RSS

X-COM

X-COM seemed to come out of nowhere. Its release was not preceded by an enormous marketing campaign with an enormous amount of hype. It had no video demo playing in the front window of Babbages, it wasn’t advertised twelve months in advance on glossy foldout magazine inserts, it had no flashing point-of-purchase kiosks. It didn’t come in a box designed by origamists from the school of abstract expressionism. It featured no full-motion video starring the best TV actors of the 80s; it had no voice-overs. It offered neither Super VGA graphics, nor General MIDI support. It wasn’t Doom-like, Myst-like, or otherwise like a hit game from the previous season; it didn’t steal the best features from several other successful games. It wasn’t even on a CD-ROM!

In short, if you plugged X-COM’s variables into the “success formula” currently in use by the majority of large game companies, you’d come up with a big, fat goose egg. According to the prevailing wisdom, there’s no way X-COM could survive in today’s gaming marketplace. And yet it sold and sold, and gamers played on and on.

— Chris Lombardi, writing in the April 1995 issue of Computer Gaming World

In the early days of game development, there existed little to no separation between the roles of game programmer and game designer. Those stalwart pioneers who programmed the games they themselves designed could be grouped into two broad categories, depending on the side from which they entered the field. There were the technologists, who were fascinated first and foremost with the inner workings of computers, and chose games as the most challenging, creatively satisfying type of software to which they could apply their talents. And then there were those who loved games themselves above all else, and learned to program computers strictly in order to make better, more exciting ones than could be implemented using only paper, cardboard, and the players’ imaginations. Julian Gollop, the mastermind behind the legendary original X-COM, fell most definitely into this latter category. He turned to the computer only when the games he wanted to make left him no other choice.

Growing up in the English county of Essex, Julian and his younger brother Nick lived surrounded by games, courtesy of their father. “Every Christmas, we didn’t watch TV, we’d play games endlessly,” Julian says. From Cluedo, they progressed to Escape from Colditz, then on to the likes of Sniper! and Squad Leader.

Julian turned fifteen in 1980, the year that the Sinclair ZX80 arrived to set off a microcomputer fever all across Britain, but he was initially immune to the affliction. Unimpressed by the simplistic games he saw being implemented on those early machines, which often had as little as 1 K of memory, he started making his own designs to be played the old-fashioned way, face-to-face around a tabletop. It was only when he hit a wall of complexity with one of them that he reassessed the potential of computers.

The game in question was called Time Lords; as the name would imply, it was based on the Doctor Who television serials. It asked two to five players to travel through time and space and alter the course of history to their advantage, but grew so complex that it came to require an additional person to serve in the less-than-rewarding capacity of referee.

By this point, it was 1982, and a friend of Julian’s named Andy Greene had acquired one of the first BBC Micros. Its relatively cavernous 32 K of memory opened up the possibility of using the computer as a referee instead of a bored human. Greene coded up the program in BASIC, staying faithful to Julian’s board game to the extent of demanding that players leave the room when it wasn’t their turn, so as not to see anything they weren’t supposed to of their opponents’ actions. The owner of the tabletop-games store where Julian shopped was so impressed with the result that he founded a new company, Red Shift Games, in order to publish it. They all traveled to computer fairs together, carrying copies of the computerized Time Lords packaged in Ziploc baggies. The game didn’t take the world by storm — Personal Computer News, one of the few publications to review it, pronounced it a “bored game” instead of a board game — but it was a start.

The two friends next made Islandia, another multiplayer strategy game of a similar stripe. In the meantime, Julian acquired a Sinclair Spectrum, the cheap and cheerful little machine destined to drive British computer gaming for the next half-decade. Having now a strong motivation to learn to program it, Julian did just that. His first self-coded game, and his first on the Spectrum, appeared in 1984 in the form of Nebula, a conquer-the-galaxy exercise that for the first time offered a computer opponent to play against.

The artificial intelligence disappeared again from his next game, but it mattered not at all. Rebelstar Raiders was the prototype for Julian Gollop’s most famous work. In contrast to the big-picture strategy of his earlier games, it focused on individual soldiers in conflict with one another in a Starship Troopers-like science-fictional milieu. Still, it was very much based on the board games he loved; there was a lot of Sniper! and Squad Leader in its turn-based design. Despite being such a cerebral game, despite being one that you couldn’t even play without a mate to hand, it attracted considerable attention. Red Shift faded out of existence shortly thereafter as its owner lost interest in the endeavor, but Rebelstar Raiders had already made Julian’s reputation, such that other publishers were now knocking at his door.

Rebelstar Raiders, the first of Julian Gollop’s turn-based tactical-combat games. Ten years later, the approach would culminate in X-COM.

It must have been a thrill for Julian Gollop the board-game fanatic when Games Workshop, the leading British publisher of hobbyist tabletop games, signed him to make a computer game for their new — if ultimately brief-lived — digital division. Chaos, a spell-slinging fantasy free-for-all ironically based to some extent on a Games Workshop board game known as Warlock — not that Julian told them that! — didn’t sell as well as Rebelstar Raiders, although it has since become something of a cult classic.

So, understandably, Julian went where the market was. Between 1986 and 1988, he produced three more iterations on the Rebelstar Raiders concept, each boasting computer opponents as well as multiplayer options and each elaborating further upon the foundation of its predecessor. Game designers are a bit like authors in some ways. Some authors — like, say, Margaret Atwood — try their hands at a wide variety of genres and approaches, while others — like, say, John Cheever — compulsively sift through the same material in search of new nuggets of insight. Julian became, in the minds of the British public at least, an example of the Cheever type of designer. “It could be said by the cruelest among us that Julian has only ever written one game,” wrote the magazine New Computer Express in 1990, “but has released various substantially enhanced versions of it over the years.”

Of those enhanced versions, Julian published Rebelstar and Rebelstar 2: Alien Encounter through Firebird as a lone-wolf developer, then published Laser Squad through a small outfit known as Blaze Software. Before he made this last game, he founded a company called Target Games — soon to be renamed to the less generic Mythos Games — with his father as silent partner and his brother Nick in an active role; the latter had by now become an accomplished programmer in his own right, in fact surpassing Julian’s talents in that area. In 1990, the brothers made the Chaos sequel Lords of Chaos together in order to prove to the likes of New Computer Express that Julian was at least a two-trick pony. And then came the series of events that would lead to Julian Gollop, whose games were reasonably popular in Britain but virtually unknown elsewhere, becoming one of the acknowledged leading lights of strategy gaming all over the world.



The road to X-COM traveled through the terrain of happenstance rather than any master plan. Julian’s career-defining project started as Laser Squad 2 in spirit and even in name, the next manifestation of his ongoing obsession with small-scale, turn-based, single-unit tactics. The big leap forward this time was to be an isometric viewpoint, adding an element of depth to the battlefield. He and Nick coded a proof of concept on an Atari ST. While they were doing so, Blaze Software disappeared, yet another ephemeral entity in a volatile industry. Now, the brothers needed a new publisher for their latest game.

Both of them had been playing hours and hours of Railroad Tycoon, from the American publisher MicroProse. Knowing that MicroProse had a British branch, they decided to take their demo there first. It was a bold move in its way; as I’ve already noted, their games were popular in their sphere, but had mostly borne the imprints of smaller publishers and had mostly been sold at cheaper price points. MicroProse was a different animal entirely, carrying with it the cachet that still clung in Europe to American games, with their bigger budgets and higher production values. In their quiet English way, the Gollops were making a bid for the big leagues.

Luckily for them, MicroProse’s British office was far more than just a foreign adjunct to the American headquarters. It was a dynamic, creative place in its own right, which took advantage of the laissez-faire attitude of “Wild” Bill Stealey, MicroProse’s flamboyant fly-boy founder, to blaze its own trails. When the Gollops brought in the nascent Laser Squad 2, they were gratified to find that just about everyone at MicroProse UK already knew of them and their games. Peter Moreland, the head of development, was cautiously interested, but with plenty of caveats. For one thing, they would need to make the game on MS-DOS rather than the Atari ST in order to reach the American market. For another, a small-scale tactical-combat game alone wouldn’t be sufficient — wouldn’t be, he said, “MicroProse enough.” After making their name in the 1980s with Wild Bill’s beloved flight simulators, MicroProse was becoming at least as well known in this incipient new decade for grand-strategy games of or in the spirit of their star designer Sid Meier, like the aforementioned Railroad Tycoon and the soon-to-be-released Civilization. The emphasis here was on the “grand.” A Laser Squad 2 just wouldn’t be big enough for MicroProse.

Finally, Moreland wasn’t thrilled by all these far-future soldiers fighting battles in unknown regions of space for reasons that were abstract at best. Who could really relate to any of that? He wanted something more down to earth — literally. Maybe something to do with alien visitors in UFOs… that sort of thing. Julian nodded along, then went home to do some research and refine his proposal.

He quickly learned that he was living in the midst of a fecund period in the peculiar field of UFOlogy. In 1989, a sketchy character named Bob Lazar had given an interview for a Las Vegas television station in which he claimed to have been employed as a civilian contractor at the top-secret Nevada military base known only as Area 51. In that location, so he said, the American Air Force was actively engaged in testing fantastic technologies derived from extraterrestrial visitors. The interview would go down in history as the wellspring of a whole generation of starry-eyed conspiracy theorists, whose outlandish beliefs would soon enter the popular media zeitgeist via such vehicles as the television series The X-Files. When Julian first investigated the subject in 1991, however, UFOs and aliens were still a fairly underground obsession. Nevertheless, he took much from the early lore and legends of Area 51, such as a supposed new chemical element — called ununpentium by Lazar, elerium by the eventual game — which powered the aliens’ spaceships.

His other major source of inspiration was the 1970 British television series entitled simply UFO. In fact, his game would eventually be released as UFO: Enemy Unknown in Europe, capitalizing on the association with a show that a surprising number of people there still remembered. (I’ve chosen to use the American name of X-COM globally in this article because all subsequent games in the franchise would be known all over the world under that name; it has long since become the iconic one.) UFO the television series takes place in the then-near-future of 1980, when aliens are visiting the Earth in ever-increasing numbers, abducting humans and wreaking more and more havoc. An international organization known as SHADO (“Supreme Headquarters Alien Defence Organisation”) has been formed to combat the menace. The show follows the exploits of the SHADO operatives, complete with outlandish “futuristic” costumes and sets and gloriously cheesy special effects. Gollop lifted this basic scenario and moved it to his own near-future: to the year 1999, thus managing to nail not only his decade’s burgeoning obsession with aliens but also its unease about the looming millennium.

The game is divided into two distinct halves — so much so that each half is almost literally an entirely separate game: each unloads itself completely from memory to run a separate executable file at the point of transition, caching on the hard drive before doing so the relatively small amount of state data which its companion needs to access.

The first part that you see is the strategic level. As the general in charge of the “Extra-Terrestrial Combat Force,” or X-COM — the name was suggested by Stephen Hand and Mike Brunton, two in-house design consultants at MicroProse UK — you must hire soldiers and buy equipment for them; research new technologies, a process which comes more and more to entail reverse-engineering captured alien artifacts in order to use your enemy’s own technology against them; build new bases at strategic locations around the world, as well as improve your existing ones (you start with just one modest base); and send your aircraft out to intercept the alien craft that are swarming the Earth. In keeping with the timeless logic of computer games, the countries of the Earth have chosen to make X-COM, the planet’s one real hope for defeating the alien menace, into a resource-constrained semi-capitalist enterprise; you’ll often need to sell gadgets you’ve manufactured or stolen from the aliens in order to make ends meet, and if you fail to perform well your sponsoring countries will cut their funding.

The “Geoscape” view, where you place your bases and use them to intercept airborne alien attackers. You can find a wealth of discussion online about where best to position your first base — but naturally, most people prefer to put it in their home town. Like the ability to name your individual soldiers, the ability to start right in your own backyard forges a personal connection between the game and its player.

This half of the game was a dizzying leap into uncharted territory for the Gollop brothers. Thankfully, then, they were on very familiar ground when it came to the other half: the half that kicks in when your airborne interceptors force a UFO to land, or when you manage to catch the aliens in the act of terrorizing some poor city, or when the aliens themselves attack one of your bases. Here you find yourself in what amounts to Laser Squad 2 in form and spirit if not in name: an ultra-detailed turn-based single-unit combat simulator, the latest version of a game which Julian Gollop had already made four times before. (Or close enough to it, at any rate: X-COM, the culmination of what had begun with Rebelstar Raiders on the Spectrum, is ironically single-player only, whereas that first game had not just allowed but required two humans to play.) Although the strategic layer sounds far more complex than this tactical layer — and, indeed, it is in certain ways — it’s actually the tactical game where you spend the majority of your time, fighting battles which can consume an entire evening each.

The “Battlescape” view, where tactical combat takes place.

For all their differences, the two halves of the game do interlock in the end as two facets of a whole. Your research efforts, equipment purchases, and hiring practices in the strategic half determine the nature of the force you lead into the tactical man-against-alien battles. Less obviously but just as significantly, your primary reward for said battles proves to be the recovery of alien equipment, alien corpses, and even live alien specimens (all is fair in love and genocidal interplanetary war), which you cart back to your bases to place at the disposal of your research teams. And so the symbiotic relationship continues: your researchers use what you recover as grist for their mill, which lets you go into tougher battles with better equipment to hand, thereby to bring back still richer spoils.

The capsule description of the finished game which I’ve just provided mirrors almost perfectly the proposal which Julian Gollop delivered to MicroProse; the design would change surprisingly little in the process of development. MicroProse thought it sounded just fine as-is.



The contract which the Gollops signed with MicroProse specified that the former would be responsible for all of the design and coding, while the latter would provide the visual and audio assets. MicroProse UK did hold up their end of the bargain, but had an oddly casual attitude toward the project in general. Julian remembers their producer as “very laid back — he would come over once a month, we would go to the pub, talk about the game for a bit, and he would go home.” Otherwise, the Gollops worked largely alone after their first rush of consultations with the MicroProse mother ship had faded into the past. Time dragged on and on while they struggled with this massively complicated game, one half of which was unlike anything they had ever even contemplated before.

X-COM‘s UFOpaedia is a direct equivalent to Civilization‘s innovative Civilopedia, its most obvious single nod to Sid Meier’s equally influential but very, very different game.

As it did so, much happened in the broader world of MicroProse. On the positive side, Sid Meier’s Civilization was released at the end of 1991. But despite this and some other success stories, MicroProse’s financial foundation was growing ever more shaky, as their ambitions outran their core competencies. The company lost millions on an ill-judged attempt to enter the stand-up arcade market, then lost millions more on baroque CRPGs and flashy interactivity-lite adventure games. After an IPO that was supposed to bail them out went badly off the rails, Wild Bill Stealey sold out in June of 1993 to Spectrum Holobyte, another American publisher. The deal seemed to make sense: Spectrum Holobyte had a lot of money, thanks not least to generous venture capitalists, but a rather thin portfolio of games, while MicroProse had a lot of games both out and in the pipeline but had just about run out of money.

