Like journals sprinkled throughout RPGs, Pause Screens go into more detail on the people, companies, and cultures that contributed to the success of the Pillars of Eternity franchise and the Infinity Engine line of roleplaying games.
Individually, John Romero and Josh Sawyer are design legends. Together, they’re a veritable dream team.
In 2003, Midway Games made the dream team a reality when Romero and Sawyer joined forces at the developer’s San Diego studio to co-direct Gauntlet: Seven Sorrows, a follow-up to 3D arcade hit Gauntlet Legends and its expansion set, Dark Legacy.
On paper, Gauntlet Seven Sorrows had all the trappings of an action-RPG apotheosis. It would boast the button-mashing madness that had put Gauntlet on the map, plus RPG mechanics and a deeper story. And it would be helmed by Romero and Sawyer, two veteran designers known for frantic action, tactical exploration, and immersive storytelling.
There was only one weak link in the chain. By 2003, the coin-op and console developer was as known for classics ranging from Mortal Kombat and NBA Jam as it was underhanded management and poor executive decisions, which led to its bankruptcy and subsequent dismantling by new owner Warner Bros. in 2009.
Seven Sorrows is not part of the Infinity Engine lineage of games, or Obsidian’s Pillars of Eternity franchise, but it was close enough—a cousin, say—that I couldn’t pass up the opportunity to learn more about it. Josh Sawyer had already given hours of his time to answering my questions and was tied up in studio work before his planned sabbatical in the late summer of 2018, so I caught up with John Romero to learn how Gauntlet: Seven Sorrows got started, what he and Sawyer aspired to do with the game, and where it went off the rails.
Craddock: You'd been in Dallas for years, since id Software relocated there for Shreveport, Louisiana. What prompted you to pick up stakes and head west?
John Romero: I had just finished Red Faction on the N-Gage. That was with my company, Monkeystone Games. I just wanted to get out of Dallas after that. I'd been there for eleven years, and I was ready to try new things. I was working with a recruiter, and I found out there was an opportunity at Midway San Diego. I thought, I love San Diego. Why wouldn't I want to live there? I went and checked the place out, interviewed with them, told them I would love to take over the Gauntlet project they were going to start.
They did mostly third-party support at that location, but they had a lot of executives. Basically, executives were at [Midway] Chicago, or in San Diego. I talked to [CEO David Zucker], who was just a contract president who'd be in the position for the foreseeable future. I really liked Gauntlet, like everybody, and I wanted to do something cooler with Gauntlet that gave you a deeper universe for these four characters, something more fleshed out than the simplistic thing you saw in the original game and even in Dark Legacy. They said, "Okay."
“Imagine Black Isle making a Gauntlet, but with action and a different camera system, and less deep-RPG stuff, but with more story. It would have been really cool.” -John Romero
Craddock: I love that you designed a game for N-Gage, because to look at your gameography, anyone can see that you've developed game software for so many platforms. That seems to reinforce that you just love developing games regardless of platform.
Romero: Yeah, pretty much on everything. Before id--before 1993--was my first time making a game for a non-8-bit computer. Before 1993, it was Apple II, Commodore 64, and [IBM compatible] PC. Those were the three platforms I worked on at that time until 1993, when we needed to make Wolfenstein 3D for the Super Nintendo. That was my first time working on a console, but it wasn't a primary [project]; it was a port. Still, I got console experience doing that. And while we were making Doom, we were developing it on NeXTSTEP workstations. So we were using one platform to make a game for another platform.
When I worked at Origin, I was porting 2400 A.D. from the Apple to the Commodore. I was programming the game on the Apple II, transferring it over, and just running it on Commodore. I was not going to develop it on a Commodore, so that was cross-development. I did a lot of that. We worked on the Jaguar version of Doom, and then, when I started Monkeystone right after Ion Storm, that was primarily to develop for the Pocket PC that was out at the time, the iPAQ. I made Hyperspace Delivery Boy for the Pocket PC, and then I made it for the PC. Then we were making games for other cellphones, and I got into the N-Gage cellphone, yet another platform. Then Midway came along, and I thought, I'm finally going to make a game [natively] for a console.
I even did Game Boy Advance programming at Monkeystone. We did Brew, Sirius 60/Nokia, which was like N-Gage, using an engine called X-Forge, by a company called Fathammer. That's what was used for Red Faction.
So, all kinds of different technology, but finally, at Midway, I could focus on consoles. After Midway, my focus was on MMOs. Just lots of different stuff over the years.
Craddock: What was the transition like, after having lived and worked in Dallas for so long?
