Adrian and I had been in touch throughout the years, but we really reconnected in 2014 while Brenda and I were driving cross-country from California to New York. We stopped at QuakeCon in Dallas, where Adrian still lived, and had a fun, laugh-filled evening. We spent hours talking about our time at id (by then, he had left, too), our plans for the future, and the current trends in games in general and, of course, FPS games specifically. We shut the restaurant down, and continued talking in the parking lot until nearly 4 A.M. We connected again in Ireland a month later while we were staying at Adrian’s hotel, The Heritage. We talked about everything under the sun, but it was the first time in a long time that I’d had extended conversations on FPS design specifically.
I had been thinking about the next big thing—not just for games, but for me. It had been fifteen years since my last first-person shooter. I still loved the genre. When I played games, more often than not, an FPS is what I was playing. I was just waiting for the right idea, the right design, the right moment. Retro-style shooters were picking up steam, but I didn’t want to design a retro shooter, exactly. I wanted a modern shooter that was cutting-edge, like the games we had made at id, with classic elements that would remind players of how fun and visceral these games could be, and that younger players could discover for the first time.
Although I was getting excited about designing a new FPS game, it wasn’t my top priority. I had to move it to the back burner to deal with everything that came with a move to Ireland: new schools for the kids, finding a place to live, getting our visas sorted. Only when the dust from our move had settled and we’d leased an office for our newly created company, Romero Games, did I finally turn my thoughts back to designing a first-person shooter.
I had a futuristic design percolating, but before I designed something new, I needed a warm-up.
As fall gave way to Christmas, I shared exciting news with Brenda: I was going to make a new DOOM level.
We both realized this was a big deal. I hadn’t made a map for DOOM since The Ultimate DOOM in 1995. On top of that, every map I’d made for a DOOM game—DOOM, DOOM II, and The Ultimate DOOM—had been part of a shareware or commercial package. This time, I would be making a DOOM level as a member of the community I had helped create back in 1993. That year was key. Most DOOM modders make maps for DOOM II. It has more weapons, enemies, and power-ups. I had my sights set on the first DOOM, though, and for good reasons.
First, before I designed a new FPS from the ground up, I wanted to get back in the right creative headspace by returning to a style of level design I knew and loved. DOOM was the perfect choice. I know the engine inside and out, and building maps in Doom Builder 2, a free editor written and maintained by Pascal van der Heiden, is fast and easy. Second, the development of DOOM was the best sort of chaos. John Carmack was writing the engine, and I was juggling DoomEd, level design, and gameplay. When you play a DOOM map, anything that happens in that level—a lift that moves, a lever that flips, a button that glows when pressed, lava or toxic slime that burns you, a light that flashes on and off, a ceiling that crushes the life from you—I’m the one who designed and programmed it. I made every map in episode one of DOOM, “Knee-Deep in the Dead,” except for a small bit of level four, which Tom had designed before he left id, and the episode’s final level. I turned that job over to Sandy. He was a great choice: He worked quickly, pounding out smaller maps every couple days at his peak. That freed me up to focus on DoomEd and our distribution plans.
In 2015, I saw an opportunity to end episode one the way I would have ended it. It’s not that there was anything wrong with Sandy’s E1M8 map. On the contrary, I really like it. He made a gloomy, claustrophobic map that focused on pitting players against a small amount of enemies before an encounter with the “Bruiser Brothers,” the Barons of Hell waiting for them in the boss room. I loved the sight of twin doors opening to reveal the Bruisers, and the two of them striding forward as they roared like tyrannosaurus rexes. I would keep that setup, but I planned to make that showdown even more epic. If I had a criticism of the map, it was that, for a final map it was too small.
Looking back at episode one, every level was bigger than the one before it. The moments, the encounters, the weapons, the map designs—everything escalated, pushing the player toward a confrontation against the biggest demons they’ve yet faced. That’s what I had in mind for E1M8b, which I called “Tech Gone Bad.” This was the final level, so I wanted to take the player on a long, grand journey that culminated in an epic boss fight. I applied the techbase theme I used for E1M1 through E1M7, keeping the continuity of the episode’s environments, and made “Tech Gone Bad” the largest map in the set.
Throughout the map, the player moves in and out of interior rooms and outdoor regions covered in slime. This forces them to adapt as they navigate between safe and hazardous floors, and to fight enemies from different vantage points. Occasionally, the player reaches high positions that give them a view of the level, driving home its scale and letting them scout for regions they haven’t explored. Eventually, the player realizes they’ve been running on top of passageways they haven’t explored yet, and they need to find a way to enter them to progress. It’s as if the level is a puzzle box, and making progress opens more avenues of exploration to you.
Of course, everything leads to a battle against the Bruiser Brothers. Sandy’s level has players race down a short hallway and jump onboard a lift that raised them into the arena. The first things players see are two closed doors. You know something bad—two bad somethings—are hiding behind them, and moving forward triggers them to open. My goal was to amplify that cool factor tenfold.
E1M8b’s boss room is a chamber with just enough room for the player—and, inevitably, the Bruisers—to maneuver when the fight breaks out. Players enter by way of a narrow ledge set high above. Stepping onto it causes the ground to break away, lowering the player into the arena like fish bait set on a hook. That was thematic. Whereas Sandy’s E1M8 has a lift that raises the player into an arena, I wanted the player to descend. Below, the walls are made from computer terminals, sticking with the map’s theme, but the rest of the room has a distinctly ominous vibe. Glowing red cracks line the floor, snaking out from two gigantic red portals with steps leading to each entryway that dominate the center of the room. I hoped the sight was so terrifying that it gave the player pause. What the hell is going to come out of those? Move a little closer, and the Bruiser Brothers appear, kicking off the most epic encounter in “Knee-Deep in the Dead.”