Spectrum Holobyte sifted carefully through their new possession’s projects in development, passing judgment on which were potential winners and which certain losers. According to Julian Gollop, Spectrum Holobyte told MicroProse UK in no uncertain terms to cancel X-COM. On the face of it, it wasn’t an unreasonable point of view to take. The Gollops had been working for almost two years by this point, and still had few concrete results to show for their efforts. It really did seem that they were hopelessly out of their depth. Luckily for them, however, Peter Moreland and others in the British office still believed in them. They nodded along with the order to bin X-COM, then quietly kept the project on the books. At this point, it didn’t cost them much of anything to do so; the art was already done, and now it was up to the Gollops to sink or swim with it.

X-COM bobbed up to the surface six months later, when the new, allegedly joint management team — Stealey would soon leave the company, feeling himself to have been thoroughly sidelined — started casting about for a game to feature in Europe in the first quarter of 1994, thereby to make the accountants happy. Peter Moreland piped up sheepishly: “You remember that UFO project you told us to cancel? Well, it’s actually still kicking around…” And so the Gollop brothers, who had been laboring under strangely little external pressure for the past 26 months or so, were now ordered to get their game done already. They managed it, just — UFO: Enemy Unknown shipped in Europe in March of 1994 — but some of the problems in the finished game definitely stem from the deadline that was so arbitrarily imposed from on high.

But if the game could have used a few more months in the oven, it nonetheless shipped in better condition than many other MicroProse games had during the recent stretch of financial difficulties. It garnered immediate rave reviews, while its sales also received a boost from another source. The first episode of The X-Files had aired the previous September in the United States, followed by airings across Europe. Just like that, a game about hostile alien visitors seemed a lot more relevant. Indeed, the game possessed much the same foreboding atmosphere as the show, from its muted color palette to MicroProse composer John Broomhall’s quietly malevolent soundtrack, which he had created in just two months in the final mad rush up to the release deadline. He couldn’t have done a better job if he’d had two years.

X-COM: UFO Defense shipped a few months later in North America, into a cultural zeitgeist that was if anything even more primed for it. Computer Gaming World, the American industry’s journal of record, gave it five stars out of five, and its sales soared well into the six digits. As the quote that opened this article attests, X-COM was in many ways the antithesis of what most publishers believed constituted a hit game in the context of 1994. Its graphics were little more than functional; it had no full-motion video, no real-time 3D rendering, no digitized voices; it fit perfectly well on a few floppy disks, thank you very much, with no need for any new-fangled CD-ROM drive. And yet it sold better than the vast majority of those other “cutting-edge” games. Many took its success as a welcome sign that gaming hadn’t yet lost its soul completely — that good old-fashioned gameplay could still trump production values from time to time.



The original X-COM‘s reputation has only grown more hallowed in the years since its release. It’s become a perennial on best-games-of-all-time lists, even ones whose authors weren’t yet born at the time of its release. For this is a game, so we’re told, that transcends its archaic presentation, that absolutely any student of game design needs to play.

That’s rather ironic in that X-COM is a game that really shouldn’t work at all according to many of the conventional rules of design. For example, it’s one of the most famous of all violators of what’s become known as the Covert Action Rule, as formulated by Sid Meier and named after one of his own less successful designs. The rule states that pacing is as important in a strategy game as it is in any other genre, that “mini-games” which pull the player away from the overarching strategic view need to be short and to the point, as is the case in Meier’s classic Pirates!. If they drag on too long, Meier tells us, the player loses focus on the bigger picture, forgets what she’s been trying to accomplish there, gets pulled out of that elusive state of “flow.”

But, as I already noted, X-COM‘s tactical battles can drag on for an hour or two at a time — and no one seems be bothered by this at all. What gives?

By way of an answer to that question, I would first note that the Covert Action Rule is, like virtually all supposedly hard-and-fast rules of game design, riddled with caveats and exceptions. (Personally, I don’t even agree that violating the yet-to-be-formulated Covert Action Rule was the worst problem of Covert Action itself.) And I would also note that X-COM does at least a couple of things extraordinarily well as compensation, better than any strategy game that came before it. Indeed, one can argue that no earlier grand-strategy game even attempted to do these things — not, at least, to anything like the same extent. Interestingly, both inspired strokes are borrowed from other gaming genres.

The first is the intriguing mystery surrounding the aliens, which is peeled back layer by layer as you progress. As your scientists study the equipment and alien corpses brought back from the battle sites and interrogate the live aliens your soldiers have captured, you learn more and more about where your enemies come from and what motivates them to attack the Earth so relentlessly. It doesn’t take long to reach a point where you look forward to the next piece of this puzzle as excitedly as you do the next cool gun or piece of armor. By the time the whole experience culminates in a desperate attack on the aliens’ home base, you’re all in. Granted, a byproduct of this sense of unfolding discovery is that you may not feel like revisiting the game after you win; for many or most of us, this is a strategy game to play through once rather than over and over again. But on the other hand, considering the fifty hours or more it will take you to get through it once, it’s hard to complain overmuch about that fact. Needless to say, when you do play it for the first time you should meticulously avoid spoilers about What Is Really Going On Here.

Learning more about the alien invaders via an autopsy. The game was ahead of its time; the year after X-COM‘s release, at the height of the X-Files-fueled UFO craze, the Fox television channel would broadcast Alien Autopsy: Fact or Fiction? in the United States. (For the record, it was most assuredly the latter.)

X-COM‘s other, even more brilliant stroke is the sense of identification it builds between you and the soldiers you send into battle. Each soldier has unique strengths and weaknesses, forcing you to carefully consider the role she plays in combat: a burly, fearless character who can carry enough weaponry to outfit your average platoon but couldn’t hit the proverbial broad side of a barn must be handled in a very different way from a slender, nervous sharpshooter. As your soldiers (hopefully) survive missions, their skills improve, CRPG-style. Thus you have plenty of practical reasons to be more loathe to lose a seasoned veteran than a greenhorn fresh out of basic training. And yet this purely zero-sum calculus doesn’t fully explain why each mission is so nail-bitingly tense, so full of agonizing decisions balancing risk against reward.

One of X-COM‘s most defining design choices is also one of its simplest: it lets you name each soldier for yourself. As you play, you form a picture of each of them in your imagination, even though the game itself never describes any of them to you as anything other than a list of numbers. Losing a soldier who’s been around for a while feels weirdly like losing a genuine acquaintance. For here too you can’t help but embellish the thin scaffolding of fact the game provides with your own story of what happened: the grizzled old-timer who went out one time too many, whose nerves just couldn’t handle another firefight; the foolhardy, testosterone-addled youth who threw himself into every battle like he was indestructible, until one day he wasn’t. X-COM provides the merest glimpse of what it must feel like to be an actual commander in war: the overwhelming stress of having the lives of others hanging on your decisions, the guilty second-guessing that inevitably goes on when you lose someone. It has something that games all too often lack: a sense of consequences for your actions. Theoretically at least, the best way to play it is in iron-man mode: no saving and restoring to fix bad outcomes, dead is dead, own your decisions as commander.

Beginning with just a name you choose for yourself and a handful of statistics which the game provides, your imagination will conjure a whole personality for each of your soldiers. Dwight here, for example, likes guitars, Cadillacs, and hillbilly music.

In one of those strange concordances that tend to crop up in many creative fields, X-COM wasn’t the only strategy game of 1994 to bring in CRPG elements to great effect. Ironically, these innovations occurred just as the CRPG genre itself was in its worst doldrums since Ultima I and Wizadry I had first brought it to prominence. Today, even as the CRPG has long since regained its mojo as a gaming genre, CRPG elements have become the special sauce ladled over a wide array of other types of games. X-COM was among the first to show how tasty the end result could be.

I have to say, however, that I find other elements of X-COM less appetizing, and that its strengths don’t quite overcome its weaknesses in my mind sufficiently to win it a place on my personal list of best games ever. My first stumbling block is the game’s learning curve, which is not just steep but unnecessarily so. I’d like to quote Garth Deangelis, who led the team that created XCOM: Enemy Unknown, the critically acclaimed franchise reboot that was released in 2012:

While [the original X-COM] may have been magnificent, it was also a unique beast when it came to beginning a new game. We often joked that the diehards who mastered the game independently belonged to an elite club because by today’s standards the learning curve was like climbing Mount Everest.

As soon as you fire up the original, you’re placed in a Geoscape with the Earth silently looming, and various options to explore within your base — including reading (unexplained) financial reports, approving manufacturing requests (without any context as to what those would mean later on), and examining a blueprint (which hinted at the possibility for base expansion), for example — the player is given no direction.

Even going on your first combat mission can be a bit of a mystery (and when you first step off the Skyranger, the game will kill off a few of your soldiers before you even see your first alien — welcome to X-COM!).

There’s certainly a place for complex games, and complexity will always come complete with a learning curve of some sort. But, again, X-COM‘s curve is just unnecessarily steep. Consider: when you begin a new game, you have two interceptors already in your hangar for bringing down UFOs. Fair enough. Unfortunately, they come equipped with sub-optimal Stingray missiles and borderline-useless cannon. So, one of the first tasks of the experienced player becomes to requisition some more advanced Avalanche missiles, put them on her interceptors, and sell off the old junk. Why can the game not just start you off with a reasonable weapons load-out? A similar question applies to the equipment carried by your individual soldiers, as it does to the well-nigh indefensible layout of your starting base itself, which makes it guaranteed to fall to the first squad of marauding aliens who come calling. The new player is likely to assume, reasonably enough, that the decisions the game has already made for her are good ones. She finds out otherwise only by being kicked in the teeth as a result of them. This is not good game design. The impression created is of a game that is not tough but fair, but rather actively out to get her.

Your starting base layout. By no means should you assume that this is a defensible one. In fact, many players spend a lot of money at the very beginning ripping it up completely and starting all over again. Why should this be necessary?

You’ll never use a large swath of the subpar weapons and equipment included in X-COM, which rather begs the questions what they’re doing in there. The game could have profited greatly from an editor empowered to pare back all of this extraneous nonsense and home in on its core appeal. Likewise, the user interface in the strategic portion operates on the principle that, if one mouse click is good, ten must be that much better; everything is way more convoluted than it needs to be. Just buying and selling equipment is agonizing.

The tactical game’s interface is also dauntingly complex, but does have somewhat more method to its madness, being the beneficiary of all of Julian Gollop’s earlier experience with this sort of game. Still, even tactical combat, so widely and justly lauded as the beating heart of X-COM, is not without its frustrations. Certainly every X-COM player is all too familiar with the last-alien-on-the-map syndrome, where you sometimes have to spend fifteen or twenty minutes methodically hunting the one remaining enemy, who’s hunkered down in some obscure corner somewhere. The nature of the game is such that you can’t relax even in these situations; getting careless can still get one or more of your precious soldiers killed before you even realize what’s happening. But, although perhaps a realistic depiction of war, this part of the game just isn’t much fun. The problem is frustrating not least because it’s so easily soluble: just have the remaining aliens commit suicide to avoid capture — something entirely in keeping with their nature — when their numbers get too depleted.

All of these niggling problems mark X-COM as the kind of game I have to rant about here all too often: the kind that was never actually played before its release. For all its extended development time, it still needed a few more months filled with play-testing and polishing to reach its full potential. X-COM‘s most infamous bug serves as a reminder of just how little of either it got: its difficulty levels are broken. If you select something other than the “beginner” difficulty, it reverts back to the easiest level after the first combat mission. In one sense, this is a blessing: the beginner difficulty is more than difficult enough for the vast majority of players. On the other, though… how the heck could something as basic as that be overlooked? There’s only one way that I can see: if you barely played the game at all before you put it in a box and shipped it out the door.

To his credit, Julian Gollop himself is well aware of these issues and freely acknowledges them — does so much more freely in fact than some of his game’s biggest fans. He notes the influence of vintage Avalon Hill and SPI board games, some of which were so demanding that just being able to play them at all — never mind playing them well — was an odd sort of badge of honor for the grognards of the 1970s and early 1980s. He would appear to agree with me that there’s a bit too much of their style of complexity-for-its-own-sake in X-COM:

I believe that a good game may have relatively simple rules, but have complex situations arise from them. Strategy games tend to do that very well, you know — even the simplest ones are very good at that. I think it’s possible to have an accessible game which doesn’t have amazingly complex rules, but still has a kind of emerging complexity within what happens — you know, what players do, what players explore. For me, that’s the Holy Grail of game design. So, I don’t think that I would probably go back to making games as complex as [X-COM].

Like poets, game designers often simplify their work as they age, the better to capture the real essence of what they’re trying to express.



But whatever their final evaluation of the first game, most players then and now would agree that few franchises have been as thoroughly botched by their trustees as X-COM was afterward. When the first X-COM became an out-of-left-field hit, MicroProse UK, who had great need of hits at the time to impress the Spectrum Holobyte brass, wanted the Gollops to provide a sequel within a year. Knowing that that amount of time would allow them to do little more than reskin the existing engine, they worked out a deal: they would give their publisher their source code and let them make a quickie sequel in-house, while they themselves developed a more ambitious sequel for later release.

The in-house MicroProse project became 1995’s X-COM: Terror from the Deep, which posited that, forty years after their defeat at the end of the first game, the aliens have returned to try again. The wrinkle this time is that they’ve set up bases under the Earth’s oceans, which you must attack and eradicate. Unfortunately, Terror from the Deep does little to correct the original’s problems; if anything, it makes them worse. Most notably, it’s an even more difficult game than its predecessor, a decision that’s hard to understand on any level. Was anyone really complaining that X-COM was too easy? All in all, Terror from the Deep is exactly the unimaginative quickie sequel which the Gollops weren’t excited about having to make.

Nevertheless, it’s arguably the best of the post-original, pre-reboot generation of X-COM games. X-COM: Apocalypse, the Gollops’ own sequel, was a project on a vastly greater scale than the first two X-COM games, a scale to which they themselves struggled to adapt. It was riven by bureaucratic snafus and constant conflict between developer and publisher, and the resulting process of design-by-fractious-committee turned it into a game that did a lot of different things — turned-based and real-time combat in the very same game! — but did none of them all that well, nor even looked all that good whilst doing them. Julian Gollop today calls it “the worst experience of my entire career” and “a nightmare.” He and Nick cut all ties with MicroProse after its 1997 release.

