Doom Video game creator reflects on 20 years of GORE!
Updated by Endah
At the stroke of midnight on December 10, 1993, an executive at id
Software uploaded a file to an FTP site on the University of
Washington’s network. The filename was doom1_0.zip. Thus did one of the
great revolutions in the world of gaming begin—not with a spectacular
launch party and a multimillion-dollar ad campaign, but with a 2MB file
transfer.
From there, gamers picked up the ball; they downloaded the shareware file and immediately uploaded it to other FTP sites and local bulletin board systems. Download by download, Doom started to make its way around the world. It would become nothing short of a cultural phenomenon, popularizing the first-person shooter genre of games and shifting the predominant aesthetic of games from “Saturday morning cartoon” to “Saturday night horror double feature.”
Id’s previous games like Commander Keen and Wolfenstein 3-D were technological marvels that were starting to make personal computers a viable alternative to game consoles for fast-paced action games. At the heart of these innovative games was the technology created by id’s programming genius and co-founder John Carmack, who in the years since Doom‘s release has continued to create increasingly stunning graphic engines for id’s games. On November 22 of this year, Carmack left id to become CTO of Oculus VR, a company developing a head-mounted gaming virtual reality display.
To commemorate the 20th anniversary of his most momentous game release, WIRED spoke with Carmack about the creation of Doom, the game development world in 1993, and his thoughts on the future of the series. (Carmack’s responses have been edited for space and clarity.)
WIRED: Doom basically created a genre. There’s a chart that tracks the use of the phrase “Doom clone” to describe games in a first-person environment, and then eventually it became the genericized “first-person shooter.” It’s interesting that shooting and first-person view became linked so quickly.
John Carmack: That definition of the genre, about what people expect in it, is something that—far more than rendering technologies, or character descriptions or anything like that—defining and laying out the boundaries of the genre is something that sticks with us. I think that first-person shooter is a stable genre that’s going to be here forever, just like there are going to be driving games forever. There’s something just intrinsically rewarding about turning around a corner and shooting at something.
Now, it is interesting that there was a cycle where FPSes seemed to be in a little bit of a wane. And I thought there was a rational reason why in many cases for gaming third-person would become the dominant genre, because third-person lets you use all the tools of the film director—there’s a large body of knowledge for what you do with control of camera to influence people. I had kind of resigned myself to the fact that that likely would be the more dominant form, and it’s still using 3-D rendering, it’s the same engine technology, but rather than direct control of the camera, it’s indirect. And I have to say I was pleasantly heartened when the Call of Duty wave came over in more recent years and really took first-person back to the top of the heap in prominence.
WIRED: The early history of id Software seems to be a pattern of developing some cool new technological trick and then developing the game around that. Would you say that was the case for Doom?
Carmack: It’s true that the technology led a lot of the things that happened there, even if you go back before that to the 2-D sidescroller games and Wolfenstein. But especially in the early days of 3-D, where we went from Wolfenstein to Shadowcaster and then Doom, it really was figuring out what was barely possible and then seeing what kind of game you could do around it.
Historically, 3-D graphics were being done at the time for simulation-type games, for flight simulators and things like that. So the decision to do three-dimensional graphics but with texture mapping and make it an action game influenced a lot of how the technology was made.
It was never a matter of just building a technology in a blank room and saying, well, now, what games will we do with this? It was always driven by, well, we would like to make a game like this and this would be really cool, let’s see what technology we can do that’s in this quadrant of game development space, and then figure out what game we can make with what we actually wind up with.
WIRED: Can you describe the main differences between Wolfenstein 3-D and Doom from a technological standpoint?
Carmack: There were some intermediate steps that I did for a technology engine that I did for Origin at the time that added lighting and floors and ceilings—it was still tile-based, but the graphics side of things that people look at, the core things where it had lighting, where you could have flashing darkened areas, was no longer tile-based. Even some of the other sophisticated games that had floors and ceilings and lighting, they were still tile-based games while Doom could be built out of these sort of arbitrary line drawings that got a lot of creative potential. I also made it more dynamic, where you could have the changing lights, the rising floors and ceilings.
But with that behind us, I knew that that was a really important thing, and with Doom I took very specific technological steps with the .WAD files and the ways that allowed people to override almost everything. I didn’t get everything right on there; there were a few things that still didn’t quite work as well as they should have for modifiability, and didn’t have scripting like we brought in later with Quake, but that was one of the things that was one of the really major contributors.
WIRED: So you were behind modding from the beginning. Was there any pushback within id at that point, from other designers thinking that if you let people make their own levels, they wouldn’t buy yours?
