Back to the past. Part 4.1. id Software. Thorny path. Interview with Sandy Petersen. From Doom to Age of Empires.

The first series of a sub-cycle of three interviews. Some of the questions will be repeated in the next two interviews, while others are relevant to the specific interviewee.

Brief information about the main characters:

2000. Photo from MobyGames.

Sandy Petersen — game designer known, in addition to level design within id Software, for board games and titles related in one way or another to the works of H. P. Lovecraft. He also had a significant influence on the development of the Age of Empires series.

Q. How did you become interested in games? Do you play games yourself?

A. I’ve always loved games. I used to stay in from recess in grammar school to play Clue or Stratego (note: both games are board games, the second one — a strategy series that has been around since the 1970s). I still play games as often as I can.

Q. D&D. Basically the fact, that it was a part of further inspiration for original Doom’s development, been covered in book, made by Kushner David back in early 00’s. Aside from the general idea of a hellish invasion, are there other details about that game that directly inspired Doom? Also what do you remember of your personal D&D experience, if you had one, even if it doesn’t directly apply to Doom?

A. I am not sure what the question is here. Dungeons & Dragons isn’t particularly about a hellish invasion though that is something that can happen in D&D. I do know that the two Carmacks, John Romero, and Tom Hall played D&D back in Louisiana, but they weren’t still doing it in Texas when I showed up. I started playing D&D in 1974, the year it was published. One of the features of D&D is that you are supposed to make complex maps full of tricks and traps and monsters and this obviously was used in Doom extensively. So yeah, Doom was in part based on the old dungeon maps we used to play on. My maps probably showed more of the trick/trap/monster D&D background. John Romero fully bought into the shoot-em-up idea and was more keen on deathmatch rather than traps and monsters. That was a good thing, because it meant Doom had a good mix of designers. John would add more traps and tricks to imitate me, and I would put bits into my maps to make it better for deathmatch.

Q. Jayextee asking: “I’d love to know what of his tabletop / RPG background he brought into DOOM level design. In other words… what commonalities and / or overlaps were there, what differences and / or new challenges and, to a further end, whether he designed with replayability in mind, or more “one and done” in the vein of a social RPG campaign.”.

A. I designed roleplaying games and boardgames before working on Doom, and of course I would steal ideas from those tabletop game and use them at every opportunity. I didn’t think of Doom as a “campaign” though, and designed each level so that they could be beaten by someone starting with just a pistol. I also tried to give “hints” about something going on behind the scenes to creep out the players. Evidence that there were wars between different demon factions, that sort of thing. As one example, on one map you can see into an area (that you can’t reach) which has the corpses of Hell Barons dangling, as though Something was torturing them to death — perhaps the Cyberdemon or the Spider Mastermind? Who knows? The whole point was to pique the imagination.

Q. Were you involved in board game development before or after joining the gaming industry? If before, is it possible to tell us about your early board games?

A. I was making my own games from the age of 14 on, but mostly I played other games. In 1981 I was hired onto Chaosium Inc. where I did roleplaying games & supplements, plus I worked on the Arkham Horror boardgame. From 1988 to 2012 I did digital games exclusively, and in 2012 I came back to board games which is once more where I am.

My earliest boardgames and to an extent even my more recent games are all attempts to fill gaps. That is, I create games when I don’t have an existing game covering the desired subject. For instance, when I was about 16 I read the science fiction story Gottlos about a powerful robot tank in a futuristic war. I really liked it so I created a tactical SF game featuring the Gottlos tank. Later in the 1970s, Steve Jackson game released their game Ogre (note: later, adaptation for IBM PC, 1986, and other systems arrived), which covered exactly the same ground as my game. It even had a cybernetic super-tank like mine. So I never went back and tried to “fix” my old Gottlos game, because Steve Jackson had covered it.

[embed]

Q. How did you get a job at MPS Labs / Microprose? Is it possible to tell how it was organized / structured from the inside?

