Maximum PC john carmack RSS Feed

John Carmack: Quake Live Will Make Consoles Green with Envy

Nathan Grayson — Thu, 26 Feb 2009 02:02:16 -0600

That John Carmack! What a gossip, huh? He goes on a beer run with PC gamers and he’s all like, “I just wanted to… I just… I love you guys!” But then, after totally crashing a console-only party, he’s singing a different tune (possibly while wearing a dog costume). And now, once again Carmack’s shacking up with PC game—oh no! He’s here! Please, please, please don’t tell him what we said. That’d be soooo awkward.

"A lot of [Quake Live] was about doing something that the PC was going to be better at than the consoles," he told Gamasutra.

"Our modern triple-A stuff has to be somewhat more console-centric, with the PC as a peer, while this is an opportunity to do something where the PC will really stand alone,” he noted.

Carmack hopes to see Quake Live blossom into a sort of social-networking service – the one toy at show-and-tell that even Microsoft, Sony, and Nintendo only wish they could get their grubby mitts on or toss in the sandbox or put in their mouths or whatever makes sense with this flimsy analogy.

"For years, I've often thought about the fact that a lot of people spend vastly more time on websites and forums about the games that they're playing than they actually spend playing the games themselves," he adds. "We hope to have some aspect of that here."

Well, that’s good enough for us, John. We’re yours forever now… wait a minute! Did you just steal this gift from our shelves – our shelves marked “1999” – and rewrap it? Is this all we are to you?

John Carmack: Keyboard-Mouse Still Better Than Controller for FPSes, Tightening Up Graphics on Level Three

Nathan Grayson — Wed, 26 Nov 2008 00:03:23 -0600

id Software boss and brain-god John Carmack may have softened his stance on consoles as of late, but don't get him wrong; his heart still has a hard-on for the PC, and he's not afraid to show it.

“The keyboard/mouse interface is definitely still the superior interface for a competitive first-person shooter experience, much better than an analog joypad,” he told PC Gamer.

But why stop with games? Clearly, the PC can do at least two other things.

"The browser environment is faster—navigating web pages on the console is a really tedious experience… And I do think there’s the whole idea of PCs being everywhere, and having a game that you can play just about anywhere. Anywhere there’s a PC, if you’ve got a few minutes you can download Quake Live content and jump in and play your game,” he said.

However, Carmack conceded that console development definitely has its perks -- for instance, acting as a hardy shelter in the hail of issues that is PC development.

"There are interesting technical things, looking across the spectrum of graphics cards, looking at the very latest stuff on there, but there are also times when I say, 'Wow, the 360 is a nicer place to develop games.' You bypass a lot of the issues there. Wouldn’t it be nice just to develop strictly for that platform?"

No.

Doom Creator John Carmack Levels Up as Part Time Rocket Scientist

Andy Salisbury — Wed, 29 Oct 2008 17:07:50 -0500

For those that don’t know, coding your favorite games isn’t the only thing that John Carmack does well, turns out he’s not half bad at rocket science. He’s proved this most recently at the Northrop Grumman Lunar Lander Challenge, which is being held by the X Prize Foundation and sponsored by NASA. The goal of the challenge is to help in the development of a fleet of lunar ferries that could one day carry people and payloads between lunar orbit and the moon’s surface.

The challenge consists of several levels and Carmack’s team, Armadillo Aerospace, has passed the first one. Level one requires that a rocket take off from a launch area, climb to an altitude of 150 feet, hover for 90 seconds and then land safely at a landing pad 150 feet away. They were then required to repeat the flight in reverse within two and a half hours. Their ability to complete this goal before any other team has won them a cool $350,000.

Level two is a bit more difficult. It requires that team double the amount of time hovered, and then land on a simulated lunar surface that’s littered with craters and boulders. Armadillo Aerospace attempted the course, but was unable to complete the task. That means that there’s still $1.65 million up for grabs.

Mr. Carmack, we’re going to have to insist that you keep doing great things, but don’t forget about us!

Image Credit: CNET

Exclusive John Carmack Interview, Part 2: Nvidia vs Intel vs ATI

Will Smith — Fri, 08 Aug 2008 15:55:25 -0500

Click here for the first part of the interview!

Here’s the second part of our exclusive QuakeCon interview with John Carmack. In the first part of our conversation, Carmack discussed his hopes for Quake Live and the id Software’s new gaming direction in Rage. This time around, he gets more into the heady technical stuff with his thoughts on Nvidia’s CUDA, physics accelerators, general purpose computing, and ATI’s rumored Fusion technology.

MaxPC –Can we talk about PhysX and GPUs and Cuda and stuff like that for a sec?

John Carmack – I was well known as not being a supporter of the PhysX accelerators. It’s always felt like a gimmicky plan with people setting up a company to be acquired. For years, the tack has been what do you do with any time Intel delivers something more with processors and more cores? It’s never really proven out right and there’re a lot of reasons for it.

For one thing you can’t scale AI and physics in general with your gameplay, while with graphics, you could scale. Without scaling, you can’t design a game that requires fancy AI and then turn off the fancy AI for the low end systems because practically that’s not possible. Similarly for physics, if it’s anything other than eye candy, you also can’t scale. If the building is going to fall down you need to know whether you’re going to be able to get past it on the high end or the low end.

So what’s happened of course is that PhysX is degenerated to fancy eye candy. You got your fields of grass, you got your walls of blocks that come tumbling down and things that aren’t crucial to the actual game, and that is just a fancy cookie that you throw at the player, which admittedly has some value. So in terms of the general purpose acceleration it was clear even when AGEIA was starting, that we knew that the graphics processors are going to be getting more generalized, and we never thought that they had any special sauce in their hardware that was fundamentally going to be better.

So what’s going on with the Cuda approach is and I think Nvidia is being very wise about their approach where they’ve brought in something early on so some people could start getting some things done with Cuda. So they’ve got a community of high performance computing research guys working with Cuda and it’s great because it’s so important to get that out of your labs and into a customer’s lab and just seeing how things work in the real world. They’re going to have several generations of extra insight over Intel by the time larrabee ships.