After that, MicroProse lost the plot entirely, stamping the X-COM label onto games that had virtually nothing in common with the first one. X-COM: Interceptor (1998) was a space simulator in the mode of Wing Commander; Em@il Games: X-COM (1999) was a casual multiplayer networked affair; X-COM: Enforcer (2001) was a mindless shoot-em-up. This last proved to be the final straw;  the X-COM name disappeared for the next eleven years, until XCOM: Enemy Unknown, the reboot by Firaxis Games.

If you ask me, said reboot is in absolute terms a better game than the original, picking up on almost all of its considerable strengths while eliminating most of its weaknesses. But it cannot, of course, lay claim to the same importance in the history of gaming. Despite its flaws, the original X-COM taught designers to personalize strategy games, showed them how to raise the emotional stakes in a genre previously associated only with cool calculation. For that reason, it richly deserves its reputation as one of the most important games of its era.

(Sources: the book Grand Thieves and Tomb Raiders: How British Video Games Conquered the World by Magnus Anderson and Rebecca Levene; Amstrad Action of October 1989; Computer Gaming World of August 1994, September 1994, April 1995, and July 1995; Crash of Christmas 1988 and May 1989; Game Developer of April 2013; Retro Gamer 13, 68, 81, 104, 106, 112, and 124; Amiga Format of December 1989, June 1994, and November 1994; Computer and Video Games of December 1988; Games TM 46; New Computer Express of September 15 1990; Games Machine of July 1988; Your Sinclair of August 1990 and September 1990; Personal Computer News of July 21 1983. Online sources include Julian Gollop’s X-COM postmortem from the 2013 Game Developers Conference, “The Story of X-COM at EuroGamer, and David Jenkins’s interview with Julian Gollop at Metro.

The original X-COM is available for digital purchase at GOG.com, as are most of the other X-COM games mentioned in this article.)

 
65 Comments

Posted by on September 18, 2020 in Digital Antiquaria, Interactive Fiction

 

Tags: , ,

Bullfrog after Populous

I’ll give you an analogy of what Populous is in my mind. Imagine if I had a blank canvas. Some people that are true artists will take a palette of paint and mix them together and carefully handcraft each and every single brushstroke until they have some beautiful and amazing picture. And then there’s me. I had the blank canvas, accidentally knocked a can of paint over, and it went splat. And an art dealer has seen it and said, “That’s brilliant.” Well, I know all I’ve done is kick a can of paint. And that’s what I believe really happened with Populous.

— Peter Molyneux

When we last met Peter Molyneux and his little database-developer-turned-games-studio Bullfrog Software, they had just made Populous and watched in disbelief as it blew up huge. The radically innovative game joined Will Wright’s SimCity as one of the progenitors of a hazily delineated new genre which the media labelled the “god game” for the way it gave you direct control over an environment but only indirect control over the people therein. As Populous became a hit on three continents and sold in the hundreds of thousands and eventually millions of copies, Bullfrog struggled to reckon with the cognitive dissonances of their changed circumstances. In a matter of months, they went from a handful of poverty-stricken dreamers pissing in the sink of their miserable, toilet-less hovel of an office to Britain’s single most successful and respected games studio of all.

The trappings of their success tended to trail behind their sales figures: when a group of Japanese executives stopped by the Bullfrog hole-in-the-wall to discuss plans for publishing Populous in their country, the senile pensioner who lived below met them at the bottom of the stairs with a mop and proceeded to beat them off the premises. “We had to get out of there as soon as possible,” says Molyneux. They wound up in a more conventional business park, whose more conventional tenants complained endlessly about their penchant for racing skateboards through the hallways and shooting BB guns out the windows. (“We were brats,” admits Molyneux today. “Horrible, horrible brats.”)

While certainly preferable to failure, success could be its own kind of mixed blessing. Expectations of Bullfrog, which had previously been nonexistent, were suddenly sky high. After a quickie add-on disk that brought additional levels and environments to Populous, they made a rather shockingly unambitious little platformer called Flood, a project of Bullfrog programmer Sean Cooper. Released only in Europe for the Commodore Amiga and Atari ST by Bullfrog’s publisher Electronic Arts, it garnered a collective shrug from the magazines; the market was already flooded with platformers much like this one, often with more compelling level designs. This sort of derivative work wasn’t at all what people had come to expect from Bullfrog after Populous.

Luckily, they had something else in the works. Released in late 1990, some eighteen months after Populous, the real-time war game Powermonger evinced a lot of innovation on its own terms even as it clearly drew from the same set of techniques and approaches that had yielded Bullfrog’s first hit. Instead of casting you as a god able to alter the very landscape of the world on behalf of your mortal followers, Powermonger cast you as an ordinary human on a mission to conquer the world — all 195 regional cross-sections of it, one region at a time. The ethos of indirect control that had made Populous so unique remained: you had to convince the people to rally to your cause, and had to work constantly to keep them loyal to you. Ditto a focus on large-scale environmental effects: you had to worry about the ecology of the land in order to feed, water, house, and equip your people. (After all, it’s hard to build much of anything if you’ve already clear-cut all of the forests…) While Computer Gaming World‘s Johnny L. Wilson, who was always eager to read meaning into games, may have been overstating the case when he called Powermonger “a dynamic treatise on the human capacity for aggrandizement and the potential consequences therein,” the same magazine’s description of it as the “thinking person’s Populous” was a good deal more tenable.

Unfortunately, it suffered from many of the same flaws as its predecessor — flaws which would become consistent hallmarks of Molyneux’s work in general. He obviously wanted to give players a lot of game by providing 195 levels, but, being all procedurally generated, they didn’t really build upon one another or force the player to reevaluate the tools at her disposal in interesting new ways. Powermonger was great fun at first — thus all the glowing reviews in the magazines — but it started to feel a little bit rote a little bit too quickly.

The graphics in Powermonger got a dramatic upgrade over those in Populous, yielding not only aesthetic but also practical benefits: it was now possible to rotate your view of the landscape and zoom it in and out as needed, while the variety of landscape features was dramatically greater. “In Populous,” noted Molyneux, “we had hills, houses, and rivers. Thanks to this new system, we’re able to generate waterfalls, cliffs, valleys, mountains, proper towns, road networks, forests… it’s a real world!”

Molyneux originally conceived of Powermonger not so much as a standalone game as an engine for running a variety of them. After the first game with its vaguely Medieval theme, he talked of making a World War I version, a high-fantasy version, an Asian version for the Japanese market, and a version focusing on the American Civil War for the punters in the United States. But Powermonger, while moderately successful, never became the sensation that Populous was, and most of those plans were abandoned; only the World War I data disk ever appeared. Powermonger “appeals to a lot of people who like very, very high strategic games, but it needed that extra element that would appeal to everybody and it didn’t have that,” said Molyneux after the dust had settled. He blamed the lack largely on the pressure Bullfrog was under from Electronic Arts to complete and release the game in time for Christmas, which meant that it didn’t get played prior to release to anywhere near the extent of Populous.

Still searching for that elusive second million-selling hit, Bullfrog opted to drink even deeper from the old Populous well next time around. Their game for the Christmas of 1991 was Populous II, which mated the improved interface and graphics of Powermonger to the literal god-game theme of Populous I. There was slightly more semblance of a plot this time out: you played a minor deity who must fight her way through a pantheon of some 35 Greek gods, culminating in Zeus himself. Your powers too were more varied than last time out; no longer could people scoff that the game was nothing more than an elaborate topography simulator, not with your ability to spawn tidal waves, whirlwinds, and lightning strikes. Yet one only had to glance at the screen, or read about its more than 1000 (!) anonymous, procedurally-generated levels to know that this was still very much Populous, for both good and bad. It sold well to the committed faithful and spawned the by-now standard expansion pack; in a sop to the Japanese market, where the first Populous had become so popular as to spawn graphic novels and symphony concerts recreating the game’s soundtrack, the expansion was set in ancient Japan rather than Greece. But even so, Populous II made relatively few new converts to the cause at home or abroad.

Populous II. Molyneux admits to feeling “ashamed” at the time to be doing a sequel at all, but he felt obligated to deliver a direct follow-up to such a massive hit. He considers Populous II a reasonable but somewhat unimaginative sequel, which in rather typical industry fashion added a lot more stuff to the template of its predecessor in the form of new godly powers, but failed to drill down on what actually made the original fun. A fair assessment, I think.

Although their latest games hadn’t sold quite as well as the original world-beating Populous, Bullfrog remained the preeminent British games studio in the minds of many. Their status was rivaled only by that of DMA Design, whose Lemmings had become upon its release in early 1991 the most successful single British game since Populous. But DMA was located way off in Dundee, Scotland, a country away from the press on Fleet Street, and when an intrepid journalist did make the trek out to those hinterlands its founder David Jones didn’t provide as many choice quotes as the gregarious Peter Molyneux, then as now one of his industry’s greatest raconteurs. The press loved him not least because he was so willing to go against the official industry position on many subjects, full speed ahead and damn the torpedoes. Asked about the effects of piracy — a subject guaranteed to produce predictions of an imminent gaming apocalypse from any other prominent industry figure — Molyneux shocked his interviewer by replying mildly that “to be honest, I don’t think that piracy hurts.” Likewise, he never hesitated to air his real feelings about competing products: “If you’ve got a crappy shoot’em-up style game, then that’s going to be pirated to hell — and it should be because it’s not even worth using this planet’s resources to produce the game.”

The chain-smoking, perpetually hyperactive Molyneux got on particularly splendidly with the laddish British Amiga magazines of the time. He was up for pretty much anything when they came calling — as when he agreed to be interviewed about a recent trip to Japan while sitting on the toilet. From here he regaled his interlocutors with anecdotes about geisha girls, and told them that “Peter Molyneux” meant “wooden tit” in Japanese: “If I wanted to break the ice anywhere I just said ‘Molyneux’ and the whole room would break up.” He cultivated the persona in such interviews of a slightly befuddled ordinary bloke who liked to spend his time down at the pub when he wasn’t making games, who had no idea how he had stumbled into this charmed career of his. Substitute playing games for making them, and remove the charmed career, and he seemed of a piece with most of the people reading the interviews. He never missed an opportunity to run down his programming skills. “When we wrote Populous,” he said, “we barely knew how to put a sprite onto the screen.” In the end, he claimed, “programming isn’t really that skillful. Anybody can learn to program, anybody, within a week.” Asked to describe Bullfrog in a single sentence, he did so in three words: “Disorganized but keen.” Or, as he put it on another occasion: “We write computer games. We’re not businessmen.”

The Peter Molyneux toilet interview for Zero magazine in December of 1991. Molyneux had by then perfected the art of press relations, which in the case of the gaming magazines often came down to the simple expedient of taking the youthful journalists down to the pub and getting them blind drunk.

In keeping with this everyman persona, Molyneux evinced no interest whatsoever in professional credentials. Recruitment at Bullfrog operated on the principle of “show us what you got,” via little classified advertisements placed in the nether reaches of the same magazines that were featuring Molyneux and his games on their covers. Bullfrog endeared themselves even more by running tutorials in said magazines, teaching graphics and programming tricks; at least one series of tutorials concluded with a contest for those who had been following along diligently, the prize a potential job with Bullfrog. “You too can make games!” was the message. And people loved Molyneux for it.

But there was also another side to Molyneux: the side that was a real businessman, whatever his claims to the contrary — a businessman who was watching his industry with eagle eyes. When someone deigned to ask him a serious question, he could deliver a cogent, sometimes even prescient response. For example, when asked whether personal computers would ultimately win out over consoles as game-playing devices, he had this to say:

Definitely not. PCs are too much bother, even with CDs. You’ve got to configure one of the 30 trillion sound cards’ 30 trillion settings. I don’t understand all these DMAs and IRQs and all that crap. I just fiddle around until I get it right. Until they sort that out, the machine is just going to terrify people.

About the general state of games in the early 1990s, he had this to say:

The current trend in games like simulations, adventures, and some sports sims is that they are getting progressively harder, cleverer, and more challenging. But that doesn’t necessarily make them better games. The trouble is that a lot of games are getting so hard that only the very best gamers can play them. The first rule of game design is that you mustn’t produce games that are too complex for people to play. Being overly complex for the sake of being complex is not a good idea. Complexity is good as long as it doesn’t get in the way of the game.

And this:

We’re into a new thing called interactive drama. Everybody’s doing interactive drama with interactive plots and interactive characters. But I think it’s going to be a tough, tough thing to do. Hollywood spends millions of pounds on scripts. They have hundreds and hundreds of scriptwriters and they get it right once or twice a year. And little game designers like us are coming along and we’re going to write this script which is going to have infinite variations, is going to be as entertaining as any Hollywood film, is going to have cinematic sequences in it, and we’re going to sell it for four times more than you can buy a video for. There’s something wrong there. Either we’re very, very clever chaps and Hollywood has been doing it wrong for the last 100 years, or perhaps we’re talking out of our arses.

Quotes like these made Molyneux into something of a spokesman for the British games industry, in the mainstream as well as the specialty press. Whatever the intrinsic merits of claims like those above, they had the advantage of poking holes in exactly the sorts of games which British studios tended to lack the resources to do as well as the Americans.

During this period, British games still largely meant Amiga games. Thus it was tough for Molyneux, both in his role as spokesman for his industry and as a proud Briton, to admit that Bullfrog just couldn’t continue to develop their games on the Amiga first and remain competitive in the international market; the latest MS-DOS machines were pulling too rapidly ahead of Commodore’s trusty old platform. Bullfrog’s next big project after Populous II would be developed first on MS-DOS and then ported to the Amiga in slightly downgraded form — the opposite of the studio’s earlier approach. For, as Molyneux put it, “you can let your imagination run wild” on an MS-DOS machine.

The same project would be a welcome, much-needed departure in both form and content from the games Bullfrog had spent the last few years making. It would be a much grittier, more down-to-earth affair of rival corporations doing battle with one another in an oligarchic worldwide dystopia of the near future. As such, it was of a piece with many of the print fictions which young men like the Bullfrog crew were reading in the early 1990s — think Neal Stephenson’s Snow Crash and, reaching just a little further back, William Gibson’s Neuromancer, the original popularizer of cyberpunk cool. (Self-effacing as ever, Molyneux claimed that “I read the first three chapters of Neuromancer, but it all went a bit above my head.”) Still, Bullfrog did add their own sprinkling of dark British humor to the mountain of cyberpunk clichés. For many months, they called the game simply Bob, after the infamously ruthless British media mogul Robert Maxwell, whose suspicious death aboard his luxury yacht and the subsequent revelation of financial malfeasance throughout his empire dominated tabloid headlines during the period.

In the end, though, Bob became known as Syndicate. Bullfrog spent a full two years working on it, marking the longest gap between games in their entire history as a studio. They claimed the end result to contain as much code and graphics as every one of their previous games combined.