Carmack: That was a decade-long fight inside id, really, about how open we should be with the technology and with the modifiability. The two things people were concerned about were, as you say: won’t people be able to make levels and sell them in competition to us? And there were certainly some specific cases, like the whole D-Zone game that came out with the package of a million or whatever different levels somebody could find scraped off the BBSes and put out there. And it’s arguable they may have made more money on that than we did on our early Doom retail deals, because at that point we were pretty naive. We did not have good publishing deals. We made a lot of money off of it, but it was far from an optimal business and publishing arrangement, the way it was pushed out.
We know some of those things sold really large numbers. So there was definitely an element of bitterness inside some corners of the company about that. I don’t think that they ever took anything from us; it’s not like we had a competing package. So that’s one of our fundamental arguments about, well, if somebody’s making money that maybe could have been yours, but if you didn’t do it—are you mad at them, or is it no skin off your back?
But then the other side of it was the technological evolution question, where people said, aren’t we giving away some of our secrets? When we released our source code to the builder and those different aspects. And certainly tons of people learned from that, and did go on to build things, and you know, there’s an argument to be made that the company could have perhaps held onto a lead and an edge in the market better without doing that. But I think we came out net positive.
I was really happy a decade later when Kevin Cloud, one of my partners, said that I had been right to be pushing for doing that. Because he had been looking at it not so much from the community and technological openness standpoint, but as a business risk. Coolly looked back at over the years, I think we benefited more than it might have hurt us. But in truth, I was just doing that at the time because it was something that felt really right to me.
I still remember, at the time I was commenting about how I remembered being a teenager sector-editing Ultima II on my Apple II, to go ahead and hack things in to turn trees into chests or modify my gold or whatever, and I loved that. The ability to go several steps further and release actual source code, make it easy to modify things, to let future generations get what I wished I had had a decade earlier—I think that’s been a really good thing. To this day I run into people all the time that say, whether it was Doom, or maybe even more so Quake later on, that that openness and that ability to get into the guts of things was what got them into the industry or into technology. A lot of people who are really significant people in significant places still have good things to say about that.
WIRED: Can you name anybody specifically?
C
armack: [laughs] There’s something that just came up a week ago that would be lovely for this story but it’s actually somebody that I can’t mention. There’s a couple of things about that that are Oculus-related. It’s funny, I almost sort of dodged this whole 20th anniversary of Doom thing because I’ve been heads-down working on stuff, I really don’t think much about this prior stuff. And it’s sort of a dangerous time for me right now, because I have all this stuff that’s secret, related to Oculus, and there’s this risk that somebody just gets me blabbing about things, and something will slip out that’s not supposed to.
WIRED: We talked about Doom being shareware. Did you have a sense that that model was going to be coming to an end?
Carmack: I don’t know if it was coming to an end so much as it was probably not optimal for a product that could generate that much attention. We knew that with Doom we were at a point that very few games had ever been before, where it was getting mainstream attention and it was something that really didn’t need to be sold, so much, because it almost sold at a glance. People looked at that and it was something they’d never seen before, and they put hands on it and very quickly, it didn’t take much to make a sale for Doom.
Shareware had been great for things that people really weren’t sure about. It was essentially a demo, it was a try before you buy sort of thing, and it was couched in the terms of, one of them’s free and you can buy more, but it was essentially a very prolonged demo—one-third of a game was the demo. And that was Scott Miller and Apogee’s innovation over the classic shareware model where you give them everything and they pay you if they feel like it. Which has never been terribly successful for games. To turn it into “give them one-third of it and they pay you for the rest of the trilogy,” a lot of that was sort of made-up words to cover what was essentially a demo arrangement.
And it worked, obviously, very well for our team. For Wolfenstein and for Doom itself. But it probably wasn’t optimal once you’ve already got that foothold there. It’s pretty clear to me that we had crummy publishing relationships at that time. If we had been in an even better position, I’m sure it would have been a couple times larger.
WIRED: When people think of you specifically, they think of someone creating all of the technology for the games and handing it off to designers — whether this is fair or unfair. What kind of a hand did you have in the gameplay design of Doom?
Carmack: The gameplay, in terms of how weapons work, how world interactive items work, all of that, how the AI works, that was all still largely my code at that time. You can trace the lineage of it back to Wolfenstein 3D, which was based on this 2-D game that I had written called Catacombs, which became Catacombs 3D. Those were things that I had written from scratch. But the personality of the game, in terms of dialing in exactly how much damage things do, tuning it, changing speed, [Doom co-creator John] Romero had a big part doing that. A lot of his fingerprints were on the core tuning elements of it. But I wrote the core feedback loop about what happens there. I was not the one making the decisions of “this should do 4 instead of 5″ and “this should be worth 20 instead of 10.” The designers were there.