A. I had told my boss at Chaosium that I was looking for a new job, because the pay was too low and Chaosium didn’t seem to be going anywhere, and I had 4 kids living in a tiny two bedroom apartment in the expensive wastelands of California. So with his blessing I sent my resume around at GenCon in 1988. Lawrence Schick, recruiting for MicroProse, coincidentally came by looking to hire me that same GenCon, so I was a pretty easy sell.

The way it worked originally was that Sid Meier worked with Bruce Shelly to create amazing games. The company leadership got the idea that the best way to make games was to combine a designer with a programmer, so they basically hired one designer per lead programmer. Lawrence Schick, Arnold Hendrick, Bruce Shelly, Jeff Briggs, and me were the original design team, and then each of us would be given to a programmer to create a game. Except that initially Sid ONLY worked with Bruce Shelly (his favorite). Later, after Bruce left MPS Labs, Sid “adopted” Jeff Briggs instead. Other designers soon joined on, such as Doug Kaufman and Paul Murphy. And more. It was an incredible pool of talent.

The company worked out well at first. Then Bill Stealey kind of went cuckoo and decided he wanted to have a stand-up arcade (note: vertical arcade machines) division. This was approximately 1990, when stand-up arcades were already kind of drooping. So suddenly half of the profits (100% generated by MPS Labs) were going to support the arcade division — which didn’t get their own designer because Bill Stealey was going to do that job. Ha. Then because MPS Labs was starting to slow down, Bill hired an astonishingly incompetent VP to run MPS Labs. This guy never even learned our names. Anyway, that new VP decided that HE wanted to do sexy adventure games like Leisure Suit Larry so he founded a whole new division to do the adventure games. Some of the designers moved over to the new adventure game division because the VP offered them lots of money (I don’t blame them!) Those guys were also excited, not by the sex adventure games, but because they were building a new cool engine to make lots of interesting games, such as DragonSphere (1994), and Phantom of the Opera. The sex adventure was just an obstacle to get past to get to the good games. But the sex adventure (Rex Nebular (1992)) was a complete flop of course.

So by 1993 we had MPS Labs trying to support two entire other company divisions all by itself because those two divisions were commercial failures. Something had to give.

Q. Is it possible to describe to our readers something about the experience of developing Sid Meier’s Pirates! (1987) for Atari ST?

A. I joined MPS Labs in 1988 so Sid Meier’s Pirates! (1987) for the Atari ST was a little before my time. I did own an Atari ST however, and loved Pirates! My first project at MPS Labs was to convert Pirates! (1988) over to the Apple MacIntosh. Part of that process was to take Sid’s horrifying spaghetti code and turn it into real programming. I didn’t do the coding, but I made sure that the final project felt like Pirates! in every single detail. It was super fun because I didn’t have to actually design anything. Instead I just played the game a million times and would explain to the programmer, “Hey, the land battles are moving too slow. We need to speed up the troops.” Or whatever. Sid took zero interest in the project — he was working on his next new thing, and rightly so.

Q. Did you do anything else during the development of Sword of the Samurai (1989) and Command H.Q. (1990) besides documentation and coordination / organization? Can you tell a few interesting / not well known facts about those two?

A. I did a lot of testing and suggestions with respect to Sword of the Samurai, because it was created by my best friend Lawrence Schick whom I spent a lot of time with, and also because SotS was modeled after the Pirates! game which I loved. But Lawrence deserves the praise. I remember him complaining that since the game used a top-down view, all the warriors looked like turtles (their helmets). I also remember Bill Stealey getting upset when he learned that the ninjas were bad guys in the game. “Aren’t ninjas good guys?” he asked. Lawrence had to explain that ninjas were professional assassins, so no, they are not heroic. For his next game, Lawrence wanted do a game he called Castle Ghast, which would be another Pirates / SotS style game, but in a fantasy world with dragons and goblins. I think he wanted to do this because the creatures wouldn’t all look similar.