Romero: I moved everything over from Dallas to San Diego. I had stuff in storage in Dallas and moved that into new storage in San Diego so I would have no reason to go back. I started in October of 2003. I was friends with the guy who was running the San Diego studio, so that was great: I knew the guy who was running the place. I'd met Michael Gottlieb years before when I was at id. He was the head of third-party [development], so there were a lot of cool people working there.
I remember when I went there and said I wanted to work on that game, and told them all the things I wanted to do. One was, "I want to take over this dev space. I want to re-paint it. I want new lights. I want a dev environment that inspires the game we're trying to make." I got this huge area. I had it painted the color I wanted, removed fluorescent lights removed and got incandescent lights all over the walls. It was this really nice, dimly lit space lit up with yellow light.
It was really cool, and forty people took up a lot of that space. I got to bring some Monkeystone people on as programmers and designers. Tom Hall came because when I said I wanted to move out of Dallas, he said, "Well, you're not going without me." He became a creative director in third-party [software].
Craddock: Was Gauntlet in progress when you started?
Romero: They'd been working on a prototype with a small team, and they'd just had a big layoff from the previous game, so they had this big building for development that was mostly empty. I was project director on the game, and they had been working for about six months on a prototype that I was not a fan of. They had a camera, a third-person camera, but it wasn't limited in any way. You could look off into the distance, which made it really chug and go slow.
Gauntlet really wasn't about that perspective. I thought it needed to be more in the classic Gauntlet perspective, but maybe we could do some cooler stuff with the camera. Maybe we could have a camera on rails that, as you moved through it, it would always have the best viewpoint for the player.
Craddock: You've talked a lot about being inspired by classic arcade games such as Pac-Man. What was your history with Gauntlet?
Romero: I was mainly a fan of the console version, the 1985 version by Ed Logg. That was just awesome. I loved it like crazy. I thought, After all these years of maze games--and maze games were a genre for a long time--here's a maze game that looks better than all these other ones. It was also the beginning of the bullet sponge: Here's a crazy amount of enemies, and you're just taking hits because they're just running into you if you can't mow them down fast enough.
The way you'd make arcade games was with infinite levels. How many levels does Pac-Man have? It has 256, but if you didn't know, you'd think it was an infinite amount. Gauntlet was the same thing: It never runs out of levels. Diablo felt like it was influenced by that [design] as well. It was asking you: How far can you go? You won't exhaust the game's content. It will just keep on making stuff.
I played some of the other Gauntlets, but I wasn't as excited about them as I was the very first one.
“When they had this quarterly meeting, they found out that all of their titles were going to slip. The president basically said, ‘No, you are going to ship. We have to ship one game because we cannot go through Christmas without shipping one of these four games that are all close.’” -John Romero
Craddock: You and Josh Sawyer co-designed this game, which would become Gauntlet: Seven Sorrows. That pairing sounds like a dream match-up for fans who grew up playing PC games in the '90s and early 2000s: John Romero, co-designer of action hits such as Doom and Quake, and Josh Sawyer, co-designer of some of the best tactical and story-driven RPGs of all time. To your recollection, how did Josh get involved?
Romero: Gauntlet is an action game, but I wanted it to have some RPG elements and a really cool story. Josh Sawyer was working at Black Isle. He had done Icewind Dale, so I knew he was a really great designer. At that time, Midway was in discussions with possibly taking on the entire Black Isle team. It was all or nothing: They weren't going to cherry-pick people. That was early, when I started. It took several months, and that fell apart. I could talk to Josh about coming aboard.
I drove up to Newport Beach to talk to him, and told him what the situation was: I would love for him to be the lead designer, and I would love for the characters to have really cool backstories. Cinematics would be a great way to show the story. The team wouldn't be too big--I think it got up to around forty people--but the president was committed to an eighty-plus Metacritic rating, which means he'd make sure we had good production values. I asked him if he was interested, and he was.
Craddock: Was recruiting Josh and the Black Isle team your initiative, or was that wheel in motion when you joined Midway San Diego?
Romero: I actually had nothing to do with the Black Isle stuff. It was a situation that happened [before I arrived]. Midway, being in San Diego in the region where Black Isle was, had the opportunity to talk with them to see what they were interested in doing. I was already working on Gauntlet, but that was a thing that was happening: This team could come into Midway, but it may or may not be on the Gauntlet team. There was no plan of, "If we get the Black Isle guys, they could work on Gauntlet." It would probably have been them making a new game.
Romero: Josh came on alone. That team shopping themselves as a block didn't work, and there's only so long people can go without paychecks. I guess that fell apart. I might be wrong, but I remember something like that happening.