Whereas Populous and its descendants were played entirely from a single interface, Syndicate was a more disparate affair. As the owner of a tiny upstart corporation bent, naturally enough, on taking over the world, you had to allocate research funds for equipment and cybernetic enhancements for your army of operatives, whilst choosing missions to send them on from a Risk-like strategic map. But it was in the missions themselves, which played out in real time from an isometric perspective, that you spent the vast majority of your time. Here you had more direct control over your operatives than you did in Populous, but they still had minds of their own, which could cause them to react with a spontaneous burst of gunfire if, for example, an enemy agent popped up in their path; it might also cause them to refuse to obey an obviously dangerous command. The missions took place in living city streets replete with civilians as well as combatants, presenting ample opportunities for mayhem. Syndicate has been called a proto-Grand Theft Auto, what with the way it tempts its player to indulge in random violence and acts of destruction for their own sake.

Syndicate

Indeed, when it finally appeared in late 1993, just weeks before id Software’s landmark DOOM, Syndicate struck many as the natural British companion to that American game, another avatar of a movement toward ever more visceral forms of violence in games. As in DOOM, blood splashed everywhere in Syndicate with gleeful abandon, and much of the appeal of acquiring new weapons was in the visible carnage they created. This prompted a brief-lived round of nervous stock-taking in the magazines of both countries — until the same magazines, seeing that hordes of players loved the violence, learned to defer to the readers who buttered their bread.

Extreme though it was by contemporary standards, the violence in the finished Syndicate was reportedly toned down from earlier versions, where you had been allowed to kill babies and pets. Play testers “told us we were going too far,” mused Molyneux. “Funnily enough, they objected most to killing the puppies…” Yet he remained unapologetic on the whole:

We made Syndicate high in gore to be more realistic. I know this sounds like a cop-out, but it’s the player who’s violent, not us. All we’ve done is give a loaded weapon to the player and it’s up to them how to use it. I’ve always hated games that gloss over violence. Surely showing the realism of a violent act is better than disguising it? It’s not that Syndicate had done anything new with violence, it’s just that it shows it like it is.

For all the changes it evinced over what had come before, Syndicate was a typical Bullfrog game in other ways. It started out thoroughly entrancing, but went on way too long, with only a handful of fixed mission types on offer as you slowly — very slowly — took over the world. By the mid-game, you had discovered most of the cool gear and cybernetic enhancements, and what had started out fresh and exciting had begun to turn into a bit of a grind. Thus Syndicate became another Molyneux game that far more players started than finished. Nevertheless, its initial appeal was enough to make it Bullfrog’s biggest hit since the original Populous, and the game and its 1994 expansion pack are still fondly remembered by many today.

By this point, Bullfrog had grown from just a few employees to about forty, enough to have multiple projects on the boil at one time, all receiving varying degrees of attention from the hardworking and endlessly enthusiastic Peter Molyneux. Thus the next game arrived barely six months after Syndicate. It would prove to be one of Molyneux’s most influential creations of all.

The project had its genesis in the first game he ever made, a text-only business simulation called Entrepreneur, of which he had sold exactly two copies — one of them quite possibly to his mother — in 1983. That failure had continued to rankle even amidst all the success he enjoyed in later years, as did the urge to make another, less dry business simulation that would appeal to more people. “Then one day the perfect idea hit me,” Molyneux says. “I’d create a game where you control a theme park.” Molyneux:

I love theme parks, and it was a great excuse to do some really good research. It’s also something where you often go back and think, “If I’d been given the chance to design this place I wouldn’t have put this here, or that there, etc.” And it’s also something that people can immediately associate themselves with. If I tell you that this game enables you to design theme parks, then you immediately know what I’m talking about.

It was indeed a brilliant stroke, one with natural appeal well beyond the typical gamer demographic who enjoyed the likes of Syndicate. Visually at least, Theme Park would be the polar opposite of that game, cheerful and bright where it had been gritty and dark.

Theme Park

Released in mid-1994, Theme Park became a monster hit — bigger than Syndicate, even bigger than Populous after it was ported to every viable or semi-viable game-playing gadget in the world. Its bright and bouncy visual aesthetic presaged the Casual Revolution in games that was still some years away, while its impact on the themes and mechanics of games to come would prove even more pronounced. In particular, Rollercoaster Tycoon, a direct heir to Theme Park which was released by MicroProse in 1999, sold even better than Bullfrog’s take on the concept — in fact, became one of the best-selling computer games in history. Today amusement parks and roller coasters remain a staple of gaming, from the more elaborate examples of the breed available at online stores like Steam to more easygoing affairs that you can play right in your browser. Almost all of them owe not just a thematic (hah!) debt to Theme Park but a direct mechanical and visual one as well, from the thought bubbles that appear over the heads of the guests wandering through the park to their whimsically cartoony graphical style.

It thus pains even more than usual to note how horribly Theme Park itself has aged, even in comparison to most of the other early Bullfrog games. Few games evidence as profound a mismatch between their surface aesthetics and their underlying gameplay as this one does. The cutesy nature of the former can confuse you for a long time, disguising the fact that the latter really doesn’t represent as great a departure from the worldview of Syndicate as it seems to let on. At bottom, Theme Park is a nasty, cynical little game, amoral if not actively immoral — a game where your concern isn’t with the happiness of your guests at all, but strictly with the amount of money you can extract from them; a profitable theme park with miserable patrons is not only possible but the only practical road to success. This is the kind of game where you over-salt the patrons’ fries to get them to buy more soda, which cups you stuff to the brim with ice to… well, you get the picture. If you come to this game wanting to build a beautiful amusement park and show everybody who visits it a great time, as the Molyneux quote above would imply you can, you’ll wind up bankrupt and disillusioned in extremely short order.

It’s really hard to know what parts of Theme Park to attribute to intentional subversiveness and what parts to simple tone-deafness. The intro video is a perfect case in point. Was Bullfrog aware of just how weird and creepy this thing is? The dog has the right idea: “Hell, no, leave me at home!”

Even if you’re willing to play the game on its own cynical terms, it has all sorts of other problems. There’s a paucity of useful feedback on both a global and granular level, which often puts you in the supremely frustrating position of failing for reasons you can’t determine. The interface in general is inscrutable in too many places, the level of micromanagement required is exhausting, and, because this is a Peter Molyneux game, winning is a task so herculean that virtually no one has ever done so: after building your first successful park in Britain, you’re expected to choose another location elsewhere in the world and do so again, ad nauseum. None of these later parks are different in any fundamental way from the first — you have the exact same rides and shops and food stands at your disposal throughout — and so the whole exercise becomes absurdly repetitive.

Theme Park was a hugely innovative and massively influential game, but it just wasn’t a very good one, even in its heyday. Its appeal was always rooted more in what it purported to be than what it actually managed to be. Because everybody loves a theme park, right?

A ride goes haywire and a kid goes flying. I’m pretty sure this part is deliberately subversive…

Bullfrog’s second game of 1994 — also the last one which we’ll be visiting as part of this little survey today — might have appeared at the time to be an attempt to jump onto the 3D-action bandwagon unleashed by Wolfenstein 3D and DOOM. In reality, though, Bullfrog had been experimenting with first-person 3D in-house for years. Those experiments finally led, after many detours and false starts, to Magic Carpet, whose namesake you got to fly — because, as Molyneux wryly put it, every other possible form of flight had already been exhaustively simulated by that point. As one of several wizards, your goal was to build up your arsenal of spells and mana in order to conquer all of the opposing wizards and take over the world. And then — remember, this was a Bullfrog game — you were expected to do the same thing in fifty or so more worlds.

The first Bullfrog game not to be ported at all to lower-powered platforms like the Amiga, Magic Carpet was a stunning technical achievement in its time. While other 3D action games segmented themselves into discrete levels made up of interior spaces only, it gave you a complete open-ended world to explore. It was a forthrightly artsy game, something DOOM and the rash of similar games which followed it certainly never aspired to be. In that spirit, it contained no words during actual gameplay, nothing to distract from the evocative wonder of its world. Bullfrog’s staffers talked in interviews about the joy they got just from drifting around above its landscapes before they’d put any enemies in — playing as they did so, they said only half facetiously, their Enya albums. Magic Carpet even had a special stereoscopic 3D mode, for those able to buy or make 3D glasses to suit.

The graphics in Magic Carpet remain strikingly beautiful to this day.

By the time you got four or five levels into it, however, it revealed itself to suffer from the standard set of Bullfrog problems. Each of its worlds was superficially different from the one before, but not in a way that really challenged you or introduced a sense of progression beyond the increasing level number on your status screen. The game shot its bolt at the beginning, then just kept giving you more of the same. Peter Molyneux spoke often in interviews about his desire to give gamers lots of value for their money by making big games. Yet, like songwriters with a knack for melody who have no clue how to take it to the bridge, he and his mates consistently struggled to find ways of varying their formulas so that their games weren’t just more of the same for hours on end. As it was, what you saw in the first hour of a Bullfrog game was what you would continue to see for the next hundred hours.

And for once, this Bullfrog game’s presentation and theme alone weren’t enough to save it on store shelves. Its abstracted and almost aggressively artsy personality combined with its high production costs and high system requirements to make it Bullfrog’s first outright money loser since Flood.



How, then, should we sum up these five busy years in the life of Peter Molyneux and his first company? We can feel certain that anything we do say must apply almost equally to his career since 1994; whether you love or hate his work, its strengths and weaknesses haven’t changed very much over the decades. An unkind assessment — of which there have been many in the last ten years in particular, as Molyneux’s real or perceived penchant for over-hyping and under-delivering has come home to roost — might peg him as a bit of a dilettante, an ideas man unwilling to do the hard work to turn his ideas into balanced games that remain playable and interesting over the long term. But the reality is, as usual, more complex than any single pejorative — or compliment, for that matter — can encompass.

Some keys to the puzzle of Peter Molyneux can undoubtedly be found in the scene from which he sprang. His design aesthetic, like that of so many British game developers, was to a large extent forged by the limited resources — in terms of both target hardware and finances — which they had at their disposal. Bullfrog’s stubborn reliance on procedurally-generated rather than handcrafted levels, often to their games’ detriment, can be traced back at least to Ian Bell and David Braben’s Elite and the vast eight-galaxy universe it packed into a 32 K BBC Micro via the magic of the Fibonacci sequence. When one didn’t have much space to store handcrafted levels and didn’t have many people to hand to make them, procedural generation seemed the only practical way forward. But Bullfrog stuck with it to the exclusion of other approaches for too long — long after other approaches became viable.

Other pieces of the puzzle are more idiosyncratic to Molyneux himself, a fellow whose own personality was always all but inseparable from that of his company. Already by the mid-1990s, his tendency to stretch himself in too many directions at once was starting to become an issue. During the period of Theme Park and Magic Carpet, Bullfrog also worked on something called Creation, where you would breed predatory fish in an underwater base to attack your rivals on the ocean floor. Molyneux even mooted linking Creation with Magic Carpet: “If you’re playing Magic Carpet, you will be able to jump off the carpet and into the ocean. The computer will then sense whether you have Creation on your hard disk and plunge you straight into that, based totally on the world you were just flying around [in].” Another work in progress, with the highly inadvisable title of MIST (My Incredible Superhero Team), would let you build and control your own superhero: “If you want to make him strong and give him rubber wings and death vision, then you can do that. But of course, they’ve all got their Achilles heel.” And then there was Biosphere, featuring a more elaborate, planet-wide take on genetic engineering along with a fictional context shamelessly ripped off from Douglas Adams, where you would “run a team of genetic and planet engineers who modify planets for shiploads of colonists. Unfortunately for you, the colonists are generally hairdressers and telephone engineers, so when they get there they’re pretty useless — and they’ll probably be eaten by dinosaurs. So you have to protect them.” None of these games were ever completed, despite a substantial amount of time and resources being devoted to each of them. Indeed, as the resources available to him increased, Molyneux’s proclivity for rushing enthusiastically down such blind alleys increased in equal measure.

Molyneux’s passionate prioritization of experimentation over the nuts and bolts of game design made him a less complete designer than, say, a Sid Meier. And yet he, along with other designers of a similar bent, have been scarcely less necessary for the evolution of their medium. Few if any designers have dared to put more new stuff out there than Molyneux, even if often in imperfect form. Such experiments can become the building blocks for more grounded designers to build upon, as the example of the badly flawed Theme Park begetting the absolutely brilliant Rollercoaster Tycoon proves in spades.

Another component of Molyneux’s claim to the status of gaming visionary is more generalized: his complete conviction during the early 1990s that, as he put it, “multiplayer games are the future of gaming.” With the exception only of Theme Park, every Molyneux game from Populous on not only supported multiplayer sessions between players on separate computers[1]Multiplayer Syndicate was made available to the public only in the expansion pack. but was literally designed for it first and foremost. That is to say that a serial or network link-up went into PopulousPowermongerPopulous II, Syndicate, and Magic Carpet long before anyone even began to think about adding a computer opponent. One might even call this fact the perfect riposte to all of my complaints about these Bullfrog games. If you played them alone, you were, in Molyneux’s mind anyway, playing them wrong in some fundamental sense. Complaints about the sameness of a game from level to level no longer carry much weight when you’re playing against that ultimate agent of unpredictability, a fellow human. While the nature of the times dictated that most people played them solo, there are nevertheless all sorts of anecdotes about the sharing of those early Bullfrog games among friends; my favorite might be the teenage next-door neighbors who made a 25-meter cable to run between their bedroom windows so that they could play Populous together every night to their hearts’ content. Stories like these, soon to be joined by tales of multiplayer DOOM, were clear signposts in their day to where much of gaming was heading, just as soon as the world’s telecommunications infrastructure caught up to the designers’ vision.

Bullfrog and Peter Molyneux have ironically suffered the opposite fate from that of the standard clichés about pioneers. Greatly appreciated in their own time for all of the bold new things they attempted to do and be, their games’ practical deficiencies seem all too obvious to our more jaded eyes of today. But, even if we can’t quite praise any one of them as a standalone masterpiece, we can recognize the purpose they served in opening up so much virgin territory for exploration by later, often better games. And if Molyneux himself has sinned by promising too much too often, it should be recognized as well that his transgressions have never had their roots in greed or guile. He just wants to make really, really amazing games — wants to make lots of them, thereby to make lots and lots of people happy. There are worse character flaws to have.

