on that broader sense of a couple million people having favorable impressions of the game and company and so on.

James Brightman: And finally, the first games to come out under this new mobile studio are expected when?

John Carmack: The Wolfenstein title is under development right now, and we expect that to follow the previous [pattern] of taking six months to develop...

Katherine: The idea is that if we do the code release of the cell phone and DS version of Wolfenstein, we'll definitely have it out for next year's holiday. If we don't do the code release, the cell phone version won't be out before the DS.

John Carmack: My ideal would be if we winded up developing two titles internally and then we probably work with maybe two other outside studio teams to develop two more products, like the Quake Arena project we're talking about for the DS and possibly deal with somebody else on another mobile title. Ideally, I would hope to see four mobile titles coming from id Mobile next year.