The Digital Antiquarian - 03 Jul 2026

Page 2 of 2

That said, Wright's next game did include more whimsy and humor than were to be found in the stolidly portentous SimEarth. Like so many of his other games, 1991's SimAnt: The Electronic Ant Colony was inspired by his readings in popular science: in this case, by the book The Ants by E.O. Wilson, the dark-horse winner of the 1991 Pulitzer Prize for General Nonfiction. This time, however, Wright was not able to work directly with the man whose theories he was trying to bring to interactive life on the computer. Wilson neglected to read closely the letter Wright sent him proposing a collaboration, and was thus left to find out about SimAnt after it was finished, just like everyone else. But when he did, he praised its "sophistication and precision," so it was all good.

E.O. Wilson was the man who first discovered that individual ants, although each possessed of only the most rudimentary forms of awareness, communicate over amazingly long distances using pheromones, resulting in a group intelligence that enormously transcends that of its constituent parts. Seen in a certain light, then, Wright's first three Sim games were all grappling with what he liked to called "distributed," communal forms of sentience: SimEarth on a macro level, SimAnt on a micro level, and SimCity somewhere in between. In SimAnt, you have to find the right ratio of breeders to workers to soldiers in the colony you build around your all-important queen. You also have to contend with a rival ant colony and other, bigger creatures that can wreak havoc on your insect society, such as a marauding tarantula.

Wright and the slowly growing team around him at Maxis did look for ways to make this simulation more relatable than SimEarth. If you need an action fix, you can take direct control of one of your soldier ants and charge into battle. Maxis even deigned to include a score and a high-score table. The ultimate achievement is to drive a hapless human and his dog right out of their house. Observing all of this, Computer Gaming World noted that there weren't actually all that many differences between the life of your ant colony and "an exciting space-opera story line": "The horrible, nearly indestructible insect-like monster in the movie Alien was, after all, a souped-up, acid-puking, survivalist ant from a mixed marriage with a scorpion."

But, for all that ravenous insects - not to mention hive minds - had been a staple of science fiction since the very beginning, the answer to the question of whether people wanted to spend hours on end seeing the world through insectoid compound photo-receptors proved mixed. Although SimAnt did considerably better than SimEarth in the long run, selling over 100,000 copies, Wright was vaguely disappointed that its primary demographic wound up being "ten-to-thirteen-year-old" boys, the group best primed to respond to its squishy sound effects and shadings of the grotesque: "I was expecting it to be more older people that would appreciate how amazingly interesting ants are as an example of distributed intelligence. But in fact, I ended up appealing to twelve-year-olds who just loved playing with ants." Ah, well... it's better to have the "wrong" customers than no customers at all. The latter would be the case to an alarming degree for the two games after SimAnt.

Those two games marked the first times that Maxis applied the Sim branding to the work of other designers than Will Wright. By now, the studio had developed into a utopian collective of thinkers and dreamers, who barely felt they were a part of the commercial games industry at all and definitely didn't feel like they had to abide by its usual rules. SimCity, that gift that kept on giving, ensured that, funding whatever esoteric experiments they wished to pursue.

A former Apple hardware engineer named Ken Karakotsios was responsible for SimLife, Maxis's most esoteric simulation yet, which abandoned all of the concessions to accessibility that were to be found in SimAnt. SimLife was responding to yet another buzz in popular science, in this case over the prospects for digital "cellular automata" - or, to use the title of a popular 1992 book by Steven Levy, Artificial Life.

This notion of self-replicating digital organisms was an old one in computer science even then, having originated in the 1940s with John von Neumann. It was given a practical spin circa 1970 by the mathematician John Conway, whose Game of Life allowed the user to set up collections of "cells" in different configurations on a computer and then watch to see what patterns they created as they lived and died. The claims that were made on the basis of simple experiments like this one were either boldly visionary or wildly overblown, depending on your point of view. But, as so often happens in fields of or adjacent to artificial intelligence, the boosters found it hard to explain how genuine consciousness was going to arise from their crude imitations of life.

This isn't to say that cellular-automata couldn't be useful and intriguing. Will Wright himself had employed some of these approaches in all three of his Sim games. Ken Karakotsios's implementation of them was at bottom a dramatically expanded version of The Game of Life, which permitted the single cells to evolve into all manner of plants and animals, then compete and cooperate in various simulated environments while the player tinkered from on-high as much or as little as she preferred. It was interesting enough conceptually even if one hadn't drunk all of the artificial-life Kool Aid - it was a decent simulation of the way that evolution is believed to work in the non-digital world as well - but the things that had made SimEarth problematic as a marketplace proposition were exaggerated even further here: it was all just so attenuated, so very abstract. And it didn't help that, on a superficial level at least, it was plowing so much of the same ground as SimEarth, which had also allowed you to watch lifeforms evolve, albeit in the context of a grander world simulation.