(Sources: Retro Gamer 39, 40, 43, 69, and 71; New Computer Express of January 20 1990, October 27 1990, and May 11 1991; CU Amiga of October 1990, February 1991, December 1991, December 1992, November 1993, January 1994, and February 1994; Computer Gaming World of January 1991, April 1991, and December 1994; The One of April 1990, July 1990, December 1990, May 1991, July 1991, December 1991, May 1992, May 1993, June 1993, December 1993, October 1994, and March 1995; Amiga Format of February 1992, October 1992, 1992 annual, June 1994, and May 1995; Zero of December 1991; Edge of January 1994, June 1994, November 1994, March 1995, May 1995, July 1995, and November 1995; PC Zone of June 1993 and November 1994; PC Review of July 1992; Next Generation premier issue. Video sources include the documentary From Bedrooms to Billions and series 3 episode 3 of Bad Influence.

Populous II, Syndicate, Theme Park, and Magic Carpet are all available as digital purchases from GOG.com.)

Footnotes

Footnotes
1 Multiplayer Syndicate was made available to the public only in the expansion pack.
 
58 Comments

Posted by on September 4, 2020 in Digital Antiquaria, Interactive Fiction

 

Tags: , , , , , ,

Death Gate

Licensed games are often dismissed as supremely cynical cash grabs. And for the most part, this judgment is correct: the majority of them are indeed dispiriting affairs.

But no thing in life is all one thing. Take, for instance, the story of Legend Entertainment, which upends all of our prejudices about the licensing racket. Rather than being some evil marketing genius’s idea for maximizing revenue, Legend’s reputation as a peddler primarily of licensed product arose all but accidentally, and at least in the beginning involved very little cynicism at all. It all stems from 1991, when Legend co-founder Mike Verdu decided that he’d really, really like to write a game set in the universe of Gateway, a series of science-fiction novels by Frederick Pohl which he happened to love for reasons pure as the driven snow. When that game was well-received, Legend designer Michael Lindner piped up to say that he would like to make an adaptation of his favorite fantasy series, the Xanth novels of Piers Anthony. And when that too went well, Glen Dahlgren asked for a license to adapt Death Gate, another series of fantasy novels, these ones by Margaret Weis and Tracy Hickman. It was all quite shockingly organic — just like one of the press releases put out by one of the other makers of licensed games, the kind that claim that they were actually passionate about Brand X since long before they were given a chance to make a game out of it. In this case, though, it’s the truth.



Margaret Weis and Tracy Hickman were among a new group of authors who began reaching bookstores shelves in the 1980s with novels that had begun as games; thus there was a certain poetic symmetry in Legend eventually completing the circle. Indeed, Weis and Hickman had at least as much impact on gaming writ large over the course of their careers as they did on the state of the art of the fantasy novel.

It all began for Tracy Hickman, as it did for so many others, with tabletop Dungeons & Dragons. He was introduced to the game by his fiancée Laura Curtis in the late 1970s, when the two were starving university students in their home town of Salt Lake City, Utah. After getting married and starting a family, the couple looked for a way to justify their hobby when they couldn’t even afford “church shoes” for their two young children. (Like a surprising number of significant figures in gaming history, they were and remain devout Mormons.) So, they founded DayStar West Media to publish two pre-written Dungeons & Dragons “adventure modules.” These sold in fairly minuscule numbers, but provided an entrée to TSR, the maker of the game. In 1982, TSR offered Tracy Hickman a job as an in-house designer. And so the couple piled their children and all the luggage that would fit into the back of their battered old Volkswagen Rabbit and set off for Lake Geneva, Wisconsin, the tabletop RPG scene’s Mecca.

Tracy has since said many times that the idea for Dragonlance, the Hickmans’ master stroke in the fields of both games and books, was born on that long drive. They’d been sent a copy of a marketing study TSR had commissioned, which said that, as Tracy remembers: “1) Dungeons & Dragons is your core product. 2) You have lots of dungeons. 3) You need more dragons.” Ergo, Dragonlance would be a series of adventure modules, each revolving around a different type of dragon.

But its short-term goal of injecting more dragons into the mix has little to do with Dragonlance‘s ultimate importance. The Hickmans conceived it as a series of adventures which would be, as tabletop-gaming historian Shannon Applecline puts it, “plot-oriented” rather than “location-based.” Instead of engaging in the mere tactical challenge of clearing a dungeon of monsters and traps and looting its contents, players would get to live out a complicated epic story worthy of Tolkien himself, set in a brand new, richly detailed fantasy world.

Dragonlance evolved into one of the most ambitious projects TSR ever undertook, one that remains to this day an enduring icon of its hobby. A groundbreaking trans-media experience at a time when such things were much rarer than they’ve become today, its story filled twelve adventure modules, supported by numerous source books — and, most importantly for our purposes, by a trilogy of thick novels which walked its readers through the modules. TSR’s management was first inclined to hire a “real” author to write them, but when the initial pool of applicants proved underwhelming they agreed to let Tracy Hickman and another staffer named Margaret Weis have the assignment. (The important contributions of Laura Hickman — née Laura Curtis — not only in the beginning stages of the project but throughout its duration, went sadly uncredited.)

Dragons of Autumn Twilight, the first volume of the Dragonlance Chronicles, appeared in late 1984. It wasn’t quite at the level of Tolkien; you could almost hear the dice rolling behind the scenes during its lovingly detailed round-by-round descriptions of Dungeons & Dragons combat. Still, it offered up a fast-paced story with appealing characters, and for many or most of its readers the Dungeons & Dragons-ness of the whole affair was actually a feature rather than a bug. It launched TSR into a whole new field of publishing, one that would eventually become more profitable than the games which continued to inform the novels so intimately.

Weis and Hickman, meanwhile, saw their names suddenly at the top of genre-fiction bestseller lists. Almost overnight, the two heretofore obscure game designers became household names among fans of fantasy fiction, bankable enough to be sought after by the old guard of American publishing. They stuck with TSR long enough to write a second, standalone trilogy of Dragonlance novels, but then departed the games industry entirely for more lucrative climes, leaving others there — perhaps most notably the company White Wolf, with their World of Darkness series of Gothic-horror RPGs — to build upon the storytelling revolution they had fostered.

After signing with the Random House imprint of Bantam Books, Weis and Hickman churned out the paperbacks at the steady pace of a novel every six to nine months: first came a Darksword trilogy, then a Rose of the Prophet trilogy. And then, beginning in 1990, came the seven-book Death Gate Cycle.

Dragon Wing, the first of the Death Gate novels. (You can’t ever go wrong with dragons, right?)

Like so many fantasy novels, the Death Gate books are most of all an exercise in world-building, and so your level of enjoyment of them must ultimately be determined by your level of interest in their elaborate castles in the air. As my regular readers know all too well by now, I haven’t been part of that target market for a long time. Nevertheless, here’s the capsule version:

The books involve two races of ultra-advanced beings, known as the Patryns and Sartans, who inhabit the far future of what would appear to be our own planet Earth. Long before the main story of the novels begins, these two races went to war with one another. The Sartans won the war, whereupon they placed the remaining Patryns in an inter-dimensional prison known as the Labyrinth. For rather convoluted reasons, they then sundered the Earth into four separate worlds, each based upon one of the Aristotelian elements — air, water, earth, and fire — and each inhabited mostly by less godly races, whom the Patryns and Sartans traditionally call the Mensch. So much for the backstory. The books proper describe how the Patryns and Sartans finally reconcile and heal their shattered realm. Weis and Hickman were just publishing the fifth of the seven books when Glen Dahlgren over at Legend decided that he’d like to make a game out of them.



Dahlgren had originally been hired to work at Legend as an assistant programmer. But, as was typical of most employees of the tiny company, his responsibilities bloomed in all sorts of unexpected directions after he arrived there. A talent for music led him to become Legend’s in-house sound man, responsible for soundtracks and sound effects alike. A talent for organization led him to become Legend’s quality-control man, overseeing a far-flung network of outside testers. And then he was selected to become one of the three men who put together Gateway, a project consciously engineered by Bob Bates, who had co-founded Legend alongside Mike Verdu, as a boot camp for training up new designers. After that, Dahlgren did half of the design work on Gateway 2, before being given permission to make his own game from start to finish; that game became Death Gate. Bates and Verdu put an enormous amount of faith in him — the sort of faith which one usually only sees manifested at small, young companies like this one, where bureaucracy is minimal and second-guessing rare. Whether or not Dalhgren realized it at the time, his bosses gave him a once-in-a-lifetime opportunity.

Of course, Legend had to secure the license before anything else could happen. Fortunately for everyone, this proved not to be terribly difficult. The negotiations were aided by the fact that Weis and Hickman remained enthusiastic gamers themselves; they made sure that their agent and publisher got the deal done with very little hassle, and even agreed to provide an original novella to include in the box. Like Bates and Verdu, they were almost bizarrely trusting of Dahlgren. They didn’t even blink when he proposed packing the entire story arc of the Death Gate series — some 3000 pages in text form — into a single game. Although only five of the planned seven novels had yet appeared when the project began, Weis and Hickman freely shared their plans for the remainder of the series — most importantly, for the big finale where the worlds would be unified once again.

In  what he called his “high concept” proposal for the game, Dahlgren laid out the plot in the broadest possible strokes:

This game is about a warrior whose race was imprisoned thousands of years ago in a place called the Labyrinth. His people were defeated by another race who broke up the world into four realms connected only by the Death Gate. In the game, the hero seeks to re-form the world by finding and assembling the scattered pieces of the World Seal.

None of it should work; it should feel like the Cliff’s Notes abridgement of the novels that it is. And yet Dalhgren made it work against all odds, largely by emphasizing the worlds of the books over the details of their plots. His most dazzling stroke was the insertion of the aforementioned World Seal as his McGuffin — one piece of it on each of the four worlds, in classic adventure-game fashion. With this scavenger hunt to serve as a straightforward motivation, you can get a taste of each world and interact with some of its characters without getting too bogged down in the more elaborate plot machinations of the books. Then fast-forward to the reunification, and you’re done.

This isn’t to say that well-nigh encyclopedic amounts of material from the books aren’t missing from the game. These leavings encompass not just countless details of the environments and cultures that Weis and Hickman created, but also nuances of theme and character. The central protagonist of both the novels and the game is a Patryn named Haplo, but he cuts a very different figure in the two mediums. In the books, he’s a morally ambiguous character at best, a cold fish full of prejudice and at times outright malevolence; in the game, he’s a naïve innocent whose mistakes all arise from the deceptions perpetrated upon him by his master Xar, a fellow the player can see is up to no good about five minutes after meeting him for the first time. It’s telling that Dahlgren refers to Haplo as his game’s “hero” in the extract above; few readers of the books would choose that word. To wit: in the first of the novels, Haplo deliberately engineers a war on the world known as Arianus; in the game, he prevents one instead with a dose of good old-fashioned peace, love, and understanding. Johnny L. Wilson, the editor-in-chief of Computer Gaming World magazine as well as precisely the fan of the books that I am not, wrote in his otherwise positive preview of the game that “I just wish Haplo had a little more anger in him.”

Sometimes Death Gate can feel a bit like a vintage Star Trek episode, with Haplo in Captain Kirk’s role of the man who swoops in to solve centuries of conflict between two commercial breaks.

Nevertheless, even Wilson admitted that Death Gate passed the acid test of any adaptation: if you didn’t know that the game was based on a series of books before playing it, you wouldn’t feel like you were missing anything. It truly works as a self-contained experience, while still capturing the feel of its source material. It manages to present the complex cosmology of the novels in an intuitive, natural way, without ever burying you under info-dumps.

Death Gate has much else going for it as well. By the time of its release in late 1994, Legend had quietly turned into the most consistent adventure studio in the industry, with a design aesthetic defined by an absolute commitment to fairness which even LucasArts couldn’t match. If at times this commitment could lead Legend to play it a bit too safe — Legend’s puzzles seldom evince the inspired lunacy of LucasArts at their best — it’s vastly preferable to the other extreme, as exemplified by the sloppy designs of Sierra, their other major competitor. Glen Dahlgren recently wrote on the subject, using terms that seem suspiciously similar to the design philosophy which Bob Bates has repeatedly explicated in his own writings and lectures:

Every puzzle should provide an “Aha!” moment, where either you feel smart for having figured it out, or you discover the answer elsewhere and smack your head for NOT figuring it out. If you don’t have enough information in the game to figure out the solution logically, you never get that moment. In an adventure game, that moment is the fun!

The design isn’t intended to torment you or to “beat” you (the designer isn’t even there to enjoy besting the player, so it isn’t a satisfying goal to begin with). My philosophy has always been that a player puts his trust in the designer to take care of him. Once the player has lost that trust, he’ll give up.

So, while there are no puzzles in Death Gate that you’ll remember for years — or even days — afterward, there are likewise none that unduly block your progress through the story. Thanks to its conversations that can go on for minutes at a time, it plays almost as much like an interactive storybook as a puzzle-solving adventure game.

Is this a puzzle I see before me?

Which isn’t to say that it doesn’t have its share of more conventionally gamey pleasures as well. A highlight is its rune-based magic system, which smacks of the classic CRPGs Dungeon Master and Ultima Underworld more than that of any other adventure game with which I’m familiar. Combining the runes into magical recipes, then trying said recipes out on anything and everything, is a consistent source of fun. A surprising number of the silly things you can try are implemented, usually to very amusing effect. You can’t lock yourself out of victory without knowing it when doing stuff like this, and, while you can die, resurrection is always just a click of the “undo” button away. Thus you can experiment to your heart’s content.

Mucking about with magic runes…

The interface is the one which Legend introduced in Companions of Xanth: fully point and click now rather than parser-based, but still retaining a pronounced literary feel that betrays Bates and Verdu’s original conception of Legend as the heirs to Infocom’s legacy. Although the characters you meet are voice-acted, descriptions and narrations are still presented only in text.

It’s easy to see which screens (like this one) are rendered in SVGA and which screens (like the previous one) are not.

Death Gate was, however, the first Legend game to make partial use of SVGA graphics, at a resolution of 640 X 480 rather than the 320 X 200 of VGA. The difference is marked. In fact, I’d go so far as to call the transition to SVGA the most significant transformation in computer-game audiovisuals since the Commodore Amiga appeared on the scene in 1985. SVGA games of the 1990s — the non-3D ones, at least — no longer have to be graded on a curve when it comes to their visual aesthetics. Much of the art in Death Gate would fit perfectly well into the latest indie sensation on Steam.

In another first for Legend, the opening movie and some of the later cut scenes are pre-rendered 3D, built using 3D Studio, a tool which was making a steady march across the games industry at the time.