Craddock: What was it about Josh Sawyer that made you feel confident that he was the right lead designer for this type of action-heavy RPG?
Romero: He was brilliant. He was super articulate. I can't even imagine a better RPG designer. He's just the best. The amount of knowledge he has about D&D and RPGs, and his own ideas--he knows what should be in an RPG. He has a wealth of experience to draw on, much like Sandy Petersen had when we hired him for Doom: Sandy was this encyclopedia of design, and so was Josh. He was easy to work with. He stands up for stuff he cares about.
Craddock: So, at this stage, you two are working together at the same studio. You seemed to know what and who you wanted on this project, starting with Josh. How did you go about recruiting the rest of your team?
Romero: I had to choose who was going to be the lead designer, which was Josh. I had to pick who was going to be the lead programmer, which was Boris Batkin. Then I needed to put together the rest of the team so they could handle making the Xbox and PS2 versions of the game.
“They had these quarterly meetings where the project heads from every studio would meet in Chicago to present. In the front row was one of the owners of Midway. He's running the show, so whatever's happening on-stage, he'll yell out anything he wants because it's his company. He'll interrupt people, saying crazy, wacky stuff. I was just like, ‘Uh... what?’” -John Romero
The lead programmer brought in a guy named Andi Smithers to take over the advanced technology group (ATG). The guy who was at the advanced technology group before Andi was a vet in the industry named Paul Edelstein, who made a game in 1982 for Sirius Software called Wayout. It was the first super-smooth maze program [with 360-degree rotation] on 8-bit computers. He was running ATG, but [Midway] did this big layoff and we needed someone in charge of that group, so we brought Andi Smithers in. He was in charge of the PS2 port. He brought over a Ukrainian guy named Sergey [Parilov], and we brought in the legendary Mike Singleton to help work on assembly language for the PS2 version.
Craddock: Gauntlet Legends revived the Gauntlet brand by starting in arcades and then coming to consoles. Gauntlet: Dark Legacy followed, and it was very similar. What did you see as big milestones for those games? What did you want to transplant into your Gauntlet, and what did you want to leave behind?
Romero: I'd say that the big thing for me was that Gauntlet had undergone the transition to 3D. That was the big thing. It had the four, classic characters, which helped it retain its identity. Dark Legacy was in 3D, and it had those characters, but I wasn't a fan of the art style so much, but they were being kind of true to the original arcade game. The arcade version was tongue-in-cheek: It was this colorful art style, and the shapes of everything were just really fun. I thought that with a new version of Gauntlet, I could deepen the story and make it more serious, which would make it different than the stuff that had come before, especially Dark Legacy being more arcade-y.
Craddock: And Gauntlet Legends was made in an era when many console games were ports of arcade titles. Mashing buttons in arcades, mashing buttons at home.
Romero: Yeah. I wanted players to be able to use combos. I mean, Street Fighter happened! [laughs] Instead of having a million enemies coming at you, could we have more interesting locations, and puzzles, and traps, and better fighting? Then, using the characters, let's make [the story] a little more serious and go deeper than button mashing.
Craddock: Was that a difficult balance to strike? Did you think that deeper systems and storytelling would muddy Gauntlet's identity, or would they have been an extension of the franchise?
Romero: Totally an extension. Like, "Here are the characters everybody's known for decades. Can we let players know where they came from? What are their names? Can we bring in some other characters who are as deep as these guys?" Let's show that they're not the only four badasses. There's a world where they came from that has a bunch of cool characters. Let's flesh the whole thing out: Instead of four characters, let's make a world.
Craddock: What was the dynamic between you and Josh as far as leadership and decision making?
Romero: I would say, "Here's the high-level stuff. Here's what I think is important about the game. You, Josh, have a design team made up of mostly senior designers who worked on a bunch of other Midway games. You get the designers to implement your vision, which hopefully lines up with my high-level [view] that will make this Gauntlet different than previous Gauntlets."
I think that was also the pitch that got him excited: Why would he want to work on Gauntlet when he creates new IP all the time? Well, this is a chance to basically make a new IP based off a classic IP, kind of reinvent it, because I already had approval from the president at Midway that I could totally do this. The president never said, "You need to retain everything from the original Gauntlet and Dark Legacy." It was, "Make it as cool as you can. Do whatever." I said, "Can I invent the world and other characters?" He said, "Yeah, do whatever you can to make it a great property."