Computer scientist Christopher Langton, the most prominent academic booster of cellular automata at the time, said that SimLife was "awfully close to being a useful tool for scientific research." Yet it was competing with games, not with scientific tools. It became Maxis's worst-selling release to date.

SimFarm, which was conceived and designed by a former Maxis playtester named Eric Albers, was more down to earth in all respects. As Maxis's marketing copy put it, it was "SimCity's country cousin," simulating the life of a farm. But it turned out that digging irrigation ditches was less exciting than building roads, and that watching cows graze was less entertaining than watching human commuters go about their day. SimCity was kinetic; SimFarm just sat there baking in the sun, succeeding mostly in demonstrating why fleeing the farm for the bright lights and big city has long been such a staple of fiction and nonfiction. It didn't sell much better than SimLife.

A meeting of great gaming minds. Shigeru Miyamoto, Will Wright, and Jeff Braun in Japan, October 1989.

Indeed, SimEarth, SimAnt, SimLife, and SimFarm combined struggled to come within an order of magnitude of SimCity's total sales. Defying the usual logic of the computer-game marketplace, SimCity's sales increased rather than decreased year by year, as more and more people who weren't traditional gamers, hackers, or nerds brought computers into their homes, thanks to the excitement about CD-ROM and multimedia, the user-friendliness of the latest versions of Microsoft Windows, and, soon enough, the chatter about the burgeoning World Wide Web. SimCity subsidized everything else that Maxis did.

It had been widely ported from its first home on the Apple Macintosh. First it had made its way to the Commodore Amiga, then to MS-DOS, the platform where it really started to drum up the sales figures. After that came just about every other viable or semi-viable computing platform in the Western world, including even such aged refugees of an earlier computing epoch as the Commodore 64 (the machine on which Wright had coded his first prototype of the game), the Sinclair Spectrum, and the BBC Micro. But the next really key port after the MS-DOS one was to the new Super Nintendo Entertainment System, the much-anticipated successor to the best-selling single games console to date. When the SNES came to North America in August of 1991, SimCity was one of its launch titles, serving as an early demonstration that this console was capable of far more than the likes of Super Mario Bros.

It's therefore ironic to note that SimCity's biggest champion inside Nintendo was none other than the creator of Super Mario Bros., the legendary Shigeru Miyamoto. Wright had first bonded with him over the course of a week-long, exploratory visit to Japan in October of 1989. Nintendo then agreed to undertake the port themselves under Miyamoto's guidance, "Nintendoizing" the game without losing its essence. Miyamoto inserted his American friend into the proceedings as "Dr. Wright," who popped up from time to time to offer comments and advice, doing much to personalize a game that was, like all of Maxis's software toys, utterly bereft of identifiable characters in its original form. Having SimCity on the SNES, which quickly became a bigger self-contained market for games than all of the existing computer platforms combined, was another huge boon for Maxis, even if the more cartoony look wasn't going to do them any favors with the furled-brow Time magazine crowd.

At the same time that they were expanding onto a living-room console, Maxis was becoming a publisher of out-of-house games. These weren't generally given the Sim branding but did tend to be aligned with it in spirit. Maxis provided an entree to the American market for any number of international studios whose abstruse concepts had been given short shrift by other publishers. Their games ranged from A-Train, a Japanese-developed, finance-heavy Railroad Tycoon competitor that gave the lie to the idea that all Japanese games were simple and cutesy like Super Mario Bros., to a passive Russian "aquarium simulator" called El-Fish that was as proudly goalless - some might opt for the term "pointless" - as anything that ever sprang from the mind of Will Wright.

Many of these were worthy, even noble efforts in their way, even if they don't float your humble author's particular boat. But that fact kind of points out the problem with them: they were ultra-nichey titles that demanded players who were seriously interested in their unusual subject matter, at a time when the distribution systems of commercial computer gaming were not well-suited to such products. Maxis struggled even to get them into the stores where they might be bought. Few retail orderers were excited about wasting precious shelf space on El-Fish when that same shelf could be filled with yet more copies of DOOM and Myst - and yes, of SimCity as well.

In the summer of 1992, Jeff Braun traded 30 percent of Maxis's equity and a seat on the company's board for a cash injection of $10 million from Warburg Pincus Ventures. The deal was predicated on the understanding that some of the money would be used for a new SimCity that was more in tune with the expectations of the 1990s, one that could be sold to all of the people who had bought the first one all over again, at the same time that it hopefully reached large numbers of new people.