Right now the switch between GPU and Cuda is a really heavyweight switch. In the next-gen stuff, it’s much more lightweight so you can toggle back and forth, and in the future, it’s all mix and match. They’ll [eventually] run GPU and Cuda processes simultaneously and it opens up a lot more avenues for computation. There are still some fundamental worries that I have about vector length on there where all of these things that are set up to be GPUs first they’ve all got very long vector lengths. So while you may have a 128 sort of banks of threads, each of them are doing 32 things at the same time. I still see a huge potential for miserable utilization where even if you could suck up all the threads, if you don’t have something that can use wide vectorization, you wind up with only 5% utilization.

MPC – that combined with the heavy switch is disastrous right now, right?

JC- Yeah, you can’t really use it in a game right now. It doesn’t make a lot of sense but it’s going to in the near future and by the time we get to next-gen console stuff all of that is going to be a nice finely integrated stuff. Right now you have this continuum from a general purpose processor like we’ve got as the main CPUs on the 360 or the PS3, then you’ve got like the Cells which are general purpose processors but they’re all wide vector with no caches special DMAs. Then you got things like Cuda thread processors and each one is more hassle than the other and the one before that.

The Cuda processors are moving up, clearly. They’re going to get caches and more general purpose programming abilities but they’re not going to be all the way to what larrabee is doing which is really independent processors with a couple of cores and a couple of threads. It’s going to be interesting to see how all that plays out where, my suspicion is that for a lot of applications they’re designing and benchmarking for, Intel will wind up having good performance. But the internals of it, the software that they write for it is pretty ugly while the code that you could write for Cuda is pretty clean.

I think Intel is going to be fine on the peak performance numbers and will probably have a process advantage, which is always one of Intel’s big hammers. So it’s going to be interesting how Nvidia’s greater experience in utilizing all this parallelism plays out versus the kind of might that Intel is going to have in their raw process advantage in applications.

MPC – Where does ATI fall into this?

JC – We’ve gotten the pitches on the Fusion project and how they’re putting it together with the more general purpose stuff, like with the AMD CPUs on it. We have less insight into that than we have into other projects. In general ATI doesn’t have quite a good developer relations support as we get from Intel and Nvidia. Again, it’s going to be interesting to see how all that plays out. I know their market share isn’t doing real well on the different PC cards.

MPC – Do you think an open API will help them?

JC – It’s a tough thing because I think that trying to spec an API for experimental hardware like this is really tough, and like I said last night it’s very different than what it was with graphics where we had examples of all that research that had been done and we knew how to do it and we were just cleaning it up and doing it better.

But in this type of situation, we really can’t say that anybody that gets up there and clearly acts like they know exactly the way things are going, is just putting up a good front because the work just hasn’t been done yet. Nobody has written major applications that are working on these things, and one of these approaches may turn out to be fundamentally better than the other. We just don’t know which one.

id Software's Programming Director and Lead Designer Explain why Rage will Kick Ass

Will Smith — Fri, 08 Aug 2008 13:50:57 -0500

John Carmack may be the face of id Software, but he’s definitely not the only person working on Rage or the next Doom. We spoke with Robert Duffy, id’s Programming Director, and Matt Hooper, Rage’s Lead Designer, about their upcoming shooter. The conversation delves into topics ranging from art design to multiplayer modes, and touches on the challenges of developing on both console and PC hardware.

MPC: It seems like you’ve done some really neat stuff with engine tools on Rage. John [Carmack] said in his keynote that you’re doing simultaneous development on three platforms so you can flip a switch to produce builds for each of the three?

Robert Duffy, Programming Director at Id Software: That is essentially true. We’re actually doing simultaneous development for four platforms. Occasionally one platform will lag a little bit behind, but generally speaking, when we do a build, it’s produced for all the platforms. We don’t have any big media packing step during development that requires people to have to wait. We’ve set everything up where even if artists are working on a console they can change media and reload to see it instantly. Even though we’ve grown a lot, we’re still a very small team compared to a lot of companies so we try to keep that workflow going [smoothly].

MPC: To get a specific concrete example: a designer comes in, makes a stamp change on a megatexture or puts new art in the game. He can then see what it looks like on PS3, Xbox, PC and OS X?

RD: Well, for stamping it’s even cooler. If the artists or designers are stamping, those changes appear real-time on all platforms for everybody in the company.

MPC: So it’s like a multiplayer game?

RD: (laughs) Sort of.

Matt Hooper, Lead Designer of Rage: We have source textures, source models and things like that, that have to make their way in the game, but the cool thing about all the platforms is that those assets are shared. So like Robert was saying, you don’t have to do anything fancy to get a model to show up on the PS3, you just have to run the PS3 build and your model will show up. Any new art that is put into our source control is just going to show up.

One of the cool things about having it all virtualized is that all of the platforms pull from this one source (our megatexture server), so it’s all stored on the network. Anybody making change to that -- if they stamp or just go crazy and they’re changing the surface texturing – updates the media for everyone else. So it’s really just a giant bank of surface texturing that we’re pulling from all the time.

MPC: Can you explain a little bit about megatexturing and the virtualization of textures, and how that differs from what happened with Doom3 or id Tech 4 engine games?

RD: In Doom3 and all games before id Tech 5, you have a whole bunch of textures that end up being used in levels or landscapes and things like that, take up texture space on the video card. You end up having to manage that really well. So you end up repeating textures on surfaces, and you can get some pretty good looks out of that. What John came up with towards the end of Doom 3 was the landscape rendering in Enemy Territory: Quake Wars. He essentially unified texturing across the entire world, so not only outdoor areas but all the indoor areas.

We build this incredibly large texture, for instance a landscape has a 128K x 128K texture that I think when uncompressed is about 192GB. Then we do all kinds of compression -- although you could run it that way – and one of the final pieces of the tech were doing is actually compressing that down to shipping size. It produces this unique layout which is essentially just a mapping that the tech reads. It looks at a surface, figures out its mapping and then goes and knows where to retrieve that [data]. The texturing spreads across every single character, the world, the indoor areas, and is dynamic. Everything is a unique pixel. The designers and artists have the freedom to actually touch every single pixel in the entire game if they want.

MPC: In Quake Wars, are bases built into the megatexture as well or are they placed on top of it?

RD: In Quake Wars that is not part of the megatexture. Quake Wars uses a more traditional texturing approach. In idtech5, building textures are completely unique, just like the landscape.

