Over the Christmas break, while the rest of us headed home to visit our families, Carmack stayed in Wisconsin. He was expecting a package, cash-on-delivery: an impressive workstation, the NeXTstation by Steve Jobs’s new company, NeXT Computer. The machine used the incredible NeXTSTEP operating system, which was light-years ahead of MS-DOS. It was the future. Carmack ordered one and needed to get an $11,000 cashier’s check for the delivery guy, but with his MG left for dead in Shreveport and all of us out of town, he trudged to the bank on foot, cursing the snow, ice, and freezing temperatures the whole way.
When we got back from vacation, we saw the NeXTstation in action. It was incredible, mind-boggling even. John told us about the research he had completed, running tests on vector quantization compression of VGA screens to see what kind of gains he could get over JPEG. I happened to have a copy of King’s Quest V by Sierra Online, which I’d picked up over Christmas, and the screens provided some good compression benchmarking data for him. Carmack and I discussed its technical prowess and what it might mean for our fledgling company. For now, though, it was just research, and the NeXT was a tool and an expensive toy. It was time for us to get back to work.
Recharged, we started on the next Commander Keen trilogy: The Universe Is Toast. We decided to test out a couple of new features to make Keen even closer to a console game. We added parallaxing layers to give the illusion of depth. With parallaxing, we draw multiple layers of background and foreground art and move them at varying speeds so that the entire scene appears to be 3D. We also decided to go with 256-color VGA graphics instead of Keen’s current sixteen-color EGA. These features were expensive memory-wise and speed-wise, however. VGA data was double the size of EGA data, and parallaxing layers drew over layers below, which could slow down the frame rate. To do some speed benchmarking, Adrian took some of Keen 4’s EGA graphics assets, turned them into VGA graphics and Carmack made a demo. It both looked good and ran fast enough. With more optimization, it could run even faster, but the initial results gave us the confidence to start developing.
Tom put together the demo map for Carmack to use. Luckily, TED5 already had the ability to edit a map’s background layer and foreground layer. It needed more than just that to get parallaxing working, though. I started updating the editor, but for now, background and foreground was all we needed.
With that, we had our quick plan, and we began work on Keen 7’s tech demo. Quick plans were normal for us. We jumped into everything headfirst. As a streamlined team, our workflow was efficient. We charged on for a week, but somehow, instead of picking up steam, we seemed to be running out of gas. It was 11 p.m. on Friday night. We should have been full of excitement after a week’s work on a new game.
“I am just not getting excited about making another Keen trilogy,” I said to the room.
“Same here,” Adrian added.
After seven episodes, we had just had enough of Commander Keen, shiny VGA parallax scrolling or not.
“The parallaxing looks great, but I just can’t get excited about another side-scroller,” I said. I knew that 3D was the way to go, and I felt like with each new 3D game, we were getting further ahead. Keen felt like a step backward. “Catacomb 3D was cool. I think we should be making 3D games instead.”
Carmack swung around in his chair, nodding in agreement.
Tom was left as the lone Keen holdout. To him, Keen was more than just a character in a video game. Tom was Keen. With Keen 7, we started a new trilogy, and if we finished this game and two additional games, Keen entered a pantheon of rare creative efforts, a trilogy of trilogies! Plus, there were the two weeks we’d invested in creating the demo.
Looking again at Keen and our new demo on the screen, none of us saw the future. Parallaxing already existed on consoles, and Keen existed in seven other games. Besides, after doing so much Keen in 1991, we needed to try something new. It was time to make a big decision again—what was our next game going to be? Beyond that, there was another question looming, a bigger one: What were we going to be as a company?
The previous year was a challenging and exciting year for id Software and for the four of us as individuals. We formed our own company, shipped thirteen games, signed our first international distribution deal, and moved from a bayou to a blizzard.
We were excited about the potential the new year held, particularly working on a brand-new game. We had all the work we wanted, and every month $50,000 was going straight into our bank account. Our success was staggering to us, but it was not something we dwelled on. Making the next game was more interesting to us than whatever we might do with the money. That said, financial security meant we could take a step back and design a better way of working.
Throughout 1991, we had been beholden to the calendar, slicing our time this way and that to get all the games done and delivered. We were focused on completing games to our standard of quality, and each provided an innovation in some way. What we had done in that year was impressive, particularly when I compared our output against any other small group of developers. However, we fought the clock the whole way and exhausted ourselves. Money was great, but money wasn’t all it took to make a great game. For that, we needed time. So in designing the next phase of id Software, we decided that our games were done when they were done. We put the focus on quality, on making something great, instead of making something as good as we could in the time allowed. It was a pivotal change that allowed us to focus on both tech and design innovations to create games we really wanted to play. We were in a fortunate position—as hard-core gamers, we knew great gameplay.* As game developers, we knew how to make it happen.
We kept things in place that were working for us. Carmack and I liked how we had divided the code responsibilities. Having Carmack focused exclusively on developing the game’s engine and AI gave us a technical edge. As a game designer and a programmer, I focused on all the design-side code, programming the game-play that we separated out from the core engine and AI code. Also on my plate were level design, music and sound effect integration, intermissions, and menus, as well as the production-side details such as the creation of the final master disks for distribution. Tom focused exclusively on creative direction, game design, sound effect creation, and level design, while Adrian, of course, handled all art.
The question remained: If not Keen 7, then what? We agreed that 3D was the future.
We brainstormed to come up with a new game that improved on Catacomb 3D’s engine. A few ideas were kicked around, including Tom’s suggestion for a game about a lethal lab with mutants running amok called It’s Green and Pissed. It was a funny title, but as I told him, it was the oldest idea this side of Frankenstein.