On the thematic side of things, I pushed certainly for the demonic aspect of it. That’s still something that I feel good about, looking back: In later games and later times, when games get attacked with some of the moral ambiguity or actual negativity about what you’re doing, I always felt good about the decision that in Doom, you’re fighting demons. There’s no gray area here, it is black and white, you’re the good guys, they’re the bad guys and everything that you’re doing to them is fully deserved.
There was the little bit of undercurrent, especially when I was younger and I was a bit more aggressive. I think at one point, I said it was fun offending the easily offended. Poking at the fundamentalists was at least a sub-current of picking the demons, and the pentagrams, and the goat’s heads and all the things in there. So I was certainly supportive in pushing that but obviously all the actual instantiations of them were done by the artists, Kevin and Adrian.
WIRED: Were you, then, surprised when it became as controversial as it did?
Carmack: Not really. I think that it was interesting being the poster child for Congress for a decade; to get up and wave a game around, it was always Mortal Kombat and Doom until Grand Theft Auto came around and made us look pretty tame in comparison for moral outrage purposes. I understand why people do that, I understand why it happens. They’re looking for a scapegoat or even a potential cause. It seems to me just part of human nature that somebody is going to take the brunt of that. And in the end it doesn’t really hurt. It’s not like there were federal laws that came out that banned any of that. It had little actual impact on us. In many ways, it’s the theater of politics.
WIRED: I remember having to wait for our family to get a 386 so we could play Doom, and Wolfenstein was the game we had been excited to play when we upgraded to a 286 years prior. Were you worried about people being able to handle these games? Because a lot of PCs in homes couldn’t handle the games.
Carmack: Wolfenstein was a 286 game that used extended memory, the old EMS/XMS different ways that we were extending past there. And one of the important decisions for Doom was, could we make it a 386-only game? And it was probably the first significant game title to use a 386 DOS extender and be written in 32-bit protected mode. That was something where I could have written the same thing still within the context of 286, and maybe a 286 person could have played it in a small window chugging along at 10 frames per second.
But that decision said, we think that it’s worth the transition over here, where the people with the better systems will get a noticeably better experience at the expense of cutting off some of the people on the low end. And this is something that to this day I still struggle a little bit with, leaving the low-ends behind. But we had just come off of some decisions in the prior games, where we made a CGA graphics version of Commander Keen: Aliens Ate My Babysitter. It was a reasonable chunk of work, it was much worse with only four colors, and it hardly got played at all.
You look at some surveys and it’s like, oh my gosh, there’s this immense part of the market that has these lower-end cards, but when we looked at people that were actually playing the games, it wasn’t significant. We made the decision on Wolfenstein that we weren’t going to support EGA graphics, it was going to be all 256 color graphics. That was stressful. And with Doom it was, well, it’s going to be 386 required. With Quake it was floating point required, which was probably the more problematic of the steps that we made. Then we made Quake 3 GPU required.
Even to this day, I struggle a little bit with that; there’s so much you can still do on the previous console generation. The 360 and PS3 are far from tapped out in terms of what a developer could do with them, but the whole world’s gonna move over towards next-gen and high-end PCs and all these other things. Part of me still frets a little bit about that, where just as you fully understand a previous generation, you have to put it away to kind of surf forward on the tidal wave of technology that’s always moving. That’s something that we’ve struggled with in every generation. And now I at least know enough to recognize that some of my internal feelings or fondness for technology that I understand or have done various things with usually has to be put aside. Because data has shown over the decades that that’s usually not as important as you think it is.
Although I keep making new arguments where now we can say that we’re past the knee of the cost-benefit curve in terms of what we get with graphics, and people are saying that with the next-gen consoles, that okay, they look better but they don’t look nearly as much better as the previous generation. So does that mean people will stay happier with the current things? And I could make that argument with a straight face and play for it, but it’s probably going to be wrong.
WIRED: When you did upgrade your computer to be able to play Wolfenstein or Doom, you were getting something that seemed almost magical, like it was somehow more powerful than your computer could really handle. Because neither of those games were 3-D games, right? They were faking people into thinking they were playing in 3-D worlds but it was really clever design on your part to do that.
Carmack: Yeah, like when you take a spiral staircase going up halfway around something and you don’t realize that they stopped after a 90-degree turn because it can’t go all the way around on top of you. And you forgive that you can’t look up and down. Those days when you were really making those fundamental systems tradeoffs, where you say, if we give up this, then it gives us that—we don’t have that so much in games today because you can do a pretty damn good job at just about anything you want to throw at a game today. And we have the hardware acceleration that automates but also channels what you choose to do. Because if you don’t choose to do what the hardware is happy with, you’ve got this order of magnitude penalty that’s almost impossible to overcome.