I was the developer of Command HQ, working closely with Dan Bunton (later Danielle Berry). I remember Dan wanting the game to be primarily played head-to-head. Which never happened, because I guess people weren’t ready for it. Surprisingly, the next year (1993) I was at Id Software and we released Doom which DID get played head-to-head a lot and in fact sparked a revolution in gaming where now almost all games have a player vs. player mode. I guess Command HQ was ahead of its time. It was also ahead in another way, which was that it was the first true Real Time Strategy game. It had production, exploration, real-time movement, everything. I knew it was special but later when I was helping create the Age of Empires series I thought back to how close we were.

Q. Similar question about Lightspeed (1990), Hyperspeed (1991), Darklands (1992), where you are credited as a designer. How much of each of these games were you responsible for? Does there some interesting stories related to those took a place back in the day?

A. I was 100% responsible for the design, both good & bad, of Lightspeed & Hyperspeed. We were being pressured by the wicked VP (mentioned previously) to get the game out as fast as possible. The original plan was to have four star clusters. But we released Lightspeed with just the first two star clusters (one “beginner” an d one “intermediate”). Then we immediately started working on Hyperspeed, which had ALL of Lightspeed PLUS two more “advanced” clusters. But we got Lightspeed out on time. Nowadays I guess we’d just add the other clusters in patches, or as in-game purchases.

Darklands was Arnold Hendrick’s game. He wanted me to come on board and make the game more fun, so that was my task. I wrote up quests and encounters, and puzzles and all manner of things. But the VP forced us to release it way way too soon. In fact, at the time of release, it could not be played for more than an hour before it locked into a hard blue-screen crash. The head of quality control refused to sign his name on the product. After its release, a lot of effort was spent to fix the bugs and I guess it was a decent game but man it left a bad taste in my mouth. It was a death march — the lead programmer, lead artist, and Arnold all feuded. The programmer & artist literally wo uldn’t speak to one another. What a nightmare. It was the worst development experience of my life, and I blame the VP entirely.

[embed]

Q. Was Darklands in development before you joined the company or during your stay? Is it possible to tell something about Arnold Hendrick, what kind of person was he? Why did you decide to leave MPS Labs / Microprose after Darklands?

A. It started after I began at MicroProse. Arnold was a very nuts-and-bolts type of designer who was keen on math and knowing the background. He had the interesting flaw that he really wasn’t sure what was fun or not fun, so his games always modeled EVERYTHING about the subject because he knew the fun was in there somewhere. For instance his awesome game M1 Tank Platoon (1989) included a screen for the loader, because maybe loading was fun. Any other designer would probably have ignored the loader but not Arnold — he had it all. I remember when I started to work on Darklands with him, I asked him about medieval German mines, because I was going to have adventures in them. He started explaining that medieval mines were stock corporations, and stuff like that. I was going in a total different direction — with kobolds, golems, and mine monsters. Not stock corporations! It was pretty funny. Arnold taught me a lot about design and I miss working with him.

I left MPS Labs under duress. Basically I didn’t immediately have a new project after Darklands, though I was acting as support on five other projects (spreading me thin). And Microprose was shedding money, with the stress of the arcade & adventure game divisions failing, so they started cutting personnel, and I was one of the guys that was cut. I guess with me being spread over 5 projects, I wasn’t essential to any one of them. It turned out to be great for me though because I went to Id Software.

Q. What attracted you to id Software? What parts of id’s company culture do you think had the biggest impact, and do you have any advice for aspiring developers about its advantages and pitfalls.

A. I had already played Castle Wolfenstein (1984) and loved it. A friend from MPS Labs told me that id was looking for a designer and I contacted them and got an interview. From there I was hired and the rest is history I guess. Id Software had great and terrible aspects to it.

One of its great features was that we were initially small and very friendly — tightly-knit. I felt able to walk into anyone’s office and discuss topics without restraint. This changed later.

One of its worst features was its inability to communicate effectively — people would talk and make decisions (which is fine) but then wouldn’t think to inform other people (like me) about those decisions which were often relevant to our jobs. I once designed an entire level because I wasn’t told that John Romero was going to design that level. It was all thrown out.