I started thinking along retro lines, too, but my idea was to make a totally new game out of a classic video game.
“Why not just make Wolfenstein in 3D,” I suggested.
Castle Wolfenstein, was a fun and revolutionary Apple II game that came out in 1981. Silas Warner, the developer, was ahead of his time in many ways. His game had an overhead view of each room in a huge, multistory castle. Stealth was an important gameplay mechanic—that is, your survival required cunning and avoidance as much as a killer instinct. It was also one of the earliest video games to have characters that spoke via audio and not just text. The game centered on a soldier, the player, who had been captured by Nazis and imprisoned in the dungeon of a castle. He obtained a gun, ten rounds of ammo, and three hand grenades and used them as he attempted to steal Nazi plans and escape. I played that game so often that I learned how to escape every one of the castle’s randomized floorplans (the exits were always at the same places). Like me, Carmack and Tom had also spent hours running from digital Nazis. Only Adrian, who grew up a non-gamer, had missed out on the masterpiece and its sequel, Beyond Castle Wolfenstein.
The response to my suggestion was immediate.
“Ah, yes! Of course!”
“That would be awesome.”
Their reaction was one of the reasons I loved working with these guys. It’s like we had lived parallel gaming lives.
Tom gave us a pitch in short order. Captured American operative William “B.J.” Blazkowicz single-handedly attacks the Nazi regime in three different episodes: “Escape from Castle Wolfenstein,” “Operation: Eisenfaust,” and “Die, Führer, Die!,” the last of which ends in battle below the Reichstag with Adolf Hitler in an armored suit—we dubbed him Mecha-Hitler—complete with chain guns. The name of the third episode mirrored Commander Keen’s third episode, “Die, Keen, Die!”
Of course, there was one sticky problem with remaking Castle Wolfenstein: the name of the game. Assuming MUSE Software owned the trademark, Tom and I started brainstorming alternative names, but nothing stuck.* So we put the name on the back burner, which we had the luxury to do since the game didn’t have a ship date. For the time being, we were content calling it Wolfenstein.
We started going over all the features we wanted in the game, starting at the beginning. We wanted a graphical loading screen, something new for our games, and one that was tech aware. Like the Keen games, we wanted it to show how much memory you had, which graphics modes you had available, and tell the player the game detected your gamepad, joystick, sound card, and other info. Since we were building upon Catacomb 3D’s source code, we started making Wolfenstein with sixteen-color EGA graphics as we had always done. Tom came up with the roster of enemies for the game, and it looked like we would need some help with character animation. Each character was more involved than a Commander Keen character, because we wanted these enemies to appear 3D even though they were sprites (2D images). We planned to draw them at every angle so you could even see their backs—something Catacomb 3D didn’t do. Wolfenstein would know which sprite to draw depending on the angle from which you viewed them. To help us, we contacted Jim Norwood, who made Bio Menace with our Keen engine. He agreed to start drawing sixteen-color Nazi rotations.
We were only a couple weeks in when Scott Miller called me to hear about the new game, eager to publish it. I gave him the rundown Wolfenstein’s cool features, and he loved the concept. However, he had one suggestion.
“Make it in VGA,” Scott said. “Forget EGA. It’s the past.”
I had to admit he had a good point. We planned to make Keen 7 in VGA, but we didn’t think about it for Wolfenstein 3-D because we were using Catacomb 3D’s EGA engine.
I shared Miller’s comment with Carmack. He thought about it for a few minutes. Then he said, “It would be cleaner code-wise.”
Art-wise, it wasn’t so clean. With the switch to 256-color VGA from sixteen-color EGA, all the art made up to that point had to be scrapped, including the art we outsourced to Jim. Adrian spent some time considering the change and assessing Jim’s style. He decided he could do better on his own, and was even feeling excited about it: His color palette just multiplied by sixteen.
Scott wasn’t the only one looking for a game. We still had one more to make to fulfill our two-game contract with FormGen. It made the most sense to use the Wolfenstein tech and make a new game with our existing Wolfenstein IP. I rang FormGen and talked to Randy MacLean, one of the owners.
“We’re developing a game called Wolfenstein,” I said. I barely got the name out before he jumped in, excited.
“That’s great! What’s it about?”
“It’s a 3D World War II Nazi blast-fest,” I answered. The enthusiasm of the whole team was in my voice.
The line went silent.
“Ah, guys, don’t go digging up that World War II stuff,” Randy said. It was a funny comment, considering the success Indiana Jones had at that point, not to mention numerous other books and movies.
“Sorry, Randy, but that’s what we’re making!” I answered.
To create the levels for Wolfenstein, we decided to use TED5 once again, since the levels were going to be based on a 2D grid like Hovertank One and Catacomb 3D. Tom got to work making the icons we’d need that represented characters, items, and features such as Nazi patrol paths, level start, and exit.
Meanwhile, Carmack got the project established and running quickly. It was able to run levels within a day or two, and a couple of weeks later, every asset was quantified. We just needed to make them and put them in the game. Carmack didn’t take long to convert the renderer to VGA and optimize it. I worked on the menu system and made levels with Tom. Adrian cranked out art constantly.