In many ways, that was the most exciting age of the decision-making process, where you’re deciding what you choose to undertake based on it. Because once GPUs took over, it took away a lot of the freedom. There’s still an infinitude of things you can do there, but the tasks you choose to undertake are a little bit more constrained. Because there’s one obviously correct way of engaging the hardware. Back in those days, up through Quake, we had people pursuing voxel engines, things that would render with all sorts of crazy deformed sprites, quad-deformed line of constancy rendering versus all the different things that we did, and it was interesting to look at how varied things were there, versus today where everything’s just vertex or fragment shaders.
In the early day, there existed systems on the high end that did exactly what everybody always wanted, where you had your image generators from Evans and Sutherland and Silicon Graphics that did things right. We would have loved to be able to have things like that, and program towards them and build things on it, but it just wasn’t possible and it wouldn’t have been possible for another five, six years or something before GPUs could get there. And GPUs might never have gotten there in the same way if we didn’t have these earlier steps.
But it was interesting where there was some pushback from—you had some of the people in the simulation business that would look at and almost be insulted by things like Doom. Because they could see that it clearly wasn’t doing everything right. There was no filtering on the pixels, the colors were banding. And they could tell you couldn’t roll your head, and you can’t do all these different things. People were a little bit put out about it because they knew all the tricks that I was pulling.
That’s one of the most important decisions: You can sacrifice something to make a net value improvement. We can make something that would not be possible any other way, and add in value for millions and millions of people here by sacrificing some degree of rightness. Putting away something that’s great to have, but deciding that it’s not as important as what we can get if we are willing to do without it.
WIRED: When you’re talking about what it was like, your recall doesn’t seem like it’s at all tinged with nostalgia.
Carmack: I’ve said before that I’m a remarkably unsentimental person. I really don’t spend much time thinking about it; only in times like this, when anniversaries and things come around. The only sort of non-prompted time when I was sort of going over this was a few years back when I did the iOS version of Doom Classic, when I said, okay, I’m going to give myself six weeks or something and I’m going to go crank this out. And that was actually a lot of fun and brought back a lot of memories where I’m just going through my old code, looking through all this stuff, playing through the game, beta testing it myself, thinking that yeah, this actually still holds up pretty good after all these years.
WIRED: Doom is famous for having been ported to every possible platform, almost. Do you want to port Doom to Oculus?
Carmack: I am greatly proud of the fact that Doom is one of those things where everything that has a 32-bit processor has had Doom run on it, and I think that’s been one of the great aspects of having it be open source, having everything out there means that people have maintained that and kept it up to date. And I got to have one of those convincing arguments when I did go do the iOS version: Rather than starting with our old code that had been released 15 years ago at that time, I went to one of the maintained open-source versions and they cleaned it up, fixed the bugs that I had left in there, and I was able to start with that, update all the iOS stuff, and release all of my changes back as open source. It was really great.
As far as Oculus specifically: If I had free time, which I absolutely do not, I wish that I could bring the Doom 3 stuff over. Sometime, years from now, all of the gory details of everything from the last year and a half or so will come out, but now is not the time to be talking about all of it. But I did feel really bad about the fact that I had pseudo-promised Doom 3 for the Rift when I was first talking about it, and now the fact that it didn’t get released, I felt personally uncomfortable with how that turned out. I wish I could be doing that, but I do not have the time now. I’m busy working on a bunch of exciting things that I can’t talk about and wish I could.
WIRED: Another question you probably can’t say much about but that everyone reading an article about Doom will want to hear—what’s up with Doom 4?
Carmack: That’s something I can’t really go into much in detail. It’s been hard—one of the things that was a little bit surprising that you might not think so from the outside, but deciding exactly what the essence of Doom is, with this 20-year history, is a heck of a lot harder than you might think. You get multiple Doom fans that have different views of what the core essence of it is, and there’s been a design challenge through all of it.
The worst aspect of the continuing pace of game development that we fell into was the longer and longer times between releases. If I could go back in time and change one thing along the trajectory of id Software, it would be, do more things more often. And that was id’s mantra for so long: “It’ll be done when it’s done.” And I recant from that. I no longer think that is the appropriate way to build games. I mean, time matters, and as years go by—if it’s done when it’s done and you’re talking a month or two, fine. But if it’s a year or two, you need to be making a different game.
Updated by Endah
Q&A: Doom’s Creator Looks Back on 20 Years of Demonic Mayhem
- 12.10.13
- 6:30 AM
John Carmack speaks at QuakeCon 2009. Photo: Drew Campbell/QuakeCon/CC BY 2.0
From there, gamers picked up the ball; they downloaded the shareware file and immediately uploaded it to other FTP sites and local bulletin board systems. Download by download, Doom started to make its way around the world. It would become nothing short of a cultural phenomenon, popularizing the first-person shooter genre of games and shifting the predominant aesthetic of games from “Saturday morning cartoon” to “Saturday night horror double feature.”