MPC: Everything? Even in indoor areas?

RD: Everything. Obviously, we don’t virtualize the user interface textures, the fonts, and some things like that. But generally speaking every single thing in the world is completely unique.

MPC: What kind of extra freedom does that give you guys? What can you do with id Tech 5 that you couldn’t have done with Doom 3, aside from the obvious giant outdoor areas?

MH: There are two main things. First of all it’s the ability to have artists who don’t necessarily need to understand all the technical ins and outs of the engine. We set the surface texture budget and then literally all they have to do is make great art and stay within that initial allocation. But they’re no longer worrying about what to call out or what textures you have to take out near end [of development] because it doesn’t matter. It’s all unique and the cost is already fixed. We’re running at 60Hz [with Rage]. You can change all the textures in this room if you want and it’s not going to hurt the framerate at all. Taking that [limitation] away allows some production freedom.

The initial kind of fear was with the [task of] uniquely texturing everything. You start to think ‘Wow that’s kind of a scary thought, we have to make all of these unique textures,’ but we still build kind of in a traditional way but without the restraints. It doesn’t take a lot for an artist to make a 4kx4k texture – they actually want to do that. What they’ve had to do [before Megatextures] is make 128’s or 256’s or worry about things like ‘that mountain in the distance can’t be seen so I have to make a lower resolution version’. Now they’re just fixing the art at this constant resolution. [Megatexturing] actually makes it easier for them.

MPC: Is it a challenge to render a scene that has more texture than you have memory for? Is this a problem when developing for consoles?

RD: The nice thing about this technology is it essentially uses a fixed video memory cost. It maintains several very large textures in video memory and it’s constantly paging in from disk.

MH: It’s streaming what it needs, when it needs it.

RD: It’s essentially paging in the mip levels that you need into these textures and mapping those out. For instance, on the Xbox360, you can govern what you use for video memory versus what you use for other memory, it’s great because we have a little more memory freedom. We can say ‘ok we need this much for video memory.’ On the PS3, because it’s a hard split [between video and system memory], we struggled a little more with memory constraints because 256mb of the memory on the PS3 is reserved for video memory and we would rather be able to switch it because we don’t even use that much.

MPC: On the PC, you have some video cards with 1GB frame buffers. Can you swing wide to accommodate them?

RD: There will be settings in the game for PC users where they can set larger memory textures and you’ll get a little crisper [visuals], but we certainly won’t require a 1GB card.

MPC: It’s nice that the people who have those high end systems will actually see something that looks a little bit better.

RD: Yeah, they’ll be able to turn on some additional filtering.

MH: The texture resolution will get a little higher [for objects] in the distance, so they can use that (additional) video memory.

MPC: Let’s discuss multiplayer in Rage.

RD: There will be a multiplayer component.

MH: We just haven’t down nailed it down exactly how. Our primary focus design-wise is making the best single player experience we can, but we do expect to have multiplayer components.

MPC: Is it safe to say there is going to be both driving and FPS multiplayer components?

MH: We haven’t committed to any [ideas] but all of those things are on the table right now. There’re some logical conclusions, we just don’t want to nail it down right now.

MPC: With the combination of driving and fps gameplay, what’s fun and exciting that we should look forward to that we haven’t seen before in games?

MH: The thing you haven’t seen is really the mix. We’re still id software and we’re still making this intense, action shooter game. Those moment to moment, finely crafted action sequences – running around with the coolest weapons and shooting guys – that’s still there. We invented that and we’re still going to do that really well. Just around the office everyone likes a lot of cool games. What we did was pull in these different elements that don’t detract from the action but add this little bit of flavor, and the vehicles are a part of that. The vehicles are almost an extension of your FPS avatar – you’re “running” around with a vehicle. It has armor on it, it carries a cool weapon, you fire that weapon, and the other car blows up in a cool satisfying explosion. It’s not as far removed as you would probably initially think. It all feels really good together.

By virtue of getting in the car and going to the next cool shooter environment, that separation allows us to do wildly different things in the different areas around the wasteland. It’s almost like the coolest 3D interface you’ve ever had in a game. You go back to town, talk to these cool memorable characters, get bits of the story, buy your next weapon, maybe take a little twirl around the track to prove something with your car, and basically just get ready for that next shooter experience. We think it’s a good mix. There’s nothing really like it which is scary at some times.

It’s hard to describe all the different elements because at first it seems like they don’t fit together. The main thing is that it’s always an action game. It's always going to be fun and arcade-y. Now were doing a lot of things that we haven’t pushed for in the past; we have all of these cool story elements. The story is so unique and diverse.

We talked about the tech, the virtualization of all the textures and how you can uniquely edit any surface in the game. We almost have that same thing on the gameplay side; we’re trying to make these really unique and diverse environments, really unique and diverse characters and story. It has some roots in things that can resonate [with people], like a giant meteor comet smacking into the earth. On top of that are all the different factions you are fighting against. It should be a really cool experience which always goes back to that unique and diverse [design]; from the technology to the gameplay fidelity.

MPC: It almost seems like what a single player MMO experience would be like. Going out and exploring this giant world, having cool stuff to do and having actual journeys to get there and not being a straight linear experience.

RD: You do have some choice so you don’t have to do things linearly. The other thing we really want to try to get really right is the vehicle combat because vehicle combat has always just kind of sucked. We are trying to make the car a true extension to your avatar where you can do some cool stuff with it. We tried to show some of that in the [preview] video, that we want it to be intense and fast paced. That way it’s a little more memorable. You take a mission to go out and do something and you may encounter some of this stuff and have to fight your way through a little bit. We don’t want that [style] to overwhelm the game but we want that to be something that is very fun and very memorable for players.

MPC: How much of the game is straight, designer created interactions and how much is dynamic? Or is there a mix between the two?

MH: It’s kind of a mix but were still doing that fine-tuned shooter experience. A lot of the action, even with the vehicles, is going to be that way. We want to have that crafted “moment to moment” feel. It doesn’t mean that you might be driving through a wasteland and see a plane that’s crashed on the side on a hill and then get out of your vehicle to explore it. There’re things like that in the game, but you always know that you are on your way to do the next mission. There’s definitely an exploration feel going on but it’s pretty direct and so it fits both needs. The people that want to explore and get their fix, and than the guys that just want to plow through and have that really straight forward experience can do that too.