I wanted Wolfenstein’s audio to stand out as well. To date, none of our games had digitized audio. We’d previously used the PC speaker, which was primitive, and FM synth audio (the sound of the 1980s). Digitized audio, on the other hand, is like listening to an MP3. I saw it as another opportunity for us to innovate and pull ahead of our competition. The Soundblaster was capable of supporting digitized audio, and we figured we could replicate the original Castle Wolfenstein’s digital audio but go over the top. Once that feature was added to the game, for players without a Sound Blaster, we still needed to support only the PC speaker and AdLib FM Synth sound effects. So, in effect, we needed to support three different sound systems. I modified MUSE to handle digital audio files so everything could be packaged cleanly.
The most important thing, of course, was the gameplay. At first, we started by reproducing several of the gameplay features of the two previous Wolfenstein games made by Silas Warner. It wasn’t a clone, but rather a reimagining of features we knew game players would expect, except this time, all in a 3D space. Players searched dead bodies, dragged bodies around to hide them from guards, and unlocked chests for loot. Our game even allowed you to wear a Nazi uniform to sneak past guards and hold up others to take their guns and ammo. They were some great features, sure, but when we played, something else was happening. We were having more fun running and gunning Nazis than using the slower-paced game features like looting. “Speed” became our operative word, the core around which the game was to be built.
As a game designer, playing the game is essential, not just to test out the features you’ve created, but to find the fun. Sometimes, that fun isn’t in the things you thought were features, as was the case with the early Wolfenstein, or worse, the features are getting in the way of the fun. This is why completing an early “first playable” of a game is so important. “First playable” is an industry term and defines the point at which the core loop of the game is completely playable. For a first-person shooter, that core loop is often something like this: Players and enemies can move, shoot at one another, take damage, and die.
The drive for speed ended up simplifying our game design. Anything that slowed the game down got cut. Goodbye, loot chests, dead guard dragging, and holding up guards. The fun of the game was in destroying screaming Nazis as fast as possible. To that, we added high production values.
Our core was this: Kill everything in sight. By the middle of February 1992, we had our template down.
Although Mark Rein was assisting us, I was still involved in the business strategy for id Software.* I had read that Sierra was thinking about entering the kids’ game market, and I thought we would be a good fit, especially since our newest game, Commander Keen in Goodbye, Galaxy, was a great all-ages title. So I wrote to one of my heroes, game designer and entrepreneur Roberta Williams. She and her husband, Ken, built Sierra Online, a gaming empire, and scored multiple hits with a series of Quest adventure games. Their company was a true industry behemoth. My letter was part fanboy, part business. As always, I was interested in making contact with such an influential game designer. More than that, though, I thought we might be able to work on a new IP for kids and publish it through them. I sent my letter off with a copy of Commander Keen in Goodbye, Galaxy.
Roberta liked the game and showed it to Ken. He liked it, too. She wrote back and requested a meeting. I was completely stoked that my initiative had paid off. I was going to meet two of my idols and convince them to fund development and publishing of more Keen games. It certainly felt possible. We were already making $50,000 a month from Keen and Gamer’s Edge. We had the team and the talent, plus we had the early levels of Wolfenstein in our back pocket. It was worth a try.
In mid-February, we flew out to California and drove to Sierra’s offices in Oakhurst, about twenty minutes from Yosemite National Park. Mark Rein flew from Toronto, too. It was a great getaway for all of us. We’d been grinding in frozen Madison, Wisconsin, for months. Now we’d been transported to beautiful, warm California. Meeting Roberta and Ken was an honor. They were warm, kind people and game developers who walked the walk. Their offices were in a huge metal building with at least two hundred people working there. We saw an adventure game in development, and Ken showed us how the hand-painted art was scanned in for each scene. He introduced us to Warren Schwader, an Apple II legend, and Tom and I went into Warren’s office, got on our knees, and bowed, saying, “We’re not worthy!” He also introduced us to the AGI (Adventure Game Interpreter) team. We got together in a meeting room to talk. They could tell Carmack was, let’s say, a more advanced programmer.
At one point, Ken wanted to talk business.
“I’m curious how the shareware model is working for you,” he said. By this time, Sierra Online sold full versions of their games exclusively at retail outlets.
“Great,” I told him.
“How much are you making?”
“A little over $50,000 a month in sales.”
He got deadly serious. “No way,” he said. “That’s not possible.”
Because we were independent and geographically isolated, we had no idea whether it was possible or not. We just knew what we had.
I showed him our earnings statements. He was stunned. His reaction made us all proud. If he was impressed, we must be doing well indeed.
Later, we went out to dinner at Erna’s Elderberry House, the fanciest restaurant in town, where Ken and Roberta were royalty. The management gave us our own dining room with a fireplace and a long banquet table. It might have been the most posh, fancy experience I’d ever had in my twenty-four years on the planet. It felt like I was in a dream. The Williamses ran a great company and made great games. It was like we were commoners dining with the king and queen.
That night, back at the hotel, Mark wanted to talk to us in my room. He was trying to make the case for getting company stock because a major deal might be happening the next day.
“C’mon man. You’re not serious,” I said. “We’ve been working incredible hours for a year.” His probationary period was meant to get him a job, not ownership. The discussion of stock ended, and we returned to talk about the experience we had at dinner with the Williamses.
The next day, Ken and Roberta took us to their house in the mountains, situated on a lake. The views were breathtaking. After putting them on a pedestal for so long, we couldn’t believe they brought us to their house! We were in shock for the next hour. Tom got a peek at Roberta’s King’s Quest VI design guide while I installed our current work in progress on Ken’s home PC. We’d only been building Wolfenstein for about a month and a half, but it was already head and shoulders above any other PC game, including Keen, which had gotten us in the front door. Plus, I knew that he and Roberta were part of the generation that worshipped the original Castle Wolfenstein, so I thought this would knock his socks off.