Id’s previous games like Commander Keen and Wolfenstein 3-D were technological marvels that were starting to make personal computers a viable alternative to game consoles for fast-paced action games. At the heart of these innovative games was the technology created by id’s programming genius and co-founder John Carmack, who in the years since Doom‘s release has continued to create increasingly stunning graphic engines for id’s games. On November 22 of this year, Carmack left id to become CTO of Oculus VR, a company developing a head-mounted gaming virtual reality display.
To commemorate the 20th anniversary of his most momentous game release, WIRED spoke with Carmack about the creation of Doom, the game development world in 1993, and his thoughts on the future of the series. (Carmack’s responses have been edited for space and clarity.)
WIRED: Doom basically created a genre. There’s a chart that tracks the use of the phrase “Doom clone” to describe games in a first-person environment, and then eventually it became the genericized “first-person shooter.” It’s interesting that shooting and first-person view became linked so quickly.
John Carmack: That definition of the genre, about what people expect in it, is something that—far more than rendering technologies, or character descriptions or anything like that—defining and laying out the boundaries of the genre is something that sticks with us. I think that first-person shooter is a stable genre that’s going to be here forever, just like there are going to be driving games forever. There’s something just intrinsically rewarding about turning around a corner and shooting at something.
Now, it is interesting that there was a cycle where FPSes seemed to be in a little bit of a wane. And I thought there was a rational reason why in many cases for gaming third-person would become the dominant genre, because third-person lets you use all the tools of the film director—there’s a large body of knowledge for what you do with control of camera to influence people. I had kind of resigned myself to the fact that that likely would be the more dominant form, and it’s still using 3-D rendering, it’s the same engine technology, but rather than direct control of the camera, it’s indirect. And I have to say I was pleasantly heartened when the Call of Duty wave came over in more recent years and really took first-person back to the top of the heap in prominence.
WIRED: The early history of id Software seems to be a pattern of developing some cool new technological trick and then developing the game around that. Would you say that was the case for Doom?
Carmack: It’s true that the technology led a lot of the things that happened there, even if you go back before that to the 2-D sidescroller games and Wolfenstein. But especially in the early days of 3-D, where we went from Wolfenstein to Shadowcaster and then Doom, it really was figuring out what was barely possible and then seeing what kind of game you could do around it.
Historically, 3-D graphics were being done at the time for simulation-type games, for flight simulators and things like that. So the decision to do three-dimensional graphics but with texture mapping and make it an action game influenced a lot of how the technology was made.
It was never a matter of just building a technology in a blank room and saying, well, now, what games will we do with this? It was always driven by, well, we would like to make a game like this and this would be really cool, let’s see what technology we can do that’s in this quadrant of game development space, and then figure out what game we can make with what we actually wind up with.
WIRED: Can you describe the main differences between Wolfenstein 3-D and Doom from a technological standpoint?
Carmack: There were some intermediate steps that I did for a technology engine that I did for Origin at the time that added lighting and floors and ceilings—it was still tile-based, but the graphics side of things that people look at, the core things where it had lighting, where you could have flashing darkened areas, was no longer tile-based. Even some of the other sophisticated games that had floors and ceilings and lighting, they were still tile-based games while Doom could be built out of these sort of arbitrary line drawings that got a lot of creative potential. I also made it more dynamic, where you could have the changing lights, the rising floors and ceilings.
That was a decade-long fight inside id, about how open we should be.
But so many other things don’t stand out so much because they weren’t
graphics-related. Obviously the multiplayer was a huge aspect of it,
where you had deathmatch for the first time, and cooperative play, just
being able to play with your friends. But one of the other things that
was not quite so obvious were the very explicit steps I took to make the
game moddable. We saw with Wolfenstein people basically
figuring everything out for themselves, and making map packs and
overwriting sprites and things, and that was all not intended by us.
That was an emergent property of the game.But with that behind us, I knew that that was a really important thing, and with Doom I took very specific technological steps with the .WAD files and the ways that allowed people to override almost everything. I didn’t get everything right on there; there were a few things that still didn’t quite work as well as they should have for modifiability, and didn’t have scripting like we brought in later with Quake, but that was one of the things that was one of the really major contributors.
WIRED: So you were behind modding from the beginning. Was there any pushback within id at that point, from other designers thinking that if you let people make their own levels, they wouldn’t buy yours?
Carmack: That was a decade-long fight inside id, really, about how open we should be with the technology and with the modifiability. The two things people were concerned about were, as you say: won’t people be able to make levels and sell them in competition to us? And there were certainly some specific cases, like the whole D-Zone game that came out with the package of a million or whatever different levels somebody could find scraped off the BBSes and put out there. And it’s arguable they may have made more money on that than we did on our early Doom retail deals, because at that point we were pretty naive. We did not have good publishing deals. We made a lot of money off of it, but it was far from an optimal business and publishing arrangement, the way it was pushed out.