At the time, Will Wright had been tinkering for months with something that even the dreamy minds around him didn't quite understand, a virtual dollhouse that he was building up from some of the core algorithms of SimAnt. By Wright's own admission, Braun had to "drag" him away from it to do SimCity 2000. He wound up devoting more than a year of his life to the project, but he did so more begrudgingly than passionately, because his trusted business partner was telling him that Maxis needed this game in order to survive and make possible all of the other ones. "I was in management mode," he says. "I had a pretty clear idea of what the design would be, since we were basically just doing a sequel, which is always easier. It was more just making sure the engineering was good and the performance was decent."

Co-designed by Fred Haslam, who had previously worked with Wright and James Lovelock on SimEarth, SimCity 2000 was a new software-development experience for Maxis all the way around. Prior to this point, Maxis had been as much a creative playground as a serious company, happily thumbing its nose at all of the established industry logic about what you needed to do to actually sell games. And, sure enough, they hadn't sold all that terribly many games beyond the sui generis SimCity. Now, Braun told everyone, they needed to make their sequel to that game match a set of bullet points that were demanded by the marketplace, needed not to challenge anyone's expectations unduly while they were doing so, and, most importantly of all, needed to have the finished product ready to go for the Christmas buying season of 1993. They managed all of these feats, but not without some friction and hard feelings. Some employees, finding the dictates to be a betrayal of everything they had thought Maxis stood for, quit rather than abide by them. Those who stayed became familiar before all was said and done with the nature of crunch, another new experience for Maxis people.

In the end, though, SimCity 2000 did exactly what was required of it, both as a computer program and as a commercial product. The most obvious improvement over its predecessor of four and a half years previous was in the visuals. An isometric view replaced the old top-down one, the onscreen color count increased from 16 to 256, and professional artists were... well, suffice to say that they were present this time. The difference really was night and day.

To the improved visuals were added an improved interface and a host of fairly commonsense gameplay extensions: hospitals and schools, new types of heavy industry and power plants, a plumbing system, subways and trains and other new transportation options. (One can detect some influence from Railroad Tycoon in these, feeding back into the game that had done so much to inspire it; game design has always been a dialog.) You could choose an historical starting date for your game and take advantage of new technologies as they became available: subways in 1910, buses in 1920, highways in 1930, etc., all the way up to microwave and fusion power in 2050. (If only...) As your city's mayor, you could issue ordinances and read about your accomplishments and your follies in an in-game newspaper. It was all done well, but at the same time not overdone, being still eminently recognizable as the same SimCity that the old fans knew and loved. In the eyes of a business professional like Jeff Braun, it was the perfect sequel.

He was amply rewarded for his foresight in getting it out before the first SimCity cash cow had dried up completely. SimCity 2000 rocketed up the bestseller charts that Christmas of 1993 in just the way that all of those other recent Maxis games hadn't. Then it kept right on selling for months and then years, another rare perennial in an intensely seasonal industry, a game so popular that it spawned three separate trade-paperback strategy guides from big outside book publishers. Although Maxis had as yet found only one way to actually make consistent money, they were fortunate in that it made them a lot of money. Having given SimCity another lease on life, they could go back to being weird... okay, no, let's call it esoteric.

Did you enjoy this article? If so, please think about pitching in to help me make many more like it. You can pledge any amount you like.

Sources: The books Games That Sell! by Mark H. Walker, Building SimCity: How to Put the World in a Machine by Chaim Gingold, SimEarth: The Official Strategy Guide by Rusel DeMaria, SimCity 2000: Power, Politics, and Plumbing (revised ed.) by Nick Dargahi and Michael Bremer, SimCity 2000 Strategies and Secrets (2nd ed.) by Daniel A. Tauber and Brenda Kienan, The Official SimCity 2000 Planning Commission Handbook by Peter Spear and Johnny L. Wilson, Game Design Theory & Practice (2nd ed.) by Richard Rouse III, and Artificial Life: A Report from the Frontier Where Computers Meet Biology by Steven Levy. Macworld of April 1990; Computer Gaming World of September 1990, January 1991, March 1992, September 1992, April 1993, July 1993, and November 1993; Retro Gamer 115 and 210.

Online sources include the collection of Maxis articles by Phil Salvador at The Obscuratory, Matt Barton's video interview with Jeff Tunnell, and Tristan Donovan's interview with Will Wright for the old Gamasutra site.

Where to Get Them: Of the Maxis games described in this article, only SimCity 2000 is currently available as a digital purchase.

Previous page | More The Digital Antiquarian | Headlines

Original: https://www.filfre.net/2026/07/the-life-and-times-of-maxis-part-1-simeverything/