All of these good qualities serve to make the rather negative review of Death Gate which Computer Gaming World published after Johnny Wilson’s positive preview seem all the more jarring. Peter Olafson dinged the game for its supposed inconsistencies of tone, for thoughtlessly mixing the serious with the comedic. This really is a problem with many adventure games then and now, which tend to collapse into comedy as a crutch for their ridiculously convoluted puzzles. And yet I don’t see Death Gate as one of those games at all. On the contrary: it strikes me as doing an admirable job of sticking to its guns and avoiding this tendency. The humor here is actually of a piece with the humor of the source material, as even Olafson admits in a tossed-off sentence that desperately needs further elaboration: “Even though this comic relief is present in the books, it seemed distracting and inappropriate in the game.” (Why exactly is that, Mr. Olafson?) Whether or not one finds the befuddled old wizard Zifnab, who wandered into the Death Gate universe from a dimensional warp in the Dragonlance world of Krynn, as hilarious as he’s obviously meant to be, his portrayal in the game isn’t notably different from his portrayal in the novels.

But another of Olafson’s criticisms is more telling, so much so that I’d like to quote it in full here:

I occasionally was haunted by a feeling that Death Gate’s technology has outstripped its genre. All the amenities lavished on the game — the enormous reservoir of digitized speech, the SVGA graphics, the animated cut scenes — build expectations for a game mechanism to complement them. And as agreeable as the engine may be, the game proper is essentially the same old object-oriented adventuring: take everything that’s not nailed down and use it in a conspiracy to take everything that is. There is something inherently trivial about inhabiting a lavish world and being stranded simply by want of a certain item. [It] is rather like having an orchestra play “Row, Row, Row Your Boat.”

If there is an overarching criticism to be raised against Death Gate, it’s this lack of will to innovate beyond the traditional confines of its genre. Indeed, the same criticism can be leveled against Legend’s catalog as a whole. While their adventures were the most consistently fair and literate in the industry during the 1990s, they were far from groundbreaking in formal terms. But then again, this eschewing of dizzying leaps in favor of keeping their feet planted firmly on solid ground does much to explain just why their games remain so playable to this day. In an industry that often fetishizes vision at the expense of craft, Legend’s games show how satisfying the tried and true can be when executed with thoroughgoing care.



Death Gate is a typical Legend game in another respect: it sold in reasonable but not overwhelming numbers; its sales figures remained in the five digits rather than the six or even seven to which the flashiest releases could aspire by 1994. Despite that, it wound up playing a role in Legend’s long-term evolution which was belied by its middling commercial performance.

Like many other small publishers, Legend found themselves increasingly threatened by the winds of change which blew through their industry as the 1990s wore on. The arrival of CD-ROM increased budgets enormously, thanks to the voice acting and richer and more elaborate graphic presentations that became not just possible but expected in this new era of 650 MB of storage. Meanwhile there arose a CD-ROM-fueled bubble in the marketplace, an ironic parallel to the bookware flash in the pan of exactly a decade earlier: titans of old media, from Disney to Random House, once again turned their attention to computer games as the potential Next Big Thing in publishing. How could little Legend hope to compete with the likes of them?

Well, as a wise person once said, if you can’t beat ’em, join ’em. By this point, Legend had established a solid relationship with Random House, first through their adaptations of Frederick Pohl’s Gateway novels and now with their work on Death Gate. With Random House eager to jump into computer games in a more concerted way, it made sense to both parties to cement the relationship — to, as they put it in their corporate-speak press release, “intensify the synergy established between them.” Mike Verdu spoke to me about it recently using more recognizably human language:

What drew Random House’s attention was the identity we had developed as a company that could take literary licenses and create properties out of them with the respect their authors thought they deserved. We had credibility as people who could take other authors’ worlds and build essentially new fictions in them, and take care of them and be a trusted partner in doing that. That had become sort of our trademark. As Random House was surveying what was then called “the new-media landscape,” they saw this as a very natural extension of their strength; it seemed like a very happy partnership.

Random House injected $2.5 million into Legend in the summer of 1994, a few months before Death Gate hit store shelves. The first fruits of the investment would appear the next year, when Legend released a three-CD full-motion-video extravaganza starring Michael Dorn, the actor best known as Worf from Star Trek: The Next Generation. That game cost far more to make than the ones Legend had been making before it, but its budget didn’t stem from arrogance or decadence. Legend was just trying to survive. And, accordingly to the conventional wisdom anyway, this was the only way to do it in the midst of the CD-ROM bubble.

In the end, the Random House deal would prove a mixed blessing at best. It gave the larger company a great deal of control over the smaller one — more control in reality than the latter had realized it would. It was structured to indemnify Random House against loss, such that Legend was on the hook to pay much of the investment back if they didn’t meet certain financial targets. For this reason not least, Random House would play an outsize role in the future in deciding which games Legend got to make. Not coincidentally, licensed games would come to dominate Legend’s output more than ever. In fact, during the entirety of their existence after the Random House deal Legend would release just one more game that wasn’t based on someone else’s preexisting intellectual property.

It was good, then, that they had a special talent for building upon the creativity of others. For they’d be doing a lot more of that sort of thing, sometimes for reasons that perhaps weren’t quite as pure as the honest love for the books in question that had spawned Gateway, Companions of Xanth, and Death Gate. Such was life if you wanted to run with the big boys.

(Sources: the book Designers and Dragons, ’70 to ’79 by Shannon Appelcline; Computer Gaming World of June 1993, November 1994, and February 1995; Questbusters 115; Retro Gamer 180. Online sources include Glen Dahlgren’s memoir of the making of Death Gate and Tracy Hickman’s interview at Mormon Artist. And my huge thanks to Bob Bates and Mike Verdu for their insights about Death Gate and all other things Legend during personal interviews.

Death Gate was briefly available for digital purchase at GOG.com, but has since disappeared again, presumably due to the difficulties associated with clearing the rights for licensed titles. It’s too large for me to host here even if I wasn’t nervous about the legal implications of doing so, but I have prepared a stub of the game that’s ready to go if you just add to the appropriate version of DOSBox for your platform of choice and an ISO image of the CD-ROM. A final hint: as of this writing, you can find the latter on archive.org if you look hard enough.)

 
 

Tags: ,

Superhero League of Hoboken

 

Please bear with me as I begin today with an anecdote from Beatles rather than gaming fandom.

Paul McCartney was having a rough time of it in 1973, three years after the Beatles’ breakup. He’d been thrown off balance by that event more than any of his band mates, and had spent the intervening time releasing albums full of far too many flaccid, underwritten songs, which the critics savaged with glee. They treated Wings, the new band he had formed, with the same derision, mocking especially the inclusion of Paul’s wife Linda as keyboard player. (“What do you call a dog with wings?” ran one of the uglier misogynist jokes of the time. “Linda!”) Then, just as it seemed things couldn’t get any worse, two of the three members of Wings who weren’t part of the McCartney family quit just as they were all supposed to fly to Lagos, Nigeria, to record their latest album.

It ought to have been the straw that broke the camel’s back. But instead, Paul, Linda, and their faithful rhythm guitarist Denny Laine went to Nigeria and delivered what many still regard as the finest album of McCartney’s post-Beatles career. “Paul thought, I’ve got to do it. Either I give up and cut my throat or get my magic back,” said Linda later. He did the latter: Band on the Run marked the return of the Paul McCartney who had crafted the astonishing medley that closed out Abbey Road, that parting shot of the greatest rock band ever. As Nicholas Schaffner wrote in The Beatles Forever, “Band on the Run more than sufficed to dispel the stigma of all that intervening wimpery. And the aging hippies all said: McCartney Is Back.”

I tell this story now because I had much the same feeling when I recently played Superhero League of Hoboken, Steve Meretzky’s game from 1994: “Meretzky Is Back.” And a welcome return it was.

Meretzky, you see, spent the period immediately following the breakup of Infocom in 1989 pursuing his own sort of peculiarly underwhelming course. The man who had strained so hard to bring real literary credibility to the medium of the adventure game in 1985 via A Mind Forever Voyaging spent the early years of the 1990s making lowbrow sex comedies seemingly aimed primarily at thirteen-year-old boys. Spellcasting 101: Sorcerers Get All the Girls, the first of the bunch, was defensible on its own merits, and certainly a solid commercial choice with which to kick-start Legend Entertainment, the company co-founded by former Infocom author Bob Bates with the explicit goal of becoming the heir to Infocom’s legacy. By the time of Spellcasting 201 and Spellcasting 301, however, the jokes were wearing decidedly thin. And the less said the better about Activision’s Leather Goddesses of Phobos 2, the comprehensively botched graphical sequel to arguably Meretzky’s best single Infocom text adventure of all. By 1992, it was time for a change. Oh, boy, was it time for a change.

Thus it was a blessing for everyone when Meretzky decided not to make the fourth game in what had been planned as a Spellcasting tetralogy, decided to do something completely different for Legend instead. The change in plans was readily accepted by the latter, for whom Meretzky worked from home on a contract basis. For, in a surprising indication that even the timeless marketing mantra that Sex Sells isn’t complete proof against a stale concept, sales of Spellcasting 301 hadn’t been all that strong.

The concept for Meretzky’s next game was fresh in comparison to what had come immediately before it, but it wasn’t actually all that new. In 1987, after finishing Stationfall for Infocom, he’d prepared a list of eight possibilities for his next project and circulated it among his colleagues. On the list was something called Super-Hero League of America:

When Marvel Comics asked if we’d be interested in a collaboration, I thought, Steve, old buddy, old pal, you could think up a lot more interesting and weird and fun superheroes than those worn-out boring Marvel Comics superheroes. Such as Farm Stand Man, who can turn himself into any vegetable beginning with a vowel. Dr. Madmoiselle [sic] Mozzarella, who can tell the toppings on any pizza before the box is even opened! I see this as a Hitchhiker’s/Rashomon type game in which you can play your choice of any of half a dozen superheroes. The story would be slightly different depending on which one you chose. If you elected to portray Annelid Man (able to communicate with any member of the worm family), you wouldn’t command as much respect as Dr. Asphalt (able to devour entire eight-lane highways), and the other superheroes wouldn’t obey you as readily. Potential for lots of interesting puzzles. Possible RPG elements.

“I like this!” scribbled one of his colleagues on the memo. “The superheroes shouldn’t be so silly, though… maybe?” But in the end, Meretzky wound up doing the safest project on the list, yet another Zork sequel. After releasing a whole pile of unique and innovative games over the course of 1987, to uniformly dismal commercial results, Infocom just didn’t feel that they had room to take any more chances.

Still, the idea continued to resonate with its originator, even through all of the changes which the ensuing five years brought with them. In late 1992, having decided not to do another Spellcasting game, he dusted it off and developed it further. He had the brilliant brainstorm of setting it in Hoboken, New Jersey, a real town of 35,000 souls not without a record of achievement — it’s the birthplace of baseball, Frank Sinatra, and Yo La Tengo — but one whose very name seems somehow to hilariously evoke its state’s longstanding inferiority complex in relation to its more cosmopolitan neighbor New York. Hoboken was the perfect home for Meretzky’s collection of low-rent superheroes with massive inferiority complexes of their own. Even more notably, the “possible RPG elements” of the first proposal turned into a full-fledged adventure/CRPG hybrid, a dramatic leap into unexplored territory for both Mereztky and Legend.

It was new territory for Meretzky the game designer, that is, but not for Meretzky the game player: he had long been a fan of CRPGs. Among the documents from the Infocom era which he has donated to the Internet Archive are notes about the games from other companies which Meretzky was playing during the 1980s. One finds there pages and pages of lovingly annotated maps of the likes of Might and Magic. Meretzky:

I’d been wanting to make an RPG for many years. But I thought that the usual Tolkienesque fantasy setting and trappings of RPGs had been done to death, and it occurred to me that superheroes was an excellent alternate genre that worked well with RPG gameplay, with superpowers substituting for magic spells. I originally planned to make it a full RPG, but Legend had never done anything that wasn’t a straight adventure game and were therefore nervous, so the only way I could convince them was to make it an RPG/adventure hybrid.

Meretzky’s characterization of Legend here is perhaps a touch ungenerous. They were a small company with limited resources, and were already in the process of moving from a parser-based adventure engine to a point-and-click one. Adapting it to work as a CRPG was a tall order.

Indeed, Superhero League of Hoboken remained in active development for more than eighteen months, longer than any Legend game to date. In the end, though, they succeeded in melding their standard graphic-adventure interface to a clever new combat engine. By the time the game was released in the summer of 1994, Meretzky had already moved on from Legend, and was working with fellow Infocom alum Mike Dornbrook to set up their own studio, under the name of Boffo Games. As a parting gift to Legend, however, Hoboken could hardly be beat. It had turned into a genuinely great game, Meretzky’s best since Stationfall or even Leather Goddesses of Phobos.

It takes place in a post-apocalyptic setting, a choice that was much in vogue in the mid-1990s; I’m now reviewing my third post-apocalyptic adventure game in a row. But, whereas Under a Killing Moon and Beneath a Steel Sky teeter a little uncertainly between seriousness and the centrifugal pull that comedy always exerts on the adventure genre, Hoboken wants only to be the latter. It’s extravagantly silly, so stupid that it’s smart — Meretzky at his best, in other words.

The premise is that a considerable percentage of the population have become “superheroes” in the wake of a nuclear war, thanks to all of the radiation in the air. But most of the actual superpowers thus acquired are, shall we say, rather esoteric. For example, you play the Crimson Tape, whose superpower is the ability to create organizational charts. That makes you ideal for the role of leader of the Hoboken chapter of the Superhero League. Your gang there includes folks like the Iron Tummy, who can eat spicy food without distress; the ironically named Captain Excitement, who puts others to sleep; Robo Mop, who can clean up almost any mess; Tropical Oil Man, who raises the cholesterol levels in his enemies; the holdover from the Infocom proposal Madam Pepperoni, who can see inside pizza boxes; and my personal favorite, King Midas, who can turn anything into a muffler. (For my non-American readers: “Midas” is the name of a chain of American auto-repair shops specializing in, yes, mufflers.) Some of these superpowers are more obviously useful than others: Captain Excitement’s power, for example, is the equivalent of the Sleep spell, that staple of low-level Dungeons & Dragons. Some of them are sneakily useful: the game’s equivalent of treasure chests are pizza boxes, which makes Madam Pepperoni its equivalent of your handy trap-detecting thief from a more ordinary CRPG setting. And some of the superpowers, including your own and that of many others, are utterly useless for fighting crime — until you stumble upon that one puzzle for which they’re perfect.

The game has a smart and very satisfying structure, playing out in half a dozen chapters. At the beginning of each of them you’re given a to-do list of five tasks in your superhero headquarters. To accomplish these things, you’ll naturally have to venture out into the streets. Each chapter takes you farther from home and requires you to explore more dangerous areas than the last; by its end, the game has come to encompass much of the Northeastern Seaboard, from Philadelphia to Boston, all of it now plagued by radiation and crime.

Your handy to-do list for Chapter 1.