We know some of those things sold really large numbers. So there was definitely an element of bitterness inside some corners of the company about that. I don’t think that they ever took anything from us; it’s not like we had a competing package. So that’s one of our fundamental arguments about, well, if somebody’s making money that maybe could have been yours, but if you didn’t do it—are you mad at them, or is it no skin off your back?
But then the other side of it was the technological evolution question, where people said, aren’t we giving away some of our secrets? When we released our source code to the builder and those different aspects. And certainly tons of people learned from that, and did go on to build things, and you know, there’s an argument to be made that the company could have perhaps held onto a lead and an edge in the market better without doing that. But I think we came out net positive.
I was really happy a decade later when Kevin Cloud, one of my partners, said that I had been right to be pushing for doing that. Because he had been looking at it not so much from the community and technological openness standpoint, but as a business risk. Coolly looked back at over the years, I think we benefited more than it might have hurt us. But in truth, I was just doing that at the time because it was something that felt really right to me.
I still remember, at the time I was commenting about how I remembered being a teenager sector-editing Ultima II on my Apple II, to go ahead and hack things in to turn trees into chests or modify my gold or whatever, and I loved that. The ability to go several steps further and release actual source code, make it easy to modify things, to let future generations get what I wished I had had a decade earlier—I think that’s been a really good thing. To this day I run into people all the time that say, whether it was Doom, or maybe even more so Quake later on, that that openness and that ability to get into the guts of things was what got them into the industry or into technology. A lot of people who are really significant people in significant places still have good things to say about that.
WIRED: Can you name anybody specifically?
C
armack: [laughs] There’s something that just came up a week ago that would be lovely for this story but it’s actually somebody that I can’t mention. There’s a couple of things about that that are Oculus-related. It’s funny, I almost sort of dodged this whole 20th anniversary of Doom thing because I’ve been heads-down working on stuff, I really don’t think much about this prior stuff. And it’s sort of a dangerous time for me right now, because I have all this stuff that’s secret, related to Oculus, and there’s this risk that somebody just gets me blabbing about things, and something will slip out that’s not supposed to.
Doom introduced the idea of ‘deathmatch,’ online battles between first-person shooter opponents. Image courtesy Bethesda
WIRED: We talked about Doom being shareware. Did you have a sense that that model was going to be coming to an end?
Carmack: I don’t know if it was coming to an end so much as it was probably not optimal for a product that could generate that much attention. We knew that with Doom we were at a point that very few games had ever been before, where it was getting mainstream attention and it was something that really didn’t need to be sold, so much, because it almost sold at a glance. People looked at that and it was something they’d never seen before, and they put hands on it and very quickly, it didn’t take much to make a sale for Doom.
Shareware had been great for things that people really weren’t sure about. It was essentially a demo, it was a try before you buy sort of thing, and it was couched in the terms of, one of them’s free and you can buy more, but it was essentially a very prolonged demo—one-third of a game was the demo. And that was Scott Miller and Apogee’s innovation over the classic shareware model where you give them everything and they pay you if they feel like it. Which has never been terribly successful for games. To turn it into “give them one-third of it and they pay you for the rest of the trilogy,” a lot of that was sort of made-up words to cover what was essentially a demo arrangement.
And it worked, obviously, very well for our team. For Wolfenstein and for Doom itself. But it probably wasn’t optimal once you’ve already got that foothold there. It’s pretty clear to me that we had crummy publishing relationships at that time. If we had been in an even better position, I’m sure it would have been a couple times larger.
WIRED: When people think of you specifically, they think of someone creating all of the technology for the games and handing it off to designers — whether this is fair or unfair. What kind of a hand did you have in the gameplay design of Doom?
I think at one point, I said it was fun offending the easily offended.
Carmack: The gameplay, in terms of how weapons work, how world interactive items work, all of that, how the AI works, that was all still largely my code at that time. You can trace the lineage of it back to Wolfenstein 3D, which was based on this 2-D game that I had written called Catacombs, which became Catacombs 3D. Those were things that I had written from scratch. But the personality of the game, in terms of dialing in exactly how much damage things do, tuning it, changing speed, [Doom co-creator John] Romero had a big part doing that. A lot of his fingerprints were on the core tuning elements of it. But I wrote the core feedback loop about what happens there. I was not the one making the decisions of “this should do 4 instead of 5″ and “this should be worth 20 instead of 10.” The designers were there.
On the thematic side of things, I pushed certainly for the demonic aspect of it. That’s still something that I feel good about, looking back: In later games and later times, when games get attacked with some of the moral ambiguity or actual negativity about what you’re doing, I always felt good about the decision that in Doom, you’re fighting demons. There’s no gray area here, it is black and white, you’re the good guys, they’re the bad guys and everything that you’re doing to them is fully deserved.