I finished the demo, eager for his response.
“That’s cool,” he said. “Let me show you Red Baron.”
I was stunned. I had just demoed a groundbreaking 3D game that represented a whole new paradigm, a VGA texture-mapped, first-person shooter with more depth and detail than any other game out there—and his response was to show me a flight simulation game in a genre that had been around since the 1970s. It felt like a replay of everyone’s initial reaction to Dangerous Dave in “Copyright Infringement,” but worse. An industry giant, a guy I considered a legend, seemed to miss the revolutionary game design and concept that was right in front of him. Or did he?
Ken may have been more impressed by our shareware earnings than Wolfenstein, but he was definitely interested in what we had to offer. The next day, when we sat down in his office to talk, he offered us a $2.5 million stock deal to buy id Software. It was a major opportunity for us—we stood to make half-a-million bucks each. I told him that I would get back to him once we were back in Madison and had a chance to talk about it.
We left Sierra feeling good about our visit. Adrian even created an image with the question, “Part of the Sierra Family?” because acquisition looked like a likely result.
When we were back in the office, I wanted to talk about the Sierra offer to get everyone’s pulse.
“Do you know how long it will take us to make that kind of money?” I told the team. “If we do this, we will make it all in one day.”
Despite my enthusiasm, I followed the script I’d used with Scott Miller. I called Ken and made a modification to the proposal by asking for $100,000 in cash up front. I had a number of reasons for doing this. First, I’d realized that we wouldn’t, in fact, all make half-a-million bucks in a single day; there are always vesting timelines for stock deals. By asking for this advance, we would get some cash immediately, but we had to wait for the rest. Second, I wanted Sierra to put some real skin in the game and $100,000 seemed like a modest amount to me for a company like Sierra Online.
Modest amounts, however, are in the eye of the beholder. Our request for a down payment was too much for Ken and Roberta Williams. He nixed the up-front cash, and we walked away. I felt a strange combination of disappointment and defiance. Doing the deal would have been gratifying, but I knew we were better coders and game developers than anyone at Sierra, so once Ken refused to add cash to the deal, my attitude was: “Too bad for them. We don’t need them.” I knew that with a few more releases, we’d be in even better financial shape, making money on new and old games and our game engine. Stacking income was a valuable lesson to learn. The more games we made, the more money we made. That could only happen if we continued owning our own company, not by selling it to someone else.
We had a game to make, so we got back to work, and I forgot about doing deals and focused on finishing Wolfenstein. At this stage of development, the engine was pretty solid. We had several Nazis in the game, like the Guard, SS, Officer, and the German shepherd. The AI was working well, and they would walk on paths that we set up in TED5. We were mostly excited about a new engine feature we could use in devious ways: sound zones.
Alerting guards is a bad idea in Wolfenstein, so you want to kill Nazis when the doors are closed. When you open a door, the sound zone you’re in will connect to the sound zone outside the open door, making a much bigger sound zone. Our sound zones were color coordinated, and we had thirty-four different sound zones at our disposal. We filled a room with one color, then made sure the halls outside were a different color, but we’d then use that same color in another room far away, thus alerting those guards anytime a sound was made in a sound zone of that color. It was a basic idea, but it allowed for some scary surprises. Sound zones increased realism because it made sure if you shot someone in a room, someone would hear, whether in that room or in a similarly colored sound zone.
We realized that there was a lot of art to make, and it would be smart to get some help for Adrian. We threw some names around for a minute, then someone mentioned Kevin Cloud. Kevin was a Shreveport native who was a calm, competent workhorse. He was Softdisk’s editorial director and had become id’s primary contact while fulfilling our obligations to the company, but he had also worked as a computer artist at Softdisk doing art for the Apple II department. Everything we saw of his we liked, and moreover, we liked him. We decided to invite him up to our office for an interview a week later. It was March and still snowing, so we warned him to be careful.
Madison’s snow was heavy and wet and prone to send those not used to it off the road and into a ditch. Tom, Carmack, Adrian, and I stayed inside and spent our weekends playing Dungeons & Dragons instead. I always enjoyed our D&D sessions in Shreveport, but that winter in Madison, as the storyline grew darker and more gripping, I found them even more exciting. Once, when I was hiding out in a pocket dimension with a semi-powerful figure, Carmack presented me with an option to give a demon The Demonicron, the book that controlled the demons’ ability to enter the prime material plane, where we lived. I took the chance. I made a deal with the demon hoping to get a magical sword, the Daikatana, along with a couple other powerful artifacts. It was a disastrous decision. I got the sword, but it didn’t matter because ALL the demons wound up teleporting onto the material plane and destroying every last thing in it, and thus the game over the next several sessions. The human race was obliterated, and it was all my fault—although a roll of the dice shares some of the blame.
I was crushed when our D&D world ended. We’d spent dozens, if not hundreds, of hours exploring this alternative universe, and Carmack spent years building it up. Now it was gone. I still have no idea why he allowed his world to end like that; maybe he was tired of playing with us or was an absolute stickler for the rules. To play again, we had to start over, and that meant waiting for our Dungeon Master to devise another world. Sadly, we never played D&D together again, but in the obliteration of our game was hidden an odd circle of life: Not too far in the future, the storyline of our D&D saga inspired both a gaming revolution and my most troubled project.