Q. Without belittling the importance of previous projects, released both before and in 1993. When you were just start thinking about the Doom as in overall, did you have a premonition that this title should or, well, will, radically change the game industry and create, in fact, a new genre direction itself as we know it now? Or was Doom just the result of a sort of “bold experiment”?

A. We knew we were onto something amazing. We did not realize HOW amazing because we didn’t predict the importance of player-to-player. We did know this was going to have a big impact on the gaming world. I had the privilege of going back and visiting MPS Labs after working at id and letting my former co-workers see the pre-release version of DOOM, which blew them away. Their reaction told me all I needed to kno w.

Q. 1993. Back in that year, did the computer at home has a significant matter for you? Or was everything revolve around work? If first, then what was it’s technical specs?

A. I did not work much from home, though I owned a home computer. It was mostly just used for fun and games. I went to work and did all my stuff there. At work I had an IBM 386 which later upgraded to an IBM 486 I think in 1994.

Q . There is an idea exist, that Doom would not be what it is, if it was not accompanied by a certain degree of surrealism, especially when it goes for the case of the environment and music. It, in particular, distinguishes it from conventional military-style shooters and, no less, conventional horror works, that have more or less action elements. Doom 64 specifically was a major example. What do you think about this?

A. We put zero effort into making Doom “realistic”. We wanted an emotional, action experience, and that’s what we got. Actually if you look at Wolfenstein it also wasn’t too realistic. The whole goal was to make it wild and weird and over the top. “Surreal” is a good way to describe i t.

[embed]

Q. Does original Doom has any influence by the Alien (1979) movie series specifically, or, perhaps, some other, not so well known one? Was there any chances that the main character supposed to be a woman, which is, basically, would be unsual choise by 90’s, especially early 90’s, standarts? A recent update of Doom featured previously unknown material, which also contained, it seems, part of early, unpublished builds.

A. Of course we were influenced by Aliens, not to mention Outland (1981), Terminator (1984), Star Wars (1977), etc. Everything went into our mental blenders. We never considered making the main character a woman. To us a “space marine” just obviously would b e a dude.

Q. Amaruq Wulfe asking: “Given your background as a mormon, was there any reason why didn’t you depict hell as described in the relevant texts instead used Lovecraft as the source material?”.

A. Texts? Even the Bible doesn’t have a lot of description of Hell though it does mention a lake of fire. My own faith teaches that the ultimate Hell is Outer Darkness. Neither image makes for a fun Hell experience. I don’t recall using Lovecraft as the main influence in Doom, except maybe his ideas about clashing dimensi ons. But certainly Lovecraft’s influence is plain in Quake (1996), which is also not directly about Hell, so Lovecraftian horrors make more sense.

Q. What inf luence, if any, do you think Adrian Carmack’s art style had in the games industry? How you look on a fact, that today’s fans continued use of this style, in particular?

A. Adrian Carmack was one of the best game artists of all time. The way he went for body horror for his demons was striking and amazing — the Cyberdemon looks naked, almost like an autopsy victim, and it makes it 10 times as scary.

Q . holaareola asking: “In Doom mapping community, infighting became a cornerstone mechanic, but your E2M9 — “Fortress of Mystery” secret map was the first map in Doom to put it front and centre. Do you remember how early in development infighting was implemented?”.

A. It was an accident. John Carmack put in code so when a monster was injured, it would “get mad” at whoever injured it, so that you couldn’t shoot them with impunity. I mean, you could hit them with one free shot from behind, but then it rolled into action. Then we discovered that when monsters shot each other, they also got angry, and started to infight. Everyone at id loved this and then I abused it in E2M9 to make it as clear as possible. Later on, it got used extensively in Tricks & Traps by me, and John Romero did a level with a cyberdemon & a spider mastermind fighting. We’d watch them fight to see who won. It w as usually the cyberdemon, but not always, because sometimes the spider could keep the cyberdemon pained long enough to take it down.