There was the little bit of undercurrent, especially when I was younger and I was a bit more aggressive. I think at one point, I said it was fun offending the easily offended. Poking at the fundamentalists was at least a sub-current of picking the demons, and the pentagrams, and the goat’s heads and all the things in there. So I was certainly supportive in pushing that but obviously all the actual instantiations of them were done by the artists, Kevin and Adrian.
WIRED: Were you, then, surprised when it became as controversial as it did?
Carmack: Not really. I think that it was interesting being the poster child for Congress for a decade; to get up and wave a game around, it was always Mortal Kombat and Doom until Grand Theft Auto came around and made us look pretty tame in comparison for moral outrage purposes. I understand why people do that, I understand why it happens. They’re looking for a scapegoat or even a potential cause. It seems to me just part of human nature that somebody is going to take the brunt of that. And in the end it doesn’t really hurt. It’s not like there were federal laws that came out that banned any of that. It had little actual impact on us. In many ways, it’s the theater of politics.
WIRED: I remember having to wait for our family to get a 386 so we could play Doom, and Wolfenstein was the game we had been excited to play when we upgraded to a 286 years prior. Were you worried about people being able to handle these games? Because a lot of PCs in homes couldn’t handle the games.
Carmack: Wolfenstein was a 286 game that used extended memory, the old EMS/XMS different ways that we were extending past there. And one of the important decisions for Doom was, could we make it a 386-only game? And it was probably the first significant game title to use a 386 DOS extender and be written in 32-bit protected mode. That was something where I could have written the same thing still within the context of 286, and maybe a 286 person could have played it in a small window chugging along at 10 frames per second.
But that decision said, we think that it’s worth the transition over here, where the people with the better systems will get a noticeably better experience at the expense of cutting off some of the people on the low end. And this is something that to this day I still struggle a little bit with, leaving the low-ends behind. But we had just come off of some decisions in the prior games, where we made a CGA graphics version of Commander Keen: Aliens Ate My Babysitter. It was a reasonable chunk of work, it was much worse with only four colors, and it hardly got played at all.
You look at some surveys and it’s like, oh my gosh, there’s this immense part of the market that has these lower-end cards, but when we looked at people that were actually playing the games, it wasn’t significant. We made the decision on Wolfenstein that we weren’t going to support EGA graphics, it was going to be all 256 color graphics. That was stressful. And with Doom it was, well, it’s going to be 386 required. With Quake it was floating point required, which was probably the more problematic of the steps that we made. Then we made Quake 3 GPU required.
Even to this day, I struggle a little bit with that; there’s so much you can still do on the previous console generation. The 360 and PS3 are far from tapped out in terms of what a developer could do with them, but the whole world’s gonna move over towards next-gen and high-end PCs and all these other things. Part of me still frets a little bit about that, where just as you fully understand a previous generation, you have to put it away to kind of surf forward on the tidal wave of technology that’s always moving. That’s something that we’ve struggled with in every generation. And now I at least know enough to recognize that some of my internal feelings or fondness for technology that I understand or have done various things with usually has to be put aside. Because data has shown over the decades that that’s usually not as important as you think it is.
Although I keep making new arguments where now we can say that we’re past the knee of the cost-benefit curve in terms of what we get with graphics, and people are saying that with the next-gen consoles, that okay, they look better but they don’t look nearly as much better as the previous generation. So does that mean people will stay happier with the current things? And I could make that argument with a straight face and play for it, but it’s probably going to be wrong.
They knew all the tricks that I was pulling.
WIRED: When you did upgrade your computer to be able to play Wolfenstein or Doom, you were getting something that seemed almost magical, like it was somehow more powerful than your computer could really handle. Because neither of those games were 3-D games, right? They were faking people into thinking they were playing in 3-D worlds but it was really clever design on your part to do that.
Carmack: Yeah, like when you take a spiral staircase going up halfway around something and you don’t realize that they stopped after a 90-degree turn because it can’t go all the way around on top of you. And you forgive that you can’t look up and down. Those days when you were really making those fundamental systems tradeoffs, where you say, if we give up this, then it gives us that—we don’t have that so much in games today because you can do a pretty damn good job at just about anything you want to throw at a game today. And we have the hardware acceleration that automates but also channels what you choose to do. Because if you don’t choose to do what the hardware is happy with, you’ve got this order of magnitude penalty that’s almost impossible to overcome.
In many ways, that was the most exciting age of the decision-making process, where you’re deciding what you choose to undertake based on it. Because once GPUs took over, it took away a lot of the freedom. There’s still an infinitude of things you can do there, but the tasks you choose to undertake are a little bit more constrained. Because there’s one obviously correct way of engaging the hardware. Back in those days, up through Quake, we had people pursuing voxel engines, things that would render with all sorts of crazy deformed sprites, quad-deformed line of constancy rendering versus all the different things that we did, and it was interesting to look at how varied things were there, versus today where everything’s just vertex or fragment shaders.