A lot of id’s biggest decisions happened late at night. Our unofficial motto was “We are the wind,” meaning we could blow out of wherever we were or change direction at any time. We worked superhuman schedules. We developed games faster than anyone. To do that, we quickly absorbed information, analyzed it, and made decisions. We also were id-driven. We wanted whatever it was we wanted—to make the best, coolest, most fun games—and we wanted it now. We called our decisions bit flips—when a computer value changes from on to off, or vice versa.
Our decision to move out of Madison, Wisconsin, happened late one night. It was instigated in part by the impending arrival of Kevin Cloud from Softdisk. He drove nineteen hours straight up to Madison with his wife, Lacey. When he arrived at the office at 8 a.m., way too early, he knocked on the door and Carmack answered in his underwear.
Carmack said, “Come back later” and closed the door on them.
Kevin came back later, interviewed with us, and we decided he would be a great addition to the team. He was hired.
It just so happened that our leases were about to be up at The Pines, and we wanted to move to better apartments down the highway near the West Towne Mall. Our apartment complex was becoming riddled with drug dealers and police visits. Kevin and Lacey came along with us to hunt for something in that area. We spent a full day looking at apartments, and finally settled on a complex that we could all move into. It was much nicer than where we were, not that The Pines set a high bar. We had dinner on State Street and decided to go and sign our leases in the morning. Kevin and Lacey went back to their hotel and planned to do the same. As usual, Carmack and I went back and started coding again, Tom worked on levels, and Adrian picked up where he left off.
Around 1 a.m., I just blurted it out: “I hate the winters here. I hate the snow and ice and shit.”
“Yeah. I hate it, too!” Adrian blasted.
“I really don’t want to stay here anymore,” I added. “We should just leave the state. Fuck this. And fuck those new apartments.”
“I agree,” Adrian said.
I went on a brief rant about not being able to walk outside all winter long. I grew up in the desert and Northern California. I knew there were other, warmer options almost everywhere.
“Remember how cool it was by the lake?” I said to Carmack, reminding him of his waterfront house in Shreveport. “I don’t want to go back there, but there must be somewhere else that’s warm.”
We started throwing around destinations like Jamaica and the Bahamas, but eventually, we decided we needed to be in the US. I offered Salt Lake City, where I had lived with my dad.
“But it has snow, so fuck that place, too.”
I mentioned Arizona, but there was no water, and it was hotter than we wanted. New Hampshire? Snow. No.
Then Adrian said, “How about Dallas?”
I grabbed a map from Tom’s desk. He had a bunch of maps from our earlier travels.
“Look! There’s a lake there,” I said and pointed to Lake Ray Hubbard. “We can just get a place on the water!”
Tom was silent. He loved Madison, and he didn’t want to go. He started making a list of pros and cons on his computer. There weren’t many pros that outweighed the snow.
We started checking off the selling points for Dallas: Scott Miller was there, and so was Apogee. It was warm. Texas had no state income tax, so we would make more money. Origin Systems was headquartered in Austin, so there was already something of a development community. That was it. At 3 a.m. our fate was decided. It was three to one. We were moving to Dallas.
Then someone remembered: “Oh, shit, Kevin’s signing a lease in the morning.”
We made frantic calls to his hotel room and left messages: “Don’t sign the lease!” and “Call us before you go anywhere!”
All we were trying to do was stop Kevin from putting down a deposit. We never considered how he might interpret the messages. When he received them, he immediately thought he was fired. Why else would we tell him not to sign a lease? He even told his wife something must have fallen through. When he showed up at our office at 10 a.m., we told him that id Software was moving to Texas. Though he was surprised at our overnight decision, he was also relieved. Dallas was a lot closer to his hometown of Shreveport, so he and Lacey loved the idea.
The wheels were in motion. We didn’t even mind when we discovered that Lake Ray Hubbard was run by the Army Corps of Engineers, so there was no lakefront property.
“Maybe we’ll just have to settle for a swimming pool,” I said.
Tom and I flew down to Dallas to scope out real estate. Scott drove us to several apartment complexes in Garland and Mesquite, walking through the model units. It took about eight hours, and we were talking about games the entire time. We settled on La Prada Club Apartments in Mesquite, just south of Apogee’s offices in Garland. It was better than all the places we looked at—two-story black-tinted windows, amazing air conditioning, and all the luxury amenities, including a big swimming pool. We rented four apartments plus a one-bedroom loft that would serve as our office. Scott loved the idea that we were closer to Apogee—he loved it so much he increased our royalty to 50 percent.
Hanging out with Scott Miller and his business partner, George Broussard, opened my eyes to living large in Texas. Scott was driving a Nissan 300 Turbo ZX. George had a red Acura NSX, which Honda had just unveiled a year earlier. They bought these cars with the money earned from all the Commander Keen games, plus their other titles. We just saved all our money. Our last night in town, we went out to dinner and then headed over to SpeedZone, a local arcade and amusement park, to play video games and race go-carts. Afterward, I got to race around in George’s NSX. Finally, Scott drove me and Tom back to our motel.
It was during this time that we decided to part ways with Mark Rein. I’m happy to report that, despite this, Mark has gone on to do great things at Epic Games, helping to turn it into an industry behemoth. In 2021, it was revealed that he was a billionaire, at least on paper.
With Mark’s departure, we were once again in need of a biz guy. Our old friend Jay Wilbur called us and said he heard we were moving to Dallas. He was ready to leave Softdisk if we’d have him, and it was a unanimous “Yes” on our side. We set April 1, 1992, as his official first day on the job. Jay became id’s biz guy on the same day Kevin Cloud started work.