In the Spellcasting games, Meretzky had a tendency to ask the player to do boring and/or irritating things over and over again, apparently in the mistaken impression that there’s something intrinsically funny about such blatant player abuse. It’s therefore notable that Hoboken evinces exactly the opposite tendency — i.e., it seeks to minimize the things that usually get boring in other CRPGs. Each section of the map spawns random encounters up to a certain point, and then stops, out of the logic that you’ve now cleaned that neighborhood of miscreants. I can’t praise this mechanic enough. There’s nothing more annoying than trying to move quickly through explored areas in a typical CRPG, only to be forced to contend with fight after mindlessly trivial fight. Likewise, the sense of achievement you get from actually succeeding in your ostensible goal of defeating the forces of evil and making a place safe again shouldn’t be underestimated. Among CRPGs that predate this one, Pool of Radiance is the only title I know of which does something similar, and with a similar premise behind it at that; there you’re reclaiming the fantasy village of Phlan from its enemies, just as here you’re reclaiming the urban northeast of the United States. Hoboken is clearly the work of a designer who has played a lot of games of its ilk — a rarer qualification in game design than you might expect — and knows which parts tend to be consistently fun and which parts can quickly become a drag.

You explore the city streets — this CRPG’s equivalent of a dungeon — from a top-down perspective. This interface yields to a separate interface for fighting the baddies you encounter, or to the first-person adventure-game interface when you wander onto certain “special” squares.

The combat system makes for an interesting study in itself, resembling as it does those found in many Japanese console CRPGs more than American incarnations of the genre. It’s simple and thoroughly unserious, like most things in this game, but it’s not without a modicum of tactical depth. Each round, each character in your party can choose to mount a melee attack if she’s close enough (using one of an assortment of appropriately silly weapons), mount a ranged attack if not (using one of an equally silly assortment of weapons), utilize her superpower (which is invariably silly), or assume a defensive stance. Certain weapons and powers are more effective against certain enemies; learning which approaches work best against whom and then optimizing accordingly is a key to your success. Ditto setting up the right party for taking on the inhabitants of the area you happen to be exploring; although you can’t create superheroes of your own, you have a larger and larger pool of them to choose from as the game continues and the fame of your Hoboken branch of the Superhero League increases. But be careful not to mix and match too much: heroes go up in level with success in combat, so you don’t want to spread the opportunities around too evenly, lest you end up with a team full of mediocrities in lieu of at least a few high-level superstars.

Combat on the mean streets of Hoboken. Here we’re up against some Screaming Meemies (“members of a strange cult that worships the decade of the 1970s, identified by their loud cry of ‘Me! Me!'”) and Supermoms (“Bred for child-bearing and child-rearing characteristics by 21st-century anti-feminist fundamentalists”).

As you explore the streets of the city, you stumble upon special locations that cause the adventure side of the game’s personality to kick in. Here the viewpoint shifts from overhead third-person to a first-person display, with an interface that will look very familiar to anyone who has played Companions of Xanth, Legend’s first point-and-click graphic adventure. In addition to conversing with others and solving puzzles in these sections, you can visit shops selling weapons and armor and can frequent healers, all essential for the CRPG side of things. That said, the bifurcation between the game’s two halves remains pronounced enough that you can never forget that this is a CRPG grafted onto an adventure-game engine. Your characters even have two completely separate inventories, one for stuff used to fight baddies and one for stuff used to solve puzzles. Thankfully, each half works well enough on its own that you don’t really care; the adventure half as well marked a welcome departure from Meretzky’s recent tendency to mistake annoyance for humor, whilst offering up some of his wittiest puzzles in years.

Curing a disease contracted in the CRPG section by visiting a healer found, complete with gratuitous Infocom references, in the adventure section.

But by my lights the funniest part of the game remains the rogue’s gallery of superheroes and villains — especially the latter. These provided Meretzky with an opportunity to vent his frustration on a wide array of deserving targets. Some are specific, like Transistor Jowl, a clone of Richard Nixon, right down to his parting line of “You won’t have Transistor Jowl to kick around anymore” (delivered perfectly in the CD-ROM version by voice actor Gary Telles). And some are more generalized, like the marketing executives who chirp, “Let’s do lunch!” in their unflappable cluelessness as you dispatch them. Either way, the social satire here has the sharp edge that was missing from the Spellcasting games:

Junk Bond Amoeba: Environmental toxins have produced these one-celled creatures, twelve feet across, bent on engulfing food morsels and defenseless companies. Beware, for during combat they can divide by mitosis, doubling their number!

Espevangelist: Similar to televangelists of the 20th and 21st centuries, except that espevangelists require no broadcasting equipment to transmit their programming, since they can project their thoughts and words directly into the minds of those around them. In addition to the damage they can thus inflict, espevangelists have been known to separate weak-willed parties from their funds. They are even more dangerous if they FUNDRAISE.

By way of attacking this last-mentioned reprobate, you “reveal details about his affair with an altar boy,” and “all the tears in the world fail to save him.” And all the aging gamers said: Meretzky Is Back.

Unfortunately, Superhero League of Hoboken‘s course after its release was markedly different from that of Band on the Run. The game got a lot of support from the all-important Computer Gaming World magazine, including an extended preview and a very positive review just a couple of issues later that proclaimed it “the first true comedy CRPG ever”; this wasn’t strictly correct, but was truthy enough for the American market at least. And yet it sold miserably from the get-go, for reasons which Legend couldn’t quite divine. Legend was no Sierra or Electronic Arts; they averaged just two game releases per year, and the failure of one of them could be an existential threat to the whole company.

But they got lucky. Just after Hoboken‘s release, the book-publishing titan Random House made a major investment in Legend; they were eager to make a play in the new world of CD-ROM, and, having been impressed by Legend’s earlier book adaptations, saw a trans-media marketing opportunity for their existing print authors and franchises. This event took some of the sting from Hoboken‘s failure. Random House’s marketing consultants soon joined in to try to solve the puzzle of the game’s poor performance, informing Legend that the central issue in their opinion was that the cover art was just too “busy” to stand out on store shelves. This verdict was received with some discomfort at Legend; the cover in question had been the work of Peggy Oriani, Bob Bates’s wife. Nevertheless, they dutifully went with a new, Random House-approved illustration for the CD-ROM release, splashed with excerpts from the many glowing reviews the game had received. It didn’t help; sales remained terrible.

The revised CD-ROM box art. (The original can be seen at the beginning of this article.)

Steve Meretzky would later blame the game’s failure on its long production time, which, so he claimed, made it look like a musty oldie upon its release. And indeed, it was the last Legend game to use only VGA rather than higher resolution SVGA graphics. Still, and while the difference in sharpness between this game’s graphics and Legend’s next game Death Gate is pronounced, Hoboken really doesn’t look unusually bad among a random selection of other games from its year; there were still plenty of vanilla VGA games being released in 1994 and even well into 1995 as software gradually evolved to match the latest hardware. The real problem was likely that of an industry that was swiftly hardening into ever more rigid genres, each of which came complete with its own set of fixed expectations. An adventure game with hit points and fighting? A CRPG with no dungeons or dragons, hurling social satire in lieu of magic spells? Superhero League of Hoboken just didn’t fit anywhere. As if all that wasn’t bad enough, unlicensed superhero games of any stripe have historically struggled for market share; it seems that when gamers strap on their (virtual) spandex suits, they want them to be those of the heroes they already know and love, not a bunch of unknown weirdos like the ones found here.

A few months after the release of the CD-ROM version, Legend received a cease-and-desist letter out of the blue from Marvel Comics. It seemed that Marvel and DC Comics were the proud owners of a joint trademark on the name of “superhero” when used as part of the title of a publication. (This sounds to my uneducated mind like a classic example of an illegal corporate trust, but I’m no lawyer…) While there was cause to question whether “publications” in this sense even encompassed computer games at all, it hardly seemed a battle worth fighting given the game’s sales figures. Already exhausted from flogging this dead horse, Legend worked out a settlement with Marvel whereby they were allowed to continue to sell those copies still in inventory but promised not to manufacture any more. In the end, Superhero League of Hoboken became the least successful Legend game ever, with total sales well short of 10,000 copies — a dispiriting fate for a game that deserved much, much better.

That fate makes Hoboken a specimen of a gaming species that’s rarer than you might expect: the genuine unheralded classic. The fact is that most great games in the annals of the field have gotten their due, if not always in their own time then in ours, when digital distribution has allowed so many of us to revisit and reevaluate the works of gaming’s past. Yet Superhero League of Hoboken has continued to fly under the radar, despite its wealth of good qualities. Its sharp-edged humor never becomes an excuse for neglecting the fundamentals of good design, as sometimes tends to happen with forthrightly comedic games. It’s well-nigh perfectly balanced and perfectly paced. Throughout its considerable but not overwhelming length, its fights and puzzles alike remain challenging enough to be interesting but never so hard that they become frustrating and take away from its sense of fast-paced fun. And then it ends, pretty much exactly when you feel like you’re ready for it to do so. A lot of designers of more hardcore CRPGs in particular could learn from this silly game’s example of never exhausting its player and refusing to outstay its welcome. The last great narrative-oriented game of Steve Meretzky’s career, Superhero League of Hoboken is also the one most ripe for rediscovery.

Some pop-culture references are truly timeless…

(Sources: the books The Beatles Forever by Nicolas Schaffner and Game Design Theory & Practice (Second Edition) by Richard Rouse III; Computer Gaming World of August 1994 and October 1994; Questbusters 113. Online sources include “The Superhero Trademark FAQ” at CBR.com and “Super Fight Over ‘Superhero’ Trademark” at Klemchuk LLC. I’m also greatly indebted to the indefatigable Jason Scott’s “Infocom Cabinet” of vintage documents from Steve Meretzky’s exhaustive collection. And my huge thanks to Bob Bates and Mike Verdu for their insights about Superhero League of Hoboken and all other things Legend during personal interviews.

Superhero League of Hoboken is available for digital purchase on GOG.com. This is wonderful on one level, but also strange, as it should still be subject to the cease-and-desist agreement which Legend signed with Marvel Comics all those years ago. There is reason to question whether Ziggurat Interactive, the digital publisher currently marketing this game, actually has the right to do so. I leave it for you to decide the ethics of purchasing a convenient installable version of the game versus downloading a CD-ROM image elsewhere and struggling to set it up yourself. Believe me, I wish the situation was more clear-cut.)

 
 

Tags: , ,

Beneath a Steel Sky

I would rather see a personal vision onscreen than filmed live-action. I have an idea that with CD technology there are going to be a lot of little-known actors photographed and appearing on our screens. I think if you have a graphic artist involved, you get something even better than reality.

— Dave Gibbons

There’s no reason why hundreds of people in California should know the future any better than ten people based in Yorkshire.

— Charles Cecil

Charles Cecil

Charles Cecil was a part of the British adventure-games scene from the beginning. Born in 1962, he began studying engineering at Manchester University in 1980. There he became friends with a fellow student named Richard Turner, who had just co-founded Artic Computing, one of the very first suppliers of software for the Sinclair ZX80, Britain’s very first mass-market personal computer. Although he was not and never would become a programmer, Cecil got pulled into other aspects of the venture, such as drawing what he describes today as “the shittiest logo.”

Chris Thornton, Richard Turner’s partner in Artic, owned an imported Radio Shack TRS-80; this allowed the group of friends to keep tabs on the American microcomputing scene, which had a few years’ head start on the British. Taking note of the success that Scott Adams was having with his text adventures in the United States, Artic developed an engine for similar games on Sinclair machines. In June of 1981, Turner and Thornton’s Adventure A: Planet of Death became the first home-grown adventure game ever to be sold in Britain.

As the name of that first game would imply, Artic intended from the beginning to make a whole line of text adventures, just as Scott Adams had done. “You like telling stories,” Turner said to Cecil. “Why don’t you write one?” Thus Cecil designed Adventure B: Inca Curse, followed by several more text adventures, all primitive enough — or, if you like, minimalist enough — to fit into a computer with just 16 K of memory. A game designer had been born, alongside a cottage industry of similarly ramshackle semi-professional text adventures that would persist for the better part of two decades. (Artic’s games were particularly noted for their atrocious spelling…)

Cecil continued to design games and do various other odd jobs for Artic for several years, but by the middle of the decade the company’s homespun products were finding the going tough in what had now become a crowded and hyper-competitive British software market. In 1985, Cecil jumped from the sinking ship to found his own Paragon Programming, which specialized in porting American games to European platforms. Two years later, he parlayed that into a short-lived gig as development manager for US Gold, then a longer-lived one in the same role for Activision’s European subsidiary.

But a series of unfortunate events and poor management decisions at the American parent company — a trend which began about the time of Cecil’s arrival, with management’s decision to change the company’s name to the hopeless corporatese “Mediagenic” — ultimately spelled disaster for that international software empire. In 1990, the 27-year-old Charles Cecil, who had recently been enjoying such luxuries as a posh company car and a mobile phone, was left high and dry by Mediagenic’s collapse. What to do now?

All his time spent porting and selling American games had given him a familiarity with goings-on across the Atlantic that was unusual among his countrymen. The one area of gaming where the Americans most obviously outdid the Brits, he realized, was the genre he still loved best: the adventure game. British and, indeed, most European developers had little that could compete with the latest graphic adventures from American publishers like Sierra and Lucasfilm Games. There was a reason for this: thanks to their need for large amounts of single-use visual and audio assets, those games were among the most expensive of all to produce; European studios for the most part simply lacked the resources to make them. The one partial exception to this rule came in the form of a few French studios like Delphine, who made games that were beautiful to look at if often atrociously designed. But Britain had absolutely nothing on offer.

So, Cecil decided for the second time in his young life to found his own company, with the intention of changing that — this despite the fact that he had very little money at all to work with even by the modest standards of British game development. He started Revolution Software in March of 1990 on the back of a £10,000 loan from his mother, and took up residence in an unheated cubbyhole above a fruit market in the struggling city of Hull — “We chose Hull because it was cheap,” admits Cecil — with a few of the folks he’d met during his previous travels through the British games industry. The setting verged on the Dickensian; during the winter months, they would huddle against their computers to try to stay warm.

Still, Cecil did soon convince the British publisher Mirrorsoft to provide some minimal funding for Revolution’s first game in return for the publication rights to the eventual finished product. When Mirrorsoft collapsed in the wake of the suspicious death of its kingpin Robert Maxwell and the postmortem revelation of financial improprieties throughout his organizations, they moved on in fairly short order to Virgin Games — a better partner on the whole, as Virgin came complete with a North American branch.

The core team at Revolution in the early days: Tony Warriner, Adam Tween, David Sykes, Stephen Oades, Dave Cummins, and Charles Cecil.