In the early day, there existed systems on the high end that did exactly what everybody always wanted, where you had your image generators from Evans and Sutherland and Silicon Graphics that did things right. We would have loved to be able to have things like that, and program towards them and build things on it, but it just wasn’t possible and it wouldn’t have been possible for another five, six years or something before GPUs could get there. And GPUs might never have gotten there in the same way if we didn’t have these earlier steps.
But it was interesting where there was some pushback from—you had some of the people in the simulation business that would look at and almost be insulted by things like Doom. Because they could see that it clearly wasn’t doing everything right. There was no filtering on the pixels, the colors were banding. And they could tell you couldn’t roll your head, and you can’t do all these different things. People were a little bit put out about it because they knew all the tricks that I was pulling.
That’s one of the most important decisions: You can sacrifice something to make a net value improvement. We can make something that would not be possible any other way, and add in value for millions and millions of people here by sacrificing some degree of rightness. Putting away something that’s great to have, but deciding that it’s not as important as what we can get if we are willing to do without it.
Carmack says he pushed hard to have the enemies in Doom be demons rather than humans, so there would be no “gray area” about the game’s morals. Image courtesy Bethesda
WIRED: When you’re talking about what it was like, your recall doesn’t seem like it’s at all tinged with nostalgia.
Carmack: I’ve said before that I’m a remarkably unsentimental person. I really don’t spend much time thinking about it; only in times like this, when anniversaries and things come around. The only sort of non-prompted time when I was sort of going over this was a few years back when I did the iOS version of Doom Classic, when I said, okay, I’m going to give myself six weeks or something and I’m going to go crank this out. And that was actually a lot of fun and brought back a lot of memories where I’m just going through my old code, looking through all this stuff, playing through the game, beta testing it myself, thinking that yeah, this actually still holds up pretty good after all these years.
Id’s mantra was, ‘It’ll be done when it’s done.’ I
recant from that. I no longer think that is the appropriate way to
build games.
And I spruced it up with the nice filtering and the 32-bit color and
all that running on iOS, and I look at that and I think, this is what
people in their minds remember Doom to be like. Not if you
actually fire up DOSBox and run the 320 by 240 resolution game, with the
bathroom-tile-sized pixels and ugly color banding. But when you play it
on something like one of the modern OpenGL versions or on iOS, you can
say, yeah, I remember that imp, I remember this hallway, I remember
pumping the shotgun here—this was how the game was and it was pretty
cool. Not realizing that it’s using like an order of magnitude more
graphics processing power now. WIRED: Doom is famous for having been ported to every possible platform, almost. Do you want to port Doom to Oculus?
Carmack: I am greatly proud of the fact that Doom is one of those things where everything that has a 32-bit processor has had Doom run on it, and I think that’s been one of the great aspects of having it be open source, having everything out there means that people have maintained that and kept it up to date. And I got to have one of those convincing arguments when I did go do the iOS version: Rather than starting with our old code that had been released 15 years ago at that time, I went to one of the maintained open-source versions and they cleaned it up, fixed the bugs that I had left in there, and I was able to start with that, update all the iOS stuff, and release all of my changes back as open source. It was really great.
As far as Oculus specifically: If I had free time, which I absolutely do not, I wish that I could bring the Doom 3 stuff over. Sometime, years from now, all of the gory details of everything from the last year and a half or so will come out, but now is not the time to be talking about all of it. But I did feel really bad about the fact that I had pseudo-promised Doom 3 for the Rift when I was first talking about it, and now the fact that it didn’t get released, I felt personally uncomfortable with how that turned out. I wish I could be doing that, but I do not have the time now. I’m busy working on a bunch of exciting things that I can’t talk about and wish I could.
WIRED: Another question you probably can’t say much about but that everyone reading an article about Doom will want to hear—what’s up with Doom 4?
Carmack: That’s something I can’t really go into much in detail. It’s been hard—one of the things that was a little bit surprising that you might not think so from the outside, but deciding exactly what the essence of Doom is, with this 20-year history, is a heck of a lot harder than you might think. You get multiple Doom fans that have different views of what the core essence of it is, and there’s been a design challenge through all of it.
The worst aspect of the continuing pace of game development that we fell into was the longer and longer times between releases. If I could go back in time and change one thing along the trajectory of id Software, it would be, do more things more often. And that was id’s mantra for so long: “It’ll be done when it’s done.” And I recant from that. I no longer think that is the appropriate way to build games. I mean, time matters, and as years go by—if it’s done when it’s done and you’re talking a month or two, fine. But if it’s a year or two, you need to be making a different game.
Comments