RSS

Monthly Archives: September 2020

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.)

 
58 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 computers1 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 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.)


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

 
52 Comments

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

 

Tags: , , , , , ,