MPC: One of the things John said last night was that ‘modern games coddle the gamer’. It’s a gentle shepherding through this single player experience. Are the hardcore gamers going to be able to take hours and hours of punishment instead of fun if they want that?

RD: We haven’t got to the point where we’re tuning different difficulty levels. One of the things that Matt and Tim [Willits] do really well is pointing out stuff. They’ll get some gameplay idea to one of my guys and we’ll talk about it and what they conclude is that they’re smashing the player over the head. They’re really great at saying ‘you know what, people hate dying’ and we want to make it challenging but we don’t get into that repetitive ‘I have to do this again’ [frustration].

I think Matt and Tim are doing a really great job at fostering that [balance] throughout the gameplay. But I would assume that whenever you get up to the hardcore nightmare levels that we end up getting into the game that they’ll be really, really tough. We haven’t balanced for any kind of driving gameplay before. We know how tweak the first person stuff, so [driving] will be a work in progress for us.

MPC: When you make a giant open-world type game, a lot of the challenge and a lot of the concern from other developers I’ve talked to in the past is that they build a whole lot of really awesome content that nobody sees. What do you feel about that and what are you doing to prevent that from happening?

MH: There’s the [main] storyline and then there’re little offshoots from it, but we’re not worried about it. There are some things that people will never see, but that’s why you just have to be smart, production wise – where you’re spending your art time. That’s actually one of the cool things about the tech: it allows us to put a lot of detail in the areas we care the most about. (For instance) the player is going be in the town a half a dozen times, so we need to make that the coolest and [as] best a looking area as possible.

It’s a weird thing for people on the outside of development because they look at a four maybe five year development cycle and they calculate ‘ok there's 20 levels and there’s five years, so it took that long to build a level,’ but it’s really not that way. It’s that last year where you do all the cool stuff and you put it all together. We want to make it as cool as possible for the people that do explore. And if they only see 20 percent of the extra stuff, we’re fine with that. I think that’s what makes games cool; maybe they tell their friend ‘Hey did you see that thing here?’ so you get that cool word of mouth from people on message boards. It’s why we’ve always had Easter eggs.

MPC: What is Rage doing on multicore CPU’s?

RD: We’re doing a lot of work on the SPUs (Synergistic Processor Unit) of the PS3; we’re using multicore pretty heavily. We expect on the PC it would run on a standard CPU, so I don’t believe we will require a true multicore. We are also doing a lot of threading for a lot of things. Most of that work is going into the PS3 and the Xbox360, although the PC certainly takes advantage of it. It’s just that on the PC you’ve got a much beefier main processor so we’re able to thread that stuff off and not stress as much about thread priorities like we do on the PS3 and the Xbox360.

MPC: Does the work that you do with the threading stuff on the consoles carry over to PC?

RD: There are certain elements that carry over, especially between the Xbox360 and the PC. The PS3, due to the nature of the SPUs, has a lot of specialized work that’s going on, but the general concepts carry over. We just have to take an extra step on the PS3 to get the performance that we need.

MPC: If you’re running the game with a quadcore or multicore CPU, is there a benefit in performance or image quality?

RD: We hope to be completely stable at 60hz across all platforms. Since the consoles are our primary target they are dictating a lot of the decisions. This is pretty far out, but I can predict that we may be able to throw some nicer effects and physics explosions, but that’s all just speculation at this point. It will definitely perform better on multicore CPU systems, it’s just how we surface that to the players to make it a neater experience. We are architecting this for the consoles primarily. The PC (users) are certainly all first class citizens; it’s just that we want it to look and feel the same on everything.

MPC: Rather than run the loop and thread it out, are you doing different subsystems on CPUs?

RD: We run sound, the renderer, background streaming, and we’re a bunch of the collision detection in their own thread these days; we have a lot of threads going on. Again, it really comes back to prioritizing and figuring out the best way to organize and maintain that on the consoles.

MPC: What about sound? How are you guys handling sound in Rage and Doom 4?

RD: We’ve got backends for all of that stuff. Doom 3 had a full software audio renderer and we’re using hardware for most everything now. We can always resurrect the software portion but we’d really just like hand that off.

MPC: Why no OpenAL on the PC?

RD: That’s not a concrete decision yet. Using the same format as on the 360 gives the two platforms parity. And there are a couple of features with the XMA stuff that we liked. We could end up basically back on OpenAL with everything; we’re still kind of back and forth on that.

MPC: Are you using external or internal physics?

RD: Internal, we don’t use any third party.

MPC: How close to true physics are you doing? Super fun or super real?

RD: We are doing super fun, as you saw from the explosions yesterday. With Doom 3, we have a really, really good physics guy and Doom 3 had good physics. We’re not trying to necessarily compete with the physics-only companies, but we have a very good physics system. Since we license the tech as well we want to give people something that’s very competitive so it’ll be a very competitive and fun physics system.

MPC: What are the benefits of doing it internally?

RD: If we have something break we understand all the code and know how to go fix it.

MPC: How about Rage as a benchmark?

RD: Like past id games, there’ll probably be settings on the PC that you can crank up and run it at whatever hertz you want. Currently, there’s a command you can run in real time and run at 30Hz, 120Hz, 87Hz – whatever you want. I doubt we take it out, but 60Hz works great.

Exclusive John Carmack Interview: The Godfather of Frag's Plan to Save PC Gaming

Norman Chan — Tue, 05 Aug 2008 20:33:08 -0500

We interviewed John Carmack back during this year's E3 when id first announced a partnership with EA to publish their next shooter, Rage. We had a chance to sit with Carmack again at this past weekend's Quakecon, where we followed up on our earlier discussion to squeeze more details out of the legendary game developer. Carmack dished out more details about their plans for Quake Live (including their high expenctations), the technology powering Rage and the next Doom, their cancelled Darkness project, and his thoughts about the current modding community.

Maximum PC – At E3, we talked about Rage and we talked about EA. Today, we’d like to go over the stuff you were talking about last night about the kind of future of PC and the state of PC gaming.