For [my part], I just needed to make sure Josh was doing that, and make sure the process is going well. I got to see what a console development process was like, and also solve issues that could come up. There were a lot of things for me to do on the project that weren't in the realm of me leading design. Josh would lead design, and there would be senior systems designers, and level designers, and we were using RenderWare, so it was a whole new tool chain. There was a lot of stuff to do, but I never had a fight with Josh about anything. That [leading design] was what he got to do, so there wasn't any stepping on toes. I never told him he couldn't do anything.
Craddock: That sounds like an ideal project director: Giving your lead designer a long lead.
Romero: The thing is, he's good. Josh was doing exactly what I was hoping he would do. Another interesting thing was, when I got there, the art director, Michael Murphy had already worked on Daikatana and Anachronox at Ion Storm. It was like, "Holy crap! You're here!" and he's like, "Holy crap! John!" I was already friends with the art director before I got there. He was great. Michael Murphy is a great art director, and he got some amazing artists. It was a really good team.
Craddock: You and Josh left before Seven Sorrows shipped, and Josh mentioned in our interviews that things weren't going well, which was why he left. From your perspective, when did the job veer off the tracks?
Romero: Midway had several studios they had acquired. There was Paradox Studios in LA, Ratbag Games in Australia, Inevitable in Austin doing Area 51, the John Woo Stranglehold team in Austin, I think. There was a team in Seattle doing the Suffering. They had lots of teams working on stuff, so there was lots of third-party activity going on. Midway had 800 people at the time. It was a big company; they had been around for decades.
They had these quarterly meetings where the project heads from every studio would meet in Chicago to present their [works in progress]. I went to one of these things, and in the front row was one of the owners of Midway. Their biggest shareholder, I think. He was on the board of directors, obviously. And he was nuts. He's running the show, so whatever's happening on-stage, he'll yell out anything he wants because it's his company. He'll interrupt people, saying crazy, wacky stuff. I was just like, "Uh... what?" I thought Midway Chicago was more managed [than San Diego]. There was a brain trust—Jack Tramiel, Sal, George, and the guys who had been at the company for decades--that made sure quality stayed high and they knew how to make decisions with cool heads. And then here's this guy who could change the course of the company because he went nuts [in these
meetings]. I thought, Whoa. This could disrupt all kinds of stuff.
Midway had a president who was on a contract, a two-year contract or something. So he's not permanent, but this guy is because he's on the board. I think I may have said hi to him, but I really didn't interface much with him. I just knew, Okay, the [contract] president isn't really the guy. It's this guy.
Craddock: Did this executive interfere with production on Gauntlet, or did something else come up?
Romero: As projects went along, we had a shitload of scheduling to do. Unfortunately, Gauntlet was waterfall scheduled. That means it took hours to revise the entire schedule. It was ridiculous. With the story we wanted, there were a lot of cinematics. Midway spent half a million dollars on cinematics to string the chapters together and tell this really cool story. That was all outsourced. The game is happening, and we're going through 2004. Around the beginning of 2005, just going by the progress we were making, we knew we weren't going to make Christmas. We thought we would make the first quarter of 2006.
Craddock: Why was that?
Romero: We were running into some issues with the way RenderWare Studio worked. By default, it's connected to a source control program called Alienbrain. Alienbrain was neat, but it wasn't a fast solution for source control. Every time you would move something in RenderWare Studio, it would throw [data] packets over the network to Alienbrain's source control server, and it was flooding the network with packets. What would you want to store forever the location of some object you're dragging? You didn't commit the data into a source control. You're just moving it, but it's being sent over the network, so it's clogging the network.
As we got more and more designers, it was taking too long to make levels, because the network was slowing down. Then we had to try what was actually making things go slow, and we found out it was RenderWare Studios, which we couldn't change. We couldn't rewrite RenderWare Studio, and this was after Burnout 3, which was made with RenderWare, so everybody wanted to license that shit. Everybody wanted to use that. We got it.
What we had to do was rewrite the interface: Instead of going to Alienbrain, we sent RenderWare to Perforce. We had to invest in Perforce, which sped everything up, but it still wasn't as fast as it should have been because of how RenderWare worked. So, that's how we projected it would take us longer to ship this game.
Around the second quarter [of 2005], we were waiting to find out our development velocity. We found out, no, we're not wrapping up to ship by Christmas. It's going to be quarter one, at our best guess. Romero: We told them during a quarterly meeting, "It looks like we're going to be slipping."
Craddock: How did Midway take the news?
Romero: When they had this quarterly meeting, they found out that all of their titles were going to slip. They had four games that were supposed to ship by Christmas. We were one of those, and all of those were going to slip. The president wasn't going to tell the board of directors this. He basically said, "No, you are going to ship. We have to ship one game because we cannot go through Christmas without shipping one of these four games that are all close."