Tacitly admitting that it would be difficult indeed for a shoestring operation like theirs to compete with a company like Sierra in terms of production values, Revolution settled on a concept and engine to power it which they called “Virtual Theatre.” They envisioned it as nothing less than the next great leap in adventure design. Cecil described it thusly at the time:

Within each game, time advances and people walk around with their own routes: the blacksmith will go into his forge and hammer away, then he’ll go into the pub to have a drink and he’ll talk to other people around the village. You could have fifteen people all walking around, all interacting with each other. So instead of being a game where you’re the key and everything reacts to you, we have a game where you’re just another person.

It was a noble vision in its way, one which aimed to push the frontiers of an oft-hidebound genre. And yet, for all that it reads well on paper, it would prove more than problematic in practice. The disadvantage of making a world which runs along of its own accord is that it can run merrily away without the player, leaving her stranded in some plotting cul de sac. And then, far from being a drawback, most players enjoy adventure games precisely because they let them be the star of the show. After all, If one wants a world where one is “just another person,” one generally need only look up from the computer.

When Cecil expanded yet further on his vision, he wound up in a place to which many designers have dreamed of venturing since the heyday of commercial text adventures, but which has yet to yield a single comprehensively satisfying game: “What we’re planning to do in the future is put in artificial intelligence whereby we set the basic parameters and then we let the characters decide what they’re going to do themselves. Fundamentally, anything could happen.”

Unsurprisingly, then, the first Revolution game — the one which most wholeheartedly embraced the Virtual Theatre concept — also proved to be the worst one they would ever make. Lure of the Temptress combined a clichéd fairy-tale setting with an awkward interface, sub-Sierra graphics, and well-nigh infuriating gameplay, which mostly entailed chasing all of those vaunted self-directed characters hither and yon through a plot line littered with potential dead ends. Published internationally by Virgin Games for the Commodore Amiga, Atari ST, and MS-DOS in the spring of 1992, it sold in reasonable quantities, doing best with Amiga owners in Europe. Charles Cecil didn’t hesitate to wave the flag on behalf of the continent. “I believe that European graphic artists are the best,” he said — an assertion which the graphics in Lure of the Temptress utterly failed to prove. Thankfully, better things were still to come from Revolution.

Lure of the Temptress did earn enough money to fund a move to better offices in York, with a corresponding uptick in the budget for their next game. Even so, much of the dramatic improvement evinced by said game was the result of a series of chance events that won Revolution the services of arguably the most respected comic-book illustrator of the era. And yes, he was a European. In fact, he hailed from Britain.

In May of 1989, a popular British gaming magazine known as The One published a feature about Watchmen, a two-year-old book which had done much to inculcate the idea of the graphic novel as a respectable literary form. Amidst much speculation about a potential Watchmen film and game — neither of which would appear until decades later — the article somehow managed to avoid mentioning the name of Dave Gibbons, the man who had drawn writer Alan Moore’s story. Understandably annoyed, Gibbons wrote to the magazine to point out the fact of his existence.

Dave Gibbons

By way of apology, The One sent Gibbons a Commodore Amiga and a copy of Deluxe Paint, then devoted five pages to an interview featuring his impressions of those things and many others. As the fact that he had seen the first Watchmen article in the magazine in the first place would indicate, Gibbons was already following the latest developments in computer gaming fairly closely. (In this respect and in many others, the down-to-earth Gibbons was unlike his sometime partner Alan Moore, an unrepentant eccentric and dyed-in-the wool Luddite.)

It seems that computer games are finding their own level in the same way as comics. I think that a lot of games, like a lot of comics it must be said, are pretty banal, and pretty repetitive — sort of like chewing gum. They won’t do you any harm, but on the other hand they aren’t likely to do much good.

I find puzzle games the most interesting. And the flight simulations… Falcon’s brilliant. You get to the point where you think you are there and you find yourself leaning in the chair. Rocket Ranger is very interesting stuff, that to me is like those role-playing gamebooks. It’s a different game every time you play it.

The magazine’s earlier slight was forgiven; Gibbons went on to draw the cover art for at least one issue of The One. More importantly, he met Charles Cecil through the magazine; Cecil was still with Mediagenic at the time and was also chummy with the staff at The One. The two started tentatively to feel one another out, until finally, after making some suggestions here and there for Lure of the Temptress, Gibbons agreed to become the principal illustrator and art director of Beneath a Steel Sky, Revolution’s second game. Not only did he bring his unique talents to the game itself, but the presence on the team of such a high-profile individual did much to drum up interest in the press. Cecil tells of the many journalists who came to the trade shows to meet Gibbons and see the game, in that order. They “began pulling out copies of the old Watchmen comics and Dave spent a while signing the lot. It was very positive, and they were dying to see what he had created in the game.”

Charles Cecil’s games have never been notable for the originality of their subject matter, and Beneath a Steel Sky is no exception to that rule of derivation. It trades in the King’s Quest-like fantasy of Lure of the Temptress for a dystopic science-fiction setting with strong cyberpunk overtones — a mixture of Blade Runner and Neuromancer, not exactly a rare blending among games of the early 1990s. Union City, where this game takes place, is the familiar authoritarian technocracy, a place where class strata have taken on a literal dimension. One has to take originality in such a setting where one can find it: upending a science-fictional trope stretching back at least to Fritz Lang’s classic silent film Metropolis, in Union City the poor and powerless live out their squabbling lives in tenements that scrape the sky, while the rich and powerful live in luxury near ground level. Union City’s most unusual wrinkle of all is the fact that it exists in the far-flung locale of Australia instead of some faded North American or European hegemony. Yet even this fact is disarmingly easy to miss entirely, especially if you happen to be playing the voice-acted CD-ROM version with its many distinctly British and American accents.

You play a young man named Robert Foster, who as the introduction begins lives with one of the nomadic tribes that inhabit a place known as The Gap, the vast wasteland separating the cities of Australia. (Said wasteland is known as the Outback today…) But then a military raid kills everyone in the tribe except Foster himself; he is spared, to be spirited away by helicopter to Union City for reasons unknown. He escapes when the helicopter crashes over the city before it can reach its final destination, whereupon the game proper begins. As Foster, you must elude your pursuers as you explore Union City’s nooks and crannies, must learn the secret that makes you of such special interest to the powers who control the city — and must bring about their downfall.

Undoubtedly the strongest aspect of the game — the one thing you’re guaranteed to still remember even years after playing it — is Dave Gibbons’s art. Despite his earlier well-publicized experiments with Deluxe Paint, he elected to draw all of the approximately 90 background scenes for which he was responsible using the same old analog techniques that he had used to bring Watchmen and countless other comics to light. He provided pencil sketches of each scene to Revolution, where an artist named Les Pace, a veteran of such Hollywood productions as Who Framed Roger Rabbit, proceeded to color them in by hand. Only then were the illustrations scanned in on an Apple Macintosh, that being the most affordable platform at the time with good support for 24-bit color. Finally, these “master plates” could be down-sampled to come within the capabilities of Revolution’s two primary target platforms for the finished game: MS-DOS machines with VGA graphics cards (which allowed a maximum of 256 onscreen colors) and the Commodore Amiga (which allowed just 32).

Even in these degraded forms, the game’s imagery is striking. Inspired to some extent by the collapsing factories of hardscrabble Hull, Revolution Software’s original home, Union City manages to be varied but also of a piece, dingy but also coldly clinical, a warren of boldly vertiginous drops and furtively claustrophobic corners. Unlike many games during this era of exploding technological innovation, when the desire for spectacle could often overwhelm consistency and coherence, there’s a thoroughgoing visual aesthetic to Beneath a Steel Sky that stems from something more than a desire to show off the technology that powers it. Charles Cecil’s comment on the subject stood out in an era obsessed with photo-realism in games: “We’re not trying to create reality. We’re trying to create a style.”

The Process


Dave Gibbons sketched each background on paper, just as he would have a comics illustration…

…to produce something like this.

Still working on paper, Les Pace painted the sketch.

Finally, it was scanned in in 24-bit color. This master copy was then down-sampled to 256 colors (MS-DOS) or 32 colors (Amiga) for inclusion in the game. (The image above is from the Amiga version; those below are from the higher-fidelity MS-DOS version.)

The End Results




The writing in the game is a touch weaker than its visuals; scriptwriter Dave Cummins isn’t incompetent by any means, but nor is he another Alan Moore. As tends to happen constantly in the adventure genre, the overarching “dark, serious” plot gets immediately overrun in the details by a collapse into comedy, a genre which seems far better suited to the outlandish puzzles that are the driving force of most adventure games, this one included.

Still, the blow of this failure of the game to stick to its dramatic guns is eased immensely simply because a lot of the humor is really, truly funny; it never feels forced, something which is by no means the case in all or even most of this game’s competitors. This is wry British humor at its best: it’s sneakily smart, and also a bit more deviously risque than what you might find in a contemporary American game of this ilk. (One running gag, for example, has to do with a skeezy character’s collection of “pussy pictures” — which, yes, turns out just to be pictures of cats.) You begin the game with a sidekick already in your inventory: your childhood friend Joey, a synthetic personality on a circuit board who can be transplanted into various robots as you go along. His sarcastic banter is a great source of fun and oblique hints, such that when he’s not with you in some sort of embodied form you genuinely miss him. In fact, I’d like the game even more if it had more of him in it. He’s prevented from joining the absolute highest ranks of classic adventure-game sidekicks only by the fact that he’s onscreen less than half the time.

If you hate convoluted adventure-game puzzles on principle, the ones here will do nothing to convince you otherwise. If you enjoy them, on the other hand, Beneath a Steel Sky is a solid implementation of their ilk. It’s not a particularly easy game, but nor is it an unusually hard one for its time, and it is consistently logical in its silly adventure-game way. (In this sense as in several others, it stood head and shoulders above its few competitors among homegrown British graphic adventures, whose grasp on the fundamentals of good game design tended to be shaky at best.) It eschews the contemporaneous interactive-movie trend, with its chapter breaks and extended cut scenes, for a more old-school non-linear approach; for the bulk of the game, you have a fairly large area to roam and multiple problems to work on. There’s never a sense that the puzzles were hasty additions inserted just to give the player something to do; they’re part and parcel of a holistic experience.

Vestiges of Revolution’s earlier rhetoric about creating more dynamic worlds do remain here. Characters are still a bit more active than you might find in a Sierra or LucasArts game, and an unusual number of the puzzles rely on analyzing their movements and timing your own actions just right. That said, the most frustrating aspects of Lure of the Temptress have been excised. For the most part, the designers opted to return to the things that were known to work in this genre rather than continuing to blaze problematic new trails — and it must be said that the game is all the better for it for their conservatism. Likewise, its straightforward one-click interface wasn’t hugely innovative in itself even at the time — this doing-away-with the old menu of verbs was becoming the norm in graphic adventures by this point — but it is a well-executed example of such an interface. All in all, if you like traditional graphic adventures, you’ll find this game to be a sturdy, perhaps occasionally inspired example of the genre.

Beneath a Steel Sky was a European game made at a time when the Commodore Amiga, although slowly sliding past its peak, was still the most popular gaming platform across much of that continent, and thus one that could not be safely ignored by any European studio. Make no mistake: the challenges of making a game that could run on an Amiga at the same time that it could stand on a reasonable par with the latest adventure games on American shelves were immense. The Amiga was slower than the latest MS-DOS machines and was lacking graphically by comparison, and most European Amiga owners didn’t even have a hard drive, much less a CD-ROM drive. And yet, remarkably, Revolution largely pulled it off. Beneath a Steel Sky shipped in March of 1994 on no fewer than fifteen Amiga floppy disks. You had to swap them constantly in order to play it on a machine without a hard drive, but it wasn’t quite aggravating enough to completely destroy the fun of the game itself for an Amiga-owning adventure fan.

Charles Cecil, whose nebbishy appearance concealed a surprisingly down-and-dirty sort of marketing savvy, cast the game not only as Britain’s answer to the adventures of Sierra and LucasArts but as the savior of adventure gaming writ large on the Amiga, coming as it did just as the aforementioned companies were abandoning the platform. He wasn’t above the occasional gratuitous slam against the Americans in interviews which he knew would remain safely ensconced on his side of the ocean: “Most American graphic adventures are a little shallow because the American public doesn’t see plot as important. However, European game players seem to want to think a lot more about what they’re doing, and we’ve tried to reflect that.” Some of his statements in this mode were just bizarre: “The engine Sierra [is] using is outdated. They introduced it five years ago and really haven’t developed it.” For the record, it should be noted that the five years in question encompass Sierra’s move from parser-driven games to point-and-click ones, along with the jump from 16-color EGA graphics to 256-color VGA and the addition of voice acting, just to list a few highlights. To further confuse the situation, Cecil was seeking and winning a contract from Sierra to port King’s Quest VI to the Amiga — something the American company otherwise had no plans to do — at the very same time he was making such comments. Naturally, the European magazines ate it up, awarding his game gold stars pretty much across the board.

Just a month later, Commodore declared bankruptcy. Beneath a Steel Sky was one of the last of its breed on the Amiga.

By way of completing the picture of a work at the crossroads between the old order and the new, Revolution released a voice-acted CD-ROM version for MS-DOS computers shortly after the floppy-based releases. The actors went for the most part uncredited, but it appears that Revolution didn’t look far from home for most of them. The eccentric citizens of Union City deliver their lines with gusto in broad Northern English, a nice contrast to the prim London accents of so many games. Their accents make the humor go down even better, and give the game that much more of a distinctive personality. Meanwhile an American refugee named Adam Henderson voices straight man Robert Foster in the neutral Midwestern tones of a prime-time news anchor, while most of the villains speak Brooklynese straight out of an episode of Law & Order. Go figure…

Helped along by positive reviews and the measure of hype which accompanied the involvement of Dave Gibbons, Beneath a Steel Sky rode Amiga loyalists in Europe and MS-DOS-computer-owning adventure fans in North America to solid sales numbers. Thus Revolution got to live on and make still more games, following a template which was the ironic opposite of their name: solidly constructed adventure games cut from a sturdy traditionalist cloth.

(Sources: Amiga Format of March 1993, December 1993, March 1994; AmigaWorld of December 1993 and August 1994; Amiga Computing of Christmas 1993 and June 1994; Computer Gaming World of July 1992; Computer and Video Games of July 1987 and January 1989; CU Amiga of March 1993 and January 1994; Edge of September 1993; Games TM 9; New Computer Express of August 4 1990; PC Review of May 1992; Questbusters 114; Retro Gamer 56 and 63; The One of May 1989, August 1989, March 1990, November 1991, February 1992, March 1993, and November 1993. Online sources include interviews with Charles Cecil on Gamasutra, Dining with Strangers, and MCV/Develop.

Charles Cecil and Revolution have released Lure of the Temptress and Beneath a Steel Sky as free downloads.)

 

Tags: , ,