John Carmack, Technical Director at id Software – And obviously that’s our roots, so we have pretty strong feelings about all of that. But at the same time, we don’t want to try to swim too much against the tide of a larger market shift.

MPC – Of course, as a business, you need to make money. What are the challenges facing PC gaming? What do you see as the biggest challenge? Is it piracy?

JC – Piracy is a part of that. [Multiplatform development] complicates the development process but not the distribution process; we just need to make sure it works on the bunch of different things. A lot of it though is just market migration where a lot of the people who would’ve bought our previous games, you know Quake 2, Doom 3, whatever sort of high end PC-based titles on there – a lot of them just prefer to play games on the consoles now. They moved on to those platforms.

It’s always hard to say how real the numbers people tout out for piracy are. Because for one, how do you estimate how many people are actually playing? We do have lots of cases where the download numbers from one piracy site is more than the retail sales numbers for Quake 4, and obviously that’s a just fraction of the pirated copies, so many times more people are at least trying a pirated version.

The question then becomes how many of those are lost sales of a real title, and it gets even really kind of ugly when you think about a cross-platform title where you have console sales and if anybody is pirating on the PC who might’ve bought a console version you start getting into this case of “well, maybe selling a few hundred thousand units on the PC is a good thing but what if we lost more than that many units off of console sales?” So that is tough. It’s going to be interesting to see how the numbers play out on Quake Live because if we do wind up with five million plus people or something playing, that means market demand is still there and it probably speaks toward the idea that piracy is a real problem.

MPC – Games are expensive now and a lot of times we’re seeing games hit shelves that are $60 on consoles. It’s a kind of scary decision if you had to buy one game a month you want to make sure you get something that’s fun out of it.

JC – Yeah that was kind of like my comment about how, you know – games that are $60, that’s a lot of money, so games are expected to have everything. They’re expected to have incredible amount of media, all these kinds of gameplay modes and it just forces everything to this “everything and the kitchen sink” sort of mentality for game development. And that’s why we really pretty much decided when we were looking at our next game for over the last year that we didn’t think that a new Quake Arena type title was particularly suited for our modern cross-platform high end game. While technically, we think we could do a great job on it – we’d make it an id tech 5 base, 60 fps, and add in all this other stuff, do great avatar modeling and lots of things we’re excited about doing -- there’s not a lot of pull from the publishers for doing a game like that and their recent data points just don’t look that good.

And we don’t want to try and ram something down publisher’s throats that they don’t really want it. We want it to be something that everyone’s happy with and the exciting thing about Quake Live is that it’s still something that caters to the PC’s strengths as a platform. It’s still mouse-keyboard on there, the best way to play that type of game. It’s something that you will be able to play at any PC, really, high-end or low-end. We think that, even though it’s not a modern graphics game, it still looks and plays great and it’s the opportunity to show, if we wind up with five million plus users, then there will be all the incentive to look at a [similar] PC-based title.

MPC – So if the users come then there’s potential for something in the future.

JC – Yeah because we would like to do that. I mean I can appreciate on some level the story-driven movie-like games that draw people through an experience, but personally I appreciated just the game activity. You know it’s something that you do just to have fun that’s not the type of game that’s going to be a modern development project. But our bet is that it may be possible to create a really successful little niche with Quake Live. We’re going to find out in 6 months or so.

MPC- And it’s safer to spend 6 months than $30 million and 4 years.

JC – And it wound up being a year. We thought it was going to be 6 months but yeah.

Tim Willits, Creative Director at id Software – But again I mean Quake 3 was released in 1999 and still, it’s still the best on-foot, 1 on 1, pure deathmatch game and has lasted the test of time. We still have a great following, if you look at the Gamespy rankings. We want to get those fans and some people that have never tried it and just put together one community.

JC – The one part where you won’t see quite yet (in the Quake Live beta) is the matchmaking, because we don’t have a large enough body of people to properly get that going. But within the next month that should be there. And we want that experience to be about saying to a friend “hey go check this out” and not saying “go pay $60 bucks for this brand new game,” which is a bigger decision. It’s just go to this URL, spend 15 minutes, download it, play through the game and see if you think it might be kind of fun.

And I think that there are a lot of people that can fall into that market. It’s a focused, fairly pure game, but it’s something that has legs that people can and have played for 9 years in a row. And with this little bit of an extra level of polish making it easier for people to get into the game, there’s every chance in the world that more people are going to wind up playing and enjoying Quake Live than they will Rage and Doom 4, because well, being free is certainly a big asset.

TW – And no updates and weird patches you have to do; it’s all just there.

MPC – So how does the service work? Is it kind of like the Halo model with different ladders or channels or something you sign into?

JC – The servers are a resource that is managed by us. That means starting them up and shutting them down. That was one of the big decisions over the year about trying out independent game server providers or letting people run it themselves, but they ended up being managed by us. They’re all fungible so the system can be started them up and shut them down remotely when needed. And I don’t know what exact rules they’re using for determining how to start things up but there’s a lot of thought that’s gone into it. You don’t want dead servers; you want to make sure that there are servers where people like to be playing. People can of course start up specific ones for their friends, but we’ve spent a lot of the time has creating the rules for the servers.

Quake Live's Chat Module (click to expand)

The chat module is built into every web page of quakelive.com. Once a player creates an account, logs in, and adds other players to their friends list, they can then always see which friends are online, chat (with friends on the site as well as friends in game), see which friends are in a game, easily obtain information about the game a friend is in, and join them with one click. The illustration demonstrates rolling over, and clicking the icon next to a friend that is currently in a game.

MPC – It sounds like all the kind of good stuff of like the Halo 2, Halo 3 style match making without the host advantage.

JC – Yes.

MPC – Very cool. Ok so, Rage and id Tech 5. You guys are going DirectX 9, which we completely understand and we think most people understand the reasons for doing that level.

JC – Well on PC we’re actually still OpenGL.

MPC – Right but DirectX 9 level of shader. Do you really think that Directx 10, and 11 now, are even necessary?

JC – They really aren’t. The things that you get in there are the geometry shaders and a few other things. There’s not a huge draw off that and that’s the danger of leaving an API that’s kind of reached a good stable level. DX 9 is a nice mature setup technology. It’s kind of the natural evolutionary peak sort of the old OpenGL model. It’s really taken that and it’s better. It’s cleaner and better defined.