The problem was that the other studios that were slipping were owned by Midway, but they weren't internal. Midway San Diego was internal, so they could control it. They said, "We're going to send a couple guys down who chop games up and ship them. We know you don't want to be around for your game to get hacked to pieces, so we're going to replace the studio head." They also replaced me, the project manager, with someone else.
At that point, I had migrated from project manager to project director because they wanted me on another game besides Gauntlet. They'd started a game called Career Criminal, and I needed to help direct that title as well. I took two weeks to make the decision to make this switch, because they had this project that had been green-lit [by executives], it's ready to go, "Please, we need you as a creative director." I said, "I came here for Gauntlet," but I said, "Okay, I'll become creative director."
So, I didn't have project manager control on Gauntlet, but nobody took over project management. I just switched roles because they needed me to focus on this other things. Not that I took much focus off of Gauntlet, because this other project didn't really go anywhere for a while.
Craddock: In other words, they knew you wouldn't want to be around to see Gauntlet get hacked into bits, and yet...
Romero: They sent these two guys down. I got to meet them and hang out with them. the president of Midway would not get his contract renewed if he didn't ship something by Christmas, so he needed to save his ass. Gauntlet was going to be the thing that got chopped up. [Mortal Kombat] Shaolin Monks was one, being done by Paradox; the Stranglehold game wasn't done, either. Then Matt Booty comes over from Chicago and basically said, "Thanks for being here working on Gauntlet. We're going to take this game in a different direction so we can ship by Christmas. It's going to be reduced in size and you're not going to like it, so we're going to let you go, and we're going to let [the studio director] go."
I said, "Well, thanks for having me. It's been a great team." I wasn't pissed off. It was just: Shit happens. They gave me three months’ severance.
Craddock: So it was an amicable parting.
Romero: Oh, yeah, it totally was. I just knew they were going to hack the game apart, and that wasn't going to be a cool time. About two weeks after I left, Josh said, "Fuck this," and he quit.
Craddock: I don't know how much you played Seven Sorrows after it came out--
Romero: As soon as I saw the big piece of cheese on the screen, I knew what happened on that game.
Craddock: According to Wikipedia, two characters created by you and Josh, Lancer and Tragedian, were removed. There's no source for that, so I thought I'd ask about it. Were they part of the game that got chopped out?
Romero: Definitely. I didn't come up with any of the characters, though. Josh came up with everything. Tragedian was a character in a play, I think. It's almost like in Final Fantasy 6 where you're on the airship, and your character's attacks are playing cards. Tragedian was a character who was an actor in tragedies, so he had his own way of fighting. Josh came up with a lot of really cool characters, and he was one of them.
It was totally a case of, "There's too much shit, and fewer things to balance means we ship faster, so chop them out." I wasn't there for any of that, but I heard about it. I believe they hacked the game to one-third the size we were going to make it, which meant that all or almost all of those cinematics were gone. Half a million dollars of cinematics just thrown away.
Craddock: This sounds like a microcosm of what went wrong at Midway: Management. It's too bad, because they made some of my favorite games.
Romero: I never got into politics at Midway because I don't do politics anywhere. I don't really give a shit. I just make games. But you can't avoid hearing things. At Midway, I couldn't figure out how they even ran that company with that one guy saying crazy shit all the time. Midway San Diego had very important executive functions that they didn't have in Chicago, so it almost seemed to be a fight over [which branch] would handle what.
“The amount of knowledge Josh has about D&D and RPGs, and his own ideas--he knows what should be in an RPG.” -John Romero
I remember mentioning, "What about the eighty-percent Metacritic [aggregate score] you wanted?" They said, "Well, production cost too much money. Metacritic ratings cost too much money. We have to ship this thing." They never told me that if it didn't ship by Christmas 2005, it would be shipped out in a shit box, but that's what they did. At E3 2005, they had a massive Gauntlet banner that went over an underpass that you had to walk to [in order to get to the Convention Center]. It was huge. I thought, This is kickass promotion. They had an amazing creative services department in San Diego that could do anything, but they screwed up this opportunity to reinvent their franchise.
It ended up being a more action-oriented version of what Gauntlet already was. They just chopped it up and made it even crazier. As soon as I saw the cheese when I played the game, I said, "What? We never would have done that? This is going directly back to this arcade-y shit. What a shame."
Imagine Black Isle making a Gauntlet, but with action and a different camera system, and less deep-RPG stuff, but with more story. It would have been really cool.