Q. What probably acted as the inspiration for the more experimental like level design for Doom II (1994)? What is your favorite Doom and / or Quake map you worked on?

A. They came from everywhere. I would think “I want a cityscape” which is where Downtown came from. Then I had a ton of ideas for environmental dangers and put them all into Tricks & Traps. Then I had a nightmare about an unending abyss, and that went into The Chasm. Etc.

Q. Futuretime23 asking: “What does he think was his / not his worst map in Doom I / II / Quake and what lessons could one learn from his / not his perceived mistakes (level design wise)?”.

A. That’s not for me to say. I will say that something I learned is that a map which one person hates can sometimes be a favorite of someone else, so the lesson from this is to mix up what you do — don’t have all the levels be the same, so everyone will have things to lo ve.

Q. Can you tell a bit about how did you design the monsters for Doom and Quake? Where did you get the inspiration and images?

A. I did not design the monsters — that was done by Adrian Carmack under the influence of the team. He did a fabulous job. A lot of the inspiration for Quake came from my book “Petersen’s Field Guide to Monsters” (1987) which I gifted Adrian a copy of. Many monster designs were drawn straight from it (Shub-Niggurath, the Shambler, Chthon, the Spawn, etc.).

Q. GreenEyesMan asking: “Why is the design and fight against Shub-Niggurath in Quake so flawed? It is, after all, one of the most powerful chthonic creatures. In Quake 1.5 the fight with this boss is many times more epic.”.

A. You are preaching to the choir, pal. I argued the same thing but I did not do the boss level so I wash my hands of it. I did argue for the Shub-Niggurath model, but thought it would be more … ac tive.

[embed]

Q. GreenEyesMan and Dimouse asking (first and second questions respectively):

“Why does Quake have such an unusual design: first one industrial level, and then several inside some castles?”

“Did your work at Chaosium Inc. and Lovecraft board games influence the design of Quake, Doom and other games you participated in?”.

A. We ended up with this whole bunch of unrelated levels and had to put them into some kind of order. It was my job to order the levels but I also wanted to mix up the levels among the designe rs, so you wouldn’t do a bunch of Sandy levels, then a whole slew of John levels in a row. I also wanted the levels to get gradually more esoteric and crazy-looking. In the en d I did the best I could with what I hand, keeping in mind that there was no one organizing the whole thing from on top — no project leader.

Q. Can you tell a bit about console ports of the main titles you have been involved into? Specifically mostly interested here in not well known detailes about Doom for Atari Jaguar, Sega 32X and PC-98, Hexen: Beyond Heretic and Quake for Sega Saturn. Still, few fact about the rest might be fine to hear too.

A. We did the Jaguar because John Carmack really liked programming for it and he insisted. My job was to shrink all the Doom levels in size so they would fit on the Jaguar. It was an unholy nightmare but I was able to get it all done in a couple of weeks. We used the same levels for the Sega. I did not do the conversions for the Sega Saturn.

Q. Immorpher asking: “Doom 64 was intended to be a N64 launch title back in 1996. However magazines reported that the initial set of levels for Doom 64 were r ejected by id Software and the game was delayed until 1997. As a level designer at id Software around that time, were you part of the team who evaluated Midway’s Doom 64 levels? If so, do you remember the issues these levels had which caused them to be rejected and perhaps any suggestions for improvement?”.

A. I was part of the team that rejected them, but sadly I remember very little about it. I remember that they were slow-moving and not very f un.

[embed]

Q. Ha ve you had any urge to follow in Romero’s footsteps by coming full circle back to Doom and perhaps create and release something new like Romero did with Sigil (2019)?

A. Have no interest in going back to Doom, but I suppose I could be bribed to do so. Frankly, the new breeds of level designers are amazing.