So you got a whole bunch of people at Microsoft that make APIs. They have the idea that “we made this API that seems to be doing everything everybody really needs but we need to keep doing a new API every year or two because that’s what we do, otherwise they might dissolve our department” or something. So it’s not the same as where up through DX 9 everybody obviously knew what needed to be in the next version. Now, it’s a lot more blind groping around [for new features] and we still don’t feel a strong pull. There’re things [in DirectX 10] we can do with the hardware. I mean any hardware that has a capability, but we can find something useful to do with it. But it’s not worth cutting off any of our market. I just looked at the recent Steam survey numbers and [DirectX 10 card adoption is] just not very good. So I mean eventually it will just be driven by the hardware option and eventually we’ll make a point to [adopt DirectX10]. But I think it’s clear that the Doom project won’t use DX 10 class hardware. We’re going to keep that engine the same with what we got right now.

MPC – With Rage and Doom 4, if you have a PC with a high-end DirectX 10 class accelerator with a lot of memory and a big GPU and a lot of stream processors, will you actually get a different experience – a better looking experience?

JC – Not as much different as used to be, the case where when we released Doom 3 most people’s cards would be playing it at twenty-something fps and eventually you could upgrade to the top of the line system and run 60fps. It’s a little worse now because we’re designing [Rage] to run at 60fps on consoles. While you would think that the consoles are basically equivalent to a 6800 in a 512mb PC the reality is that all layers of inefficiency on the PC and the drivers mean that you really need twice the PC to get the equivalent play of the consoles. But when we have PCs that are 4 or 8 times as powerful as the latest consoles, it’ll help.

But what that means is obviously the game will still run at 60Hz, but you’ll be able to run at higher resolutions. On the PC, things stay at higher resolutions faster as you move around. You’ll be able to install everything onto the hard drive on the PC so it’ll page in better as well. We’ll probably have a couple of optional shader things to just turn on. Say, if you got the graphics power to burn, we can do better sampling on some of the textures but none of them are world beater changes. None of them are things that really make you say “wow I’m glad I paid $500 for my video card.”

It’s unfortunate because I always love being the kind of the application that people thought was important enough to go out and upgrade their hardware for. And old previous games were of a great generation where it’s this clear thing when you buy a high-end graphics card and it’s just so much more awesome. With the Doom project it’ll be a case where we’re going to be throwing a whole lot more through the same pipe so the consoles will be down to 30Hz, but the high end PC s will be up at 60hz and that’ll be a case where the PCs will have a clear advantage over then the console generation because by that time it’ll be 10x faster on the high end.

MPC- so one of the things from our last conversation that people had a lot of questions about is that the next Doom was going to be a 30Hz game. Does that mean the input is running at 30Hz as well?

JC – Yeah I would expect it’s going to be 30Hz. It’s possible we could have the playing input run at the higher tick rate than the base game and frame rate, but that’s probably not going to be the case. And essentially no games, you know, no modern games will wind up doing that type of thing. Most of the games are 30Hz although it’s great to see Call of Duty 4 being a 60hz game. It’s an interesting question. We can’t do scientific studies on this but how much does that superior feel contribute to people enjoying the game more? And certainly, we’re going to be catering that feel on Rage. It’s going to be this perfect silky smooth experience across all platforms as well as having really awesome visuals.

TW – I mean consistent framerate is also very important like John said. But with that, we still have some of the greatest artists in the world so the game will look beautiful.

JC – and that’s the great thing. That’s all stuff set up for 60Hz gameplay. But what we didn’t want was to have people look at Doom and say is Rage re-skinned? So we want to be able to take that extra step and it’s a hard fought battle to be a 60hz game and we’re going to be struggling with that all the way through to shipping we have to keep it at 60Hz when the artists add some more stuff to the game. We have to keep optimizing through that. We’re not going to fight that same battle through Doom. We’re just going to say it’s going to run at 30hz but we’ve got three times the resources to throw at it.

MPC – How much do you think the tools for artists are kind of holding back gaming in general?

JC – On the traditional modeling side we’re basically all using the movie industry tools so you can’t really say that that’s holding them back because you can build movies with the same sets of tools. I do think that the stamping stuff that we’ve gotten – that is a pretty fundamental new advancement for what we could do with gaming and it lets us bring a look to our games that you don’t see in others. Like after you’ve looked at Rage stuff for a while you start looking at other game trailers and you’ll notice they’re mostly these big flat areas of repeated textures. Those look more like a game than like the lived-in world that we’ve got in Rage.

But when you think about the core code development that runs this stuff, it’s a page of code to do the megatexture lookup. And even the management of it all is two files of code or something. But here it is three years of work later and now our challenge is all about making everything work in production and getting the production processes together. So we’re at least preparing ourselves to go through that same set of challenges as we extend the stuff to geometry hopefully in the next generation.

TW – But as far as tools go, the challenges that the mod community have faced, even in the Doom 3 generation, have gotten much more difficult.

JC – And that’s unfortunate, but it’s one of those courses of history. I’d say that the golden age of the mods passed on the PC because it used to be anybody could make something that at least resembled the commercial product, and a talented person could make something that could even stand in for a commercial product. And that’s just not the case anymore. I do have hopes that there may be other kind of platforms that it migrates to like the mobile platforms where you may have a similar kind of modability.

MPC – There are user-generated Team Fortress 2 maps that may not have the level of polish and the props and all that stuff that you guys or Valve or anyone might add, but they are fundamentally very playable maps that Valve is picking up and bringing into their game. What do you guys think about that?

JC – It’s great. [I’ve] always been a big supporter of that type of thing. We’re going to have a lot of that with Quake Live. While we will eventually support some other completely different mod game types, early on the plans are that we’re going to be advocating development of fresh new levels.

It’s really suitable on Quake Live because given the distribution method, hopefully the large number of users that are going to be playing there it gives people a big stage to play on. Lots of people will get to see the content. It’ll be trivial for people to download it and rank it. Just making the whole user-content experience isn’t something that you have to kind of know the quirky lore to know how to access.

MPC – And it’s an option if one of your friends is playing, you hit the button and join?

JC – It just goes and gets you in the game automatically.