Set up in new offices, making Wolfenstein was some of the most fun we’d ever had making a game. A lot of that had to do with our liberation from Madison and our new life in Texas. It was warm. It was sunny. We had apartments at the La Prada Club Apartments and a separate apartment that served as the id office with a swimming pool right behind it. You could open the door and take a dip and lie in the sun. We’d had fun in Shreveport, too, going kneeboarding on the lake, but we were consistently working seven days a week back then and had all been in survival mode.
The truth is that we relished the enormous pressure. You know that moment in a game where you’re on your last life? Sometimes that’s what it felt like. We had to keep going, keep playing, or the dream would be over. Now, in Texas, we had a financial cushion from Keen’s success, and Wolfenstein was bound to make far more money. If we wanted to hang out and play games, we could. Street Fighter II and Fatal Fury were our two favorites at the time. There was pressure and a lot of work, but life felt a lot more balanced now that we were focused on delivering just one game.
The hardest thing about making Wolfenstein was finishing the levels. The level design was not visually sophisticated, but that is because we made a trade-off, sacrificing visual nuance—textures, colors, angles—in favor of rapid-fire speed, which was part of the thrill of our FPS killfest. Adding those features would have slowed the game down and required more work on the engine. We felt that the slimmed-down feature set was exactly the type of gameplay we wanted. That kept the core loop clean and easy to understand, which was important since it would be the first fast 3D game that anyone had seen.
Earlier in March, Tom and I noticed Wolfenstein was missing something, something fun that we were used to, but that this engine didn’t support: secret areas. They were a staple of exploration and of our games. We lobbied Carmack to add some way of pushing a wall to reveal a hidden area, but he didn’t want to violate the purity of his engine to hack in a secrets-revealing “pushwall.” The next couple of months, we pressed him on it until he finally agreed in April to hack it in. It was a trying situation for Tom because, as the creative director, his design was expected to be supported by tech, and now there was pushback that affected the gameplay. The functionality of the engine should support fun gameplay, provided it was possible, not be constrained by it.
Wolfenstein was filled with more artwork than any game we had made. The walls featured great-looking textures, more detailed than those in previous games, while the floors and ceiling were solid colors to maximize engine speed. Room after room, level after level, the walls were a variety of brick, turquoise-blue metal, and wood paneling. Yes, Adrian and Kevin did a great job adding warning signs, swastikas, iron crosses, and Hitler portraits, but from a level design perspective, this creativity was limited. Tom and I found the 3D space exciting, but the levels themselves somewhat tedious to create. Fortunately, the gameplay was fun. The blood-and-guts details made it the most violent game around, and the premise—blowing away Nazis to save yourself and steal their plans—was hilarious.
As our new biz guy, Jay Wilbur’s first order of business was trying to hunt down the owner of the Wolfenstein trademark. This was before the internet, so it was not easy. Eventually, after many, many phone calls, Jay discovered that a woman in Baltimore had bought all of MUSE Software’s intellectual property. Jay offered her $5,000 for the Wolfenstein trademark, and she accepted. No more trying out alternate names, none of which stuck—our game would be an official Wolfenstein sequel! We decided to call it Wolfenstein 3-D. It was the third in the series and 3D.
We felt like we were getting close to the end of the shareware episode’s development, but we didn’t have all the music and sound effects we needed. Jay called up Bobby Prince, with whom we’d collaborated before, to see if he would fly down to Texas with his equipment and do the work on site. Bobby brought what seemed like a whole studio with him: a huge sampling keyboard, professional monitor speakers with mounts, a studio microphone setup, and a rack of audio processors. Bobby was the real deal.
We let Bobby use the entire bottom living room area in our La Prada office, and he made the music right there. For the voice over (VO), Tom wrote down all the German phrases that needed to be said by the Nazis. We had a German-to-English translation book, and Tom and I spent an afternoon yelling, “Spion!,” “Achtung!,” “Mein Leben!,” “Mein Gott in Himmel!,” and all the other lines except “Schutzstaffel!”* We cajoled Adrian into saying that one. The death screams were the funniest ones to voice. We got good at using Bobby’s setup to record stuff.
By the end of April, we knew we were in the home stretch with Wolfenstein 3-D. It was getting close. Part of the final polish was making sure all the title screens, end screens, and help screens were in the game with the correct text. One of the extra screens we added just before we showed the title screen was a light-blue screen that said:
This program has been voluntarily rated
PC-13: PROFOUND CARNAGE
By id Software
We designed the screen to look like the 1970s rating screen shown at the beginning of a movie. We added it as a joke, but little did we know it was the first instance of a game being voluntarily rated.
One last detail remained. We had no size limitations on Wolfenstein 3-D, so we just made the game and decided to figure it out later. Well, later was now, and the game was bigger than a single disk. We had never made a game bigger than one disk, so I needed to figure out how players were going to get the game off two disks and onto their hard drives. There were several games that were distributed on multiple floppies by 1992, but they had their own proprietary install systems. Today, it’s even a business that makes decent money. There was no market for this in 1992, however, and so I had to write it myself. My idea was to create an install system and give it away, to get rid of the proprietary nature of this tool that, to me, should be free for everyone to use. First, I needed to write a tool to take a single ZIP file and split it into 1.44 MB-size chunks that could be copied onto 3.5" disks. Next, I needed to write the installer program that took all the chunks, put them together, and decompressed the game to the destination folder.
The program became known as ICE, the Installation Creation Editor, and took about six hours. Then I started writing DEICE, which would copy files off floppies, put them together, and UNZIP the files at a destination the player specified. I wanted it to be bulletproof, just in case a player pulled the floppy out of the drive mid-copy. I got it all working by the end of the next day. It wasn’t long before I gave that system to Apogee so they could use it for all their games.