Despite using someone else’s map editor, I found designing maps for DOOM as natural as it had been more than twenty years earlier. However, everything about my environment had changed. For one thing, DoomEd was written on the NeXTSTEP operating system, so I couldn’t use that. The original DOOM code doesn’t run on Windows, so I used a source port named GZDoom. I ran Doom Builder 2 on a 2013 Mac Pro with 128 GB of RAM, a terabyte solid-state drive, and two 3D graphics cards. Doom Builder 2 isn’t compatible with Mac OS, so I ran it through a virtual machine (VM) that had Windows 10 installed. As I designed E1M8b, I dropped new versions of other files I needed into a Dropbox account and retrieved them in the Windows 10 VM environment. Doom Builder 2 and GZDoom ran at full speed with no slowdown, because I wasn’t emulating either program. My Mac Pro ran on an Intel processor, giving me access to GZDoom in sixty-frames-per-second glory.
In short, designing E1M8b was easy and felt like going home. I didn’t develop a test level first. E1M8b was the test level, and the version you play was my first draft, created in about twenty hours over two weeks. Every time I sat down to Doom Builder 2, I felt things coming back to me. It was muscle memory, with no gap between 1995 and 2015.
On January 16, 2016, I uploaded E1M8b to my Dropbox and tweeted out the link. To set the mood for my take on episode one’s final level, I wrote a story in a text file that accompanied the map:
After exiting the Computer Station, you knew the worst was up ahead. You still hadn’t reached the place where the demons were coming from. The steel door shuts behind you as you realize you’re there; you’re at the Phobos Anomaly. Cracks from hell are all over the place as seepage from the portal invades the entire installation.
Now it’s time to find the portal and stop the demons from coming through. You know UAC had hundreds of scientists working at a high-tech lab somewhere in this area, and the portal must be connected to it somehow. Time to lock and load.
I’d hoped the community would be excited about my first DOOM map in more than two decades. As it turned out, that excitement extended far beyond just the community. Within hours, gaming outlets like PC Gamer, Polygon, IGN, Shacknews, and Eurogamer had written stories about it. Soon, mainstream tech outlets like PC Magazine covered it, too. Within ten minutes, the map had been downloaded thousands of times, and players were already sharing their thoughts with me on Twitter, in forum posts on doomworld.com, and in playthroughs live on Twitch and uploaded to YouTube. Word of my return to DOOM went beyond the gaming sphere, as well: “New DOOM level released by game creator John Romero” read a headline by the BBC.
Behind the scenes and out of the spotlight, I had already started concepting my next game, a new first-person shooter. At the time, Kickstarter was established as a viable means to fund a game, and I decided that was the way to go. E1M8b was the confidence boost I needed to feel that gamers would embrace a new FPS from me.
While I worked on the new shooter and its Kickstarter campaign, I was itching to create more maps for DOOM. I decided to strike while the proverbial portal to hell was still hot. Once again, I was more interested in replacing a map from episode one than creating a level disconnected from the original game. Tom Hall had started E1M4 before leaving id. I had stepped in and re-textured it, adding things like weapons, power-ups, and enemies to balance the map flow on every difficulty setting. My issue with E1M4 was that there was only one thing special about it. Near the end, you use the blue keycard to access a maze. Tom did a great job on this part: It’s darkly lit, and the walls of the main artery are so tight, the ceiling so low, the player almost feels like they’ll have to crawl on their belly to get through it. Side passages unspool from the main path, and each passageway leads to supplies, or dead ends, or a fight against imps and pinkies. The maze is memorable, but the rest of the level felt like yet another futuristic lab.
E1M4b, Phobos Mission Control, is my vision of episode one’s fourth level. My goal was to do something that none of DOOM’s original maps had done—to build a map of layers. At the start, players think the map is small. Every button they press, every switch they flip, and, in some areas, every step they take peels back layers. Elevators come to life, stairs rise to let them reach previously inaccessible terrain, and walls drop to show new paths. Every new path hits the player with a surge of excitement as they move deeper into unexplored terrain. E1M4b is especially fun to play in deathmatch, because bringing down walls reveals things like openings players can shoot rockets through to gain a tactical advantage over their opponents.
I dropped E1M4b as a surprise on April 26, 2016, just over three months after E1M8b, and one day after launching the Kickstarter for Blackroom, the new FPS I’d been working on over the last few months. The release timing of E1M4b was no coincidence. In my tweet, I described it as more of the classic FPS gameplay I promised to deliver in Blackroom. Fans could download it from my Dropbox, along with a readme file on how to play the map and a story to get them excited:
With the Toxin Refinery in the rearview, you make your way to Phobos Mission Control, where the computers crunching the data from the Phobos Anomaly are located. You need to use them to gain access to the Phobos Lab—but you remember hearing that the computers were tied into all areas of the installation and that you never knew when the environment around you would change. You need to keep your eyes alert to all movement—this place is not what it seems …
I was still in a groove from “Tech Gone Bad,” so the development of E1M4b flowed similarly. It was a few hours in my spare time over two weeks, between working on the design and Kickstarter page for my upcoming FPS. My goal was to take what I knew about level design and give fans a style of map they had never experienced in the original game.
The positive reception to “Tech Gone Bad,” the unveiling of Blackroom’s Kickstarter, and the surprise announcement of another free DOOM map was a triple whammy of positivity for me. DOOM fans were ecstatic to experience new levels made by the game’s original designer, and the hype around those was transferred to Blackroom’s crowdfunding campaign.