TW- I was actually talking to a reporter last night. We’re talking about the history of modding and I thought “wow, that’s really interesting. When Doom came out John allowed the game to be modded and changed and that has affected id [as a company]. Because myself, our lead designer, our art director, and our programmer director, all came from the mod community. Modding has actually shaped what id is today based on our modding games in the beginning.

JC – And it’s the best way to do it. I remember being a teenager and sending a letter to EA saying “here’s the game that I want to create in pascal record structures” and obviously they blew me off because they had no reason to think any other way about me. But mods are just the best way to do it because it’s a way for people that are outside the industry to put something together to show why they should be in the industry and it lets the people on the other side to actually judge new talent fairly.

MPC – Would you say that id Tech 5 is a more general purpose kind of game engine? Up to this point you’ve been kind of well known for making awesome first-person shooter engines. Are you trying to make id tech 5 more general purpose?

JC – Well because we knew that we were doing outdoor stuff in Rage it shaped all the internal decisions about the engine where there’re really no optimizations for interiors like portals. Everything is done in a way that will work for the outside stuff so the indoor stuff, which is the easier case, just falls out of it. But there’re difference between id Tech 5 vs. id Tech 4.

It’s an interesting thing where I was so happy with id Tech 4 where everything became completely universal in general. What I mean is all the lighting on all the surfaces, and it seemed like we were moving towards this general purpose thing and away from all the special case hacks. But Rage forced us to do at least a 90 degree turn and say “ok, we’re leaning on the pre-generated, pre-rendered stuff for the megatexture and we’re doing a lot of the game-ism classic design type things not unifying the lighting and shadowing but dimming things down with shadowmaps and brightening things up for lightmaps rather than doing true proper lighting.

The game actually runs in 2 modes. There’s a development mode which is very similar to a Doom 3 type renderer that gets slower the more lights you have on. But then we go into the production mode with what we call combo maps where it takes everything and digests everything down, cuts it all up, analyzes everything. And at that point it’s running in this much more specialized mode which is several times faster and that’s how we have a chance to get up to 60Hz.

And it wasn’t the direction that I thought id Tech would be going toward at the end of Doom 3. The direction we had started with the [cancelled] Darkness project was still doing these more general purpose things, in some ways, adding more ambient volumes and stuff we could do in different ways. But we made a real strategic change in the kind of implementation in Rage to just say we’re not about being pure or being correct or being mathematically elegant in some way.

We’re going to do the things that make the game good – that cater to the things that we made possible [with the technology]. The big play was the megatexture stuff which is how we think that we can differentiate ourselves from all the other games out there. We’ve spent a lot of effort to go do this. I’m sure there are lots of people working on copying it right now but there’s a lot of work for them to catch up and it’s something that just so different.

There were a whole set of techniques that I looked at post Doom 3, for example things that shaped area lights, special shadowing, different ways to do specular highlights, displacement bump mapping, and a whole raft of things. And the real takeaway that I came away with was most of these are things that you have to point out to people. You have to be able to say “isn’t it cool that that highlight there is square instead of circular?” Things like that that really don’t make that much of a difference.

The difference that you want to make is somebody walking by outside your office looking in and seeing something on the screen from there that looks cooler than what they’re use to seeing. Some people are of the opinion that you could put together a thousand of these little things and make something that becomes “next gen” that looks like a new technology. While the megatexture technology would allow us to look different than what people are used to seeing and will have a qualitatively different perspective.

TW – One of the great things that it does is allow us to have true unique areas and that help enrich the story and the setting and it makes you feel like you’re more inside the world. When you walk through some of the towns and places that we’ve already made and finished you may not consciously realize “ok everything is unique” but it feels different. It really does and that really helps the story in-game.

Check back later for the second part of our exclusive interview with John Carmack, in which he gives his thoughts on Nvidia's Cuda, Intel's Larrabee, and ATI's rumored Fusion!

E3 2008: The John Carmack Interview. Rage, id Tech 6, Doom 4 Details, and More!

Will Smith — Tue, 15 Jul 2008 12:00:35 -0500

We had an opportunity to speak with id co-founder John Carmack after the big EA press conference yesterday (where id surprisingly announced a partnership with EA to publish Rage). We grilled the legendary game developer (and part-time rocket scientist) about id's post-apocalyptic shooter, the state of gaming graphics, and what his plans are after id Tech 5. Rage looks be a drastic departure from the traditional id FPS, not only in gameplay style (open worlds with vehicles vs. claustrophobic indoor environments) but also in the way Carmack has designed the code-base. id has already announced that Doom 4 is in development (no publisher has yet been annonced), and Carmack confirmed that it'll run at 30Hz and run with several times the graphics power as Rage, a 60Hz game.

MPC: Can you lead off by telling us a little bit about Rage and id Tech 5?

John Carmack: id Tech 5 is the next major technology generation after the Doom 3 [engine]. Doom 3 was targeted at high-end PCs and the original Xbox. It’s been one of our big learning experiences working with our partner companies about just how miserable the porting process is. Doing a game, then delivering for the PC, the Xbox 360, the PS3, the Mac or whatever. It’s just something horrible that we suffered through with Enemy Territory, or rather one of our partners did.

The target that we had, we knew when we were starting this generation of technology that the consoles were taking over the preeminence for the types of games we’d been making—the AAA media-rich blockbusters were on the consoles. We needed a solution that would be portable across 360, PS3, PC, and OS X, but we didn’t want to abandon our PC roots. And we have enough Apple boosters internally that the Mac remains a platform that we have some fondness for.

On a technical level, we had to make design decisions that would let us target all of these platforms, very importantly, from a single source-base, and developed [that] internally. We didn’t want to have to go out to any other companies to produce these games. We wanted to say OK, here’s the build and run the same build on the PC, the 360, and the PS3. We still have to go through one extra step to build it on the Mac. But it really does work that way on the other [console] platforms. Right now, you check the stuff in, the next day you get the build reports, which says, “here it is [ready to go] on these three platforms.” So that’s one of the core technical aspects.

MPC: So one team is developing for four platforms at the same time (PC, 360, PS3, and OS X)? That’s really cool!