Everything was ready to go on May 5. The master disk was made, and all of us were testing the game on our PCs. No errors, no crashes, no bugs—the game was solid. After testing every kind of weird scenario we could think of, we knew it was time to upload. It was nearly 3 a.m. We called Scott and George and invited them over.
At 4 a.m., we dialed into Software Creations to upload the 2 MB shareware version. Scott had been talking to Software Creation’s owner Dan Linton that day to prepare him for the game’s imminent release. For Dan, it was 5 a.m., but he was ready, knowing he was getting the hottest game around. The upload finished, we high-fived, and left the office to sleep until whenever. We knew everything was going to change from that day onward. We felt it.
Wolfenstein 3-D’s shareware release was ten times bigger than both Keen releases. The first month, we sold 4,000 preorder copies of the Wolfenstein 3-D bundle at $60 each, which included two disks with three episodes each, plus a hint book we still had to write. Some players bought just the disks and came back to order the hint book later. Wolfenstein 3-D generated $240,000 in sales in May, and our first royalty check for the game was $120,000. Even Randy MacLean at FormGen got behind Wolfenstein 3-D when he saw the reviews of the shareware version. Seems digging up that World War II stuff was a good idea after all. We felt it was going to be big, but not this big. We didn’t change much, but we did decide it was time for raises.
At this point, we had only uploaded the shareware version and still needed to finish the other five episodes. Apogee was accepting orders for all six episodes and the hint book, so we needed to get moving. Even so, it was great to see the reaction from players and reviews in the press from our free shareware release. Four years later, Scott replicated this marketing tactic with Duke Nukem 3D.
Based on the promise of the shareware version, Wolfenstein 3-D was selling like hotcakes, and it wasn’t even done. We had to finish five more episodes, each one containing ten levels. New episodes meant new bosses and new secret levels, and we had to write a hint book. All as fast as possible!
Fortunately, we were good at this. Unfortunately, it was difficult to get the levels done quickly because they were so boring to make. Creatively, Tom and I were used to more visually interesting design spaces with Commander Keen 4 through 6’s lush visuals, EGA graphics notwithstanding. Tom was easily distracted, and to get him pumped up again, I would chant, “NSX! NSX! NSX!”—a promise that he could get a car like George Broussard’s if only we could finish these damn maps!
We buckled down and tore through them. Bobby Prince wanted to make a level, too, and so we let him. It was one fewer for us to create. After the maps were done and tested, we jumped onto the creation of the hint book. Kevin had already started on the hint guide using the NeXTstation to create it. The NeXT was the perfect machine for it since, in addition to its superior tech, it also had superior graphic design programs. It was made for desktop publishing, and even used Display PostScript, so what you saw on the screen was exactly what you saw printed on the paper. Compared to MS-DOS, it was significantly more advanced. Kevin also created a Commander Keen brochure on it.
Tom and I started having fun again. The hint book was more fulfilling creatively because it was our job to inject it with some humor. How many ways can you describe filling a Nazi full of lead to go get a key?
“Get the key, after the SS find themselves more horizontal than normal” was one instruction. Others included:
“The key is hidden behind a wall of uniforms. Lay the uniforms on the floor and you can easily grab the key.”
“Here we are at the fabled Elevator of Floor Six. I see no white-bearded magi, golden unicorns and majestic crystalline gate. I guess the ancient description was somewhat … embellished.”
Not your typical hint book instructions, but they made us laugh.
Kevin wrote the intro text describing the characters in the game, showing Tom’s sketches and hints, and then it was on to the maps themselves. At the top of each level’s description, we put my fastest time through the level—the first printed instance of FPS speedrunning. Page 25 describes my fastest speed through the first episode: five minutes, twenty seconds.
On June 15, after testing, I made the master disks using ICE and drove them to Scott at Apogee, the first time we had handed over a game in person. Previously, the disks were sent through the mail, and so we were removed from the final process. Scott had his disk duplication people ready to start copying. Soon, they were sending customers an envelope with disks and a manual in it. It would be a year before we sold the game in a retail box through GT Interactive.
We were finally done. The shareware episode took us four months from start to ship, and the other five episodes of Wolfenstein 3-D were done in one-and-a-half months. In total, just shy of six months for a six-episode game.
It was time for a short rest. Tom and I spent time in the pool, Carmack bought some nice leather furniture for his place (having had enough of highly optimized but minimal furniture), and Bobby bid us adieu and went back home after a job well done.
Now that customers were getting their orders, we started paying attention to the reviews. None may have been more laudatory than Chris Lombardi’s write-up in issue 98 of Computer Gaming World. It ended this way:
Castle Wolfenstein 3-D* is, with Ultima Underworld, the first game technologically capable of creating a sufficient element of disbelief-suspension to emotionally immerse the player in a threatening environment, even when viewing it on a flat screen. I can’t remember a game making such effective use of perspective and sound and thereby evoking such intense physiological responses from its players. I recommend gamers take a look at this one, if only for a cheap peek at part of interactive entertainment’s potential for a sensory immersed virtual future.
Lombardi also praised the sound and music, even name-checking Bobby Prince’s spy-thriller soundtrack and his fantastic stylized, minimalist versions of fight-song favorites, like the “The Marines’ Hymn” and “Anchors Aweigh.” Wolfenstein 3-D also won many awards, including being inducted into the Computer Gaming World Hall of Fame. Many positive reviews cautioned readers about the graphic violence, and in Germany, where Nazi imagery and paraphernalia were illegal, the ratings board refused to give it a rating, which meant it was banned.