Q. Why did you decide to move to Ensemble Studios after Quake? You have participated in the creation of one of the most popular RTS series — Age of Empires (1997). What is your opinion about other well-known and popular (now or at one time) competitors in the genre — Command & Conquer (1995), WarCraft (1994), StarCraft (1998), Total Annihilation (1997), Homeworld (1999), Company of Heroes (2006)? In the first Age of Empires you are listed as being responsible for scripts and QA. How did that manifest itself?

A. Quake destroyed id Software. After Quake, we lost John Romero, Shawn Green, myself, Mike Abrash, and American McGee. That gutted the company and it was never the same. I was just part of the fallout. Later they also lost Adrian Carmack, and in fact had lawsuits about it.

I was always interested in other RTS games which we viewed as essential to understand if we were to advance the Ensemble Studios skills. This is in direct contrast to id Software which rarely, if ever, played other company’s games and the culture was to sneer at everyone else because we were the best. I tried a couple of times to talk about what Unreal or Valve was doing but got hooted down because Not Invented Here.

Ensemble Studios, refreshingly, was keen on learning from others. I joined Ensemble a few months before Age of Empires was due to be released. I wrote scripts for the campaigns and scenarios, and I helped run playtests. I didn’t do much more because I was supposed to be working on a magic-based RTS game to be called Sorceress. It got killed because literally all my personnel except one programmer & one artist were taken off it to work on Age 2. So instead I proposed that I take my tiny team and make Rise of Rome instead.

Q. What was your work look alike as a designer during the Age of Empires & Age of Mythology related titles during developing stage?

A. In the early stages of each Age game, I was the guy busy making the expansion to the previous game (I was lead on Rise of Rome, The Conquerors, and The Warchiefs). When I finished the expansion I’d be seconded to the main project. By that time they had the game engine done so my task became to help script the campaigns & scenarios, to balance the game, and to create the various historical parts, such as coming up with the civ unique units & techs.

Q. There is information that Ensemble had an idea to make the fourth and fifth parts dedicated to the 20th century and the future, respectively, which is reflected on one of the pages of the book from the collector’s edition of Age of Empires III (2005). How true is this and were any ideas for the fifth part implemented in Halo Wars (2009)?

A. We did in fact consider that. We also thought about having 19th century wars that would end with WWI, then have WWI & WW2 be a game. In the end one of the obstacles we faced was simple. In Age 1 you were building up a new society from the stone age. In Age 2 you were recovering from the fall of Rome. In Age 3 you were colonizing the New World. In ever y case, it made sense to start from scratch and build up your empire. But in the 19th and 20th centuries, starting from a single town center makes no sense.

Q. Subquestion. Have you been involved / what was your role during the development of the re-releases — Age of Empires II: HD Edition (2019), Halo Wars: Definitive Edition (2017), Age of Empires: Definitive Edition (2013), Age of Empires II: Definitive Edition (2018)?

A. I did nothing directly for any of the post-2009 versions of Ensemble Studios games. I’m sure they have lots of my creative work inside those products though, of which I’m proud.

Q. taufan99 asking: “Given your love for tabletop RPG’s and RPG video games and the fact that you’ve worked for both Doom and Quake (in addition to countless RPG projects, tabletop or video games), do you have any thoughts on the so-called “immersive sim” subgenre of video games, which often mixes RPG and FPS games to a large extent (e.g. System Shock (1994), Thief: The Dark Project (1998), Deus Ex (2000), BioShock (2007))?”.

A. I always felt one of the first and best was H alf-life but yeah System Shock counts too. I thought these were terrific ideas, and more could be done with them. I do remember other people at id Software (not all of them) mocking the idea of Half-Life because why would you want an immersive story in a shooter? I guess history has shown who was right.

Q. A ny final words you would like to say to our readers?

A. Game needs to be fun even if you don’t win. The victory is ma ybe 10 seconds at the game’s end. The play is an hour or more of enjoyment. Don’t make your players have to earn the right to have fun — make your games great from the get-go. It’s okay to pay for a game, even if it’s on your phone. I agree that micro-transactions are an invention straight from Hell.

[embed]