JC: This is the first time we’ve done this at id, and we think we’ve done it better than many other companies. This is one source tree and one environment, you build it and it works well on all these. Different platforms have had different amounts of elbow grease on them. Certain things are easier on the 360 than the PS3, but you put the extra effort in where it’s required. You don’t want to minimum common denominator it, and just do what’s easy on the different platforms. The intersection of that is not so great.

Tim Willits, id's Lead Designer: One of the great things about the new id Tech 5 engine is that as a developer, you use the same set of assets on all of the platforms. You can immediately see what your game looks like on all the different platforms. For us, that’s a great way of looking at development of games.

JC: Our history as a PC developer lead us to a lot of development technologies that console developers generally didn’t do. Through the Doom 3 generation we had all of this live update stuff, where a modeler could save out a model or save out an image. You could just hit one key in game, and it refreshes everything in the game around you. While we went through the Doom 3 porting process to the Xbox, our designers were appalled at how painful the process was. “Ew, this takes hours to just make tiny changes there”. We’ve brought all of the goodness of PC development there to console developers.

During the development process [now], you can do the exact same thing: save this thing out, hit a key, and it magically pops up in the world around you. With a stamping on the Megatexture stuff, you’ve got even cooler demos, where you can be doing graffiti on the texture on the PC, and a guy walks around the corner on the PS3, and it’s magically there.

TW: So Rage, the other part of your question. Rage is a new franchise built on a first-person shooter foundation. But yet, it expands on that with more exploration, more action, and more adventure with some driving and vehicle combat.

JC: It’s explicitly not the corridor shooter that people have come to expect from id. It was a very conscious decision that this won’t just be the next revision of the same game that id’s made before.

MPC: It seems like you’ve taken inspiration for Rage from Mad Max and everything else that’s post apocalyptic and awesome. What would you say that you’re juiced about and pulling on to make Rage?

TW: For us, it’s all about fun and making the game experience as enjoyable as possible. During development, whenever we hit a crossroads between realism and fun, we take the fun path. We would love for game players, after they finish Rage to say, “that game was fun. I had a great time.”

JC: And that actually, there are some technical points that we tie in there. I was very proud of the Doom 3 generation, where I unified all these things: static and dynamic lighting, static and dynamic geometry. It was technically elegant and wonderful, and was this thing that I was quite proud of as a setup. We’re making completely different decisions on Rage. I’m not trying at all to be perfectly uniform or elegant in whatever way. We’re doing a lot more of the traditional gaming hacks in the technology, because we’re a 60Hz game. We’re totally blazing fast because Rage is a 60Hz game. We want to be responsive for the driving side of things, and that carries over into a silky-smooth sense of play even with all the other first-person sides of things.

That also plays into our grand strategic plan with all this generation of technology. We’re ramping up to do another Doom game, built on id Tech 5. But it’s going to be a 30Hz game. Even though we’re not changing the engine, we get to throw three times as much horsepower at it, so it’s going to look like a totally new game engine on there, even though it’s going to be built on the four years of effort that we spent developing this generation of technology.

MPC: So, you said Rage is a 60Hz game. Is it an OpenGL or DirectX game?

JC: It’s still OpenGL, although we obviously use a D3D-ish API [on the Xbox 360], and CG on the PS3. It’s interesting how little of the technology cares what API you’re using and what generation of the technology you’re on. You’ve got a small handful of files that care about what API they’re on, and millions of lines of code that are agnostic to the platform that they’re on.

John Carmack conjures up graphics magic for Rage

MPC: Are you using DirectX 9 equivalent? For Doom 4 as well?

JC: Yes to both. It’s one of those things I get asked a lot. What’s big and exciting for DirectX 10 or DirectX 11? There’s not a whole lot of… really not a whole lot. The big touted geometry shaders were in many ways, a mistaken belief that people desperately wanted to create stencil shadow volume.

There’s a tough thing with that. You get a bunch of people who make APIs, and they think “it’s my job to make APIs. I make new APIs every year.” There’s a reality of approaching a functionality curve, and the DX9 level gives us a whole lot of stuff where it’s not like before, even at the DX7 functionality level. Graphics programmers have tried every possible configuration, and they’ve tried every state and know what happens when. But, as soon as you get programmability in there (as happened with DX9) you’re writing code now. The code is limited, but we’re so far from exhausting the possibilities.

Sure, when DX10 hardware is ubiquitous and that’s our baseline platform, we’ll find something useful to do with all that extra hardware. It’s not like we’re saying “no, we won’t use this”.

MPC: Do you think we’ve reached a point of diminishing returns with regard to graphics?

JC: There’re interesting things to talk about in that direction, [for example] with Quake Live. We’re taking this ancient graphics technology, it’s nine years old, but we’re wrapping it in this other way to innovate, with the website interface for all of that. It’s clear that there are certain types of games that we’re past the curve for the benefit. For the highly competitive games, competitors would crank the detail all the way down, sometimes going too far. It’s cool that we’re running those games now at 60Hz on 2 million-pixel monitors.

There’s still value to be gained at the high end with graphics. We’ve got some wonderful looking stuff with Rage where we can do things with the environments that people have never seen before. Rage and id tech 5 will make a lot of games start to look plain. We’ve seen that phenomenon with previous games, where people don’t know exactly what they’re missing until they’re shown it, but it makes some of the other things look shabby in comparison.

I still think there’s one more generation to be had where we virtualize geometry with id Tech 6 and do some things that are truly revolutionary.

MPC: How long will we see games based on id Tech 5?

JC: Through this console generation at least, I am beginning some of the preliminary research work on what we’re going to follow this up with on the following console generation. Lots of questions are unanswered about that, depending on how all the players choose their technology—whether it ends up being a Larrabee based, CUDA hybrid, or Fusion-based, there’re lots of unanswered questions. I know we can deliver a next-gen kick, if we can virtualize the geometry like we virtualized the textures; we can do things that no one’s ever seen in games before. It’s worth doing a new generation for.

MPC: So this isn’t your last engine?

JC: No. It’s interesting though, that a couple of years ago, I’d thought that maybe we were approaching something that would be a regular tool, that the performance is a driving factor. You could write something really general, but the performance is hard, even for a 30Hz game.

In some ways, our big advances are in tools. Our innovations are beyond the offline renderers. We’re doing stuff that even the offline renderers don’t.