While readers may have heard that there was an outcry over the violence in Wolfenstein 3-D, it wasn’t actually the case, at least not that I saw. On the contrary, most reviews sounded like this one from Peter Olafson in Electronic Entertainment: “Wolfenstein 3-D is drop-dead gorgeous, outrageous to the ear, stay-up-all-night addictive, and easily the best action game available for the IBM.” If they wrote about the violence, it was simply as a disclaimer—“if you find blood and shooting offensive, don’t get the game.” I don’t recall any articles bemoaning the decline of video games, or how horrible Wolfenstein 3-D was. That said, I am sure there were some. We were still in the height of the Satanic Panic, and any new media, particularly successful new media, was viewed with suspicion.
The first true FPS was, in its way, the perfect shooter game. Players were escaping a Nazi dungeon as an American war hero blowing Nazis away. Who is going to complain about a game where you’re killing Nazis, the universally accepted symbol of evil? The people who complained about shooting the dogs surprised me. They chose to ignore the fact that you’re mostly shooting people.
One of the positive messages we received came from an unlikely source: a former Vietman POW who noted that he hesitated to play, fearing flashbacks. He and his friend had dared an escape of their own, making maps and hoarding food in preparation. He made it out, but his friend did not. He said Wolfenstein 3-D allowed him to “face the past” and cured him of endless nightmares.
Another unlikely effect of Wolfenstein was finding out that Jewish kids loved playing it because they felt they were getting revenge for their ancestors’ suffering at the hands of the Nazi regime.
In retrospect, the id founders should probably thank Ken Williams for rejecting our counteroffer—id’s $100,000 in-cash demand—just nine months earlier. We made a lot more money by remaining an independent company. Instead of thanking him, we razzed him just a few months later when we met at the 1993 CODiE Awards, the black-tie gala that was the game industry’s equivalent of the Oscars, run by the Software Publishers Association. The four of us were decked out in tuxedos to attend the festivities, and we strutted away with the Best Action/Arcade Game of the year for Wolfenstein 3-D: It was an historic victory—Wolfenstein 3-D was the first shareware title to nab a CODiE, and id Software was the smallest company to win the honor. If we had joined Sierra Online, the trophy would have been theirs. We saw Ken afterward and I couldn’t resist teasing him. “This could have been yours, Ken!” I said, waving the trophy at him. It was a good-natured ribbing.
“I know, I know. I’ve made mistakes like that before, believe me! It wasn’t the first, and it won’t be the last.”
We laughed, although maybe Ken wasn’t laughing quite as hard as we were.
The success of Wolfenstein 3-D provided a lot of great moments. We now had three best-selling shareware games all earning money and accolades. The amount of cash pouring in was both gratifying and liberating. As a kid who grew up with his mother using food stamps and clipping coupons, it was an enormous, unfathomable shift. Financial pressures ceased to exist. So did the normal rules of work: The kid who used to wake up at 3 a.m. to deliver newspapers to earn Pac-Man money could now sleep until noon, and when he woke up, there was a Pac-Man machine in his office. Rest assured, though, if I got in at noon, it was because I had been there until 5 a.m.
However, one of the sweetest moments tied to Wolfenstein 3-D’s release occurred during the July 1992 KansasFest. This was the annual computer jamboree held at Avila College in Kansas City, Missouri. It evolved out of AppleFest—the short-lived Apple II series of conferences, the one that landed me a job with Origin Systems. Tom, Carmack, and I had attended previous KansasFests and loved them, because they were filled with Apple II programmers. To us, this was the best thing ever; we’d stay up all night in the dorms just talking with members of our unofficial Apple II-loving tribe. In ’92, we showed up and learned that Silas Warner, the creator of Castle Wolfenstein, was the guest of honor and was going to give a talk about his former company, MUSE Software. This was a great coincidence. In preparation for the festival, we bought the world’s first color laptop—a $5,000 Toshiba—and installed Wolfenstein 3-D on it. Now we would get to show it to the man who inspired the game.
It was a great experience. Silas’s talk was about the history of MUSE Software and all the different tools and games he wrote. Afterward, he was swarmed with coders who had questions. We waited until the crowd died down and showed him Wolfenstein 3-D. He was impressed and happy that the game lived on, even though we left out the stealth aspects and accentuated the violence. Later that evening, a group of us were sitting on the floor in the hallway hanging out. Silas came over and sat on the floor with me, Tom, Lane,* and Carmack. It was an incredible moment: The visionary who inspired us, who we idolized, had literally lowered himself to our level, ready to just hang out and talk about anything and everything. At the end of our long talk, Silas signed the Wolfenstein 3-D manual that accompanied the game. That manual is still at id Software’s office. I hope someone framed it.
* We were so busy trying to define the cutting edge and making a ton of games that we didn’t do a lot of competitive analysis. During 1991, we played a lot of the new NEO-GEO console games, and each one cost us more than $100. We played SNES and Sega Genesis games, and also bought a handheld Atari Lynx. We played everything new on the consoles because PC games had nothing like that. To us, they felt dated, and we were trying to change that.
* The similarity in names between our MUSE tool and MUSE software are coincidental.
* Mark was still in his voluntary six-month probationary period.
* “Spy!,” “Danger!,” “My life!,” “Oh my god!,” and “Protection squad!”
* Computer Gaming World incorrectly identified the name of the game.
* Lane was still at Softdisk but regularly attended KansasFest.