In the very early days of Ultima, Richard Garriott made a public promise which would eventually come back to haunt him. Looking for a way to differentiate his CRPG series from its arch-rival, Wizardry, he said that he would never reuse an Ultima engine. Before every new installment of his series, he would tear everything down to its component parts and rebuild it all, bigger and better than ever before. For quite some time, this policy served Garriott very well indeed. When the first Ultima had appeared in 1981, it had lagged well behind the first Wizardry in terms of sales and respect, but by the time Ultima III dropped in 1983 Garriott’s series had snatched a lead which it would never come close to relinquishing. While the first five Wizardry installments remained largely indistinguishable from one another to the casual fan, Ultima made major, obvious leaps with each new release. Yes, games like The Bard’s Tale and Pool of Radiance racked up some very impressive sales of their own as the 1980s wore on, but Ultima… well, Ultima was simply Ultima, the most respected name of all in CRPGs.

And yet by 1990 the promise which had served Richard Garriott so well was starting to become a real problem for his company Origin Systems. To build each new entry in the series from the ground up was one thing when doing so entailed Garriott disappearing alone into a small room containing only his Apple II for six months or a year, then emerging, blurry-eyed and exhausted, with floppy disks in hand. It was quite another thing in the case of a game like 1990’s Ultima VI, the first Ultima to be developed for MS-DOS machines with VGA graphics and hard drives, a project involving four programmers and five artists, plus a bureaucracy of others that included everything from producers to play-testers. Making a new Ultima from the ground up had by this point come to entail much more than just writing a game engine; it required a whole new technical infrastructure of editors and other software tools that let the design team, to paraphrase Origin’s favorite marketing tagline, create their latest world.

But, while development costs thus skyrocketed, sales weren’t increasing to match. Each new entry in the series since Ultima IV had continued to sell a consistent 200,000 to 250,000 copies. These were very good numbers for the genre and the times, but it seemed that Origin had long ago hit a sales ceiling for games of this type. The more practical voices at the company, such as the hard-nosed head of product development Dallas Snell, said that Origin simply had to start following the example of their rivals, who reused their engines many times as a matter of course. If they wished to survive, Origin too had to stop throwing away their technology after only using it once; they had to renege at last on Richard Garriott’s longstanding promise. Others, most notably the original promise-maker himself, were none too happy with the idea.

Origin’s recently arrived producer and designer Warren Spector was as practical as he was creative, and thus could relate to the concerns of both a Dallas Snell and a Richard Garriott. He proposed a compromise. What if a separate team used the last Ultima engine to create some “spin-off” games while Garriott and his team were busy inventing their latest wheel for the next “numbered” game in the series?

It wasn’t actually an unprecedented idea. As far back as Ultima II, in the days before Origin even existed, a rumor had briefly surfaced that Sierra, Garriott’s publisher at the time, might release an expansion disk to connect a few more of the many pointlessly spinning gears in that game’s rather sloppy design. Later, after spending some two years making Ultima IV all by himself, Garriott himself had floated the idea of an Ultima IV Part 2 to squeeze a little more mileage out of the engine, only to abandon it to the excitement of building a new engine of unprecedented sophistication for Ultima V. But now, with the Ultima VI engine, it seemed like an idea whose time had truly come at last.​

The spin-off games would be somewhat smaller in scope than the core Ultimas, and this, combined with the reuse of a game engine and other assets from their big brothers, should allow each of them to be made in something close to six months, as opposed to the two years that were generally required for a traditional Ultima. They would give Origin more product to sell to those 200,000 to 250,000 hardcore fans who bought each new mainline installment; this would certainly please Dallas Snell. And, as long as the marketing message was carefully crafted, they should succeed in doing so without too badly damaging the Ultima brand’s reputation for always surfing the bleeding edge of CRPG design and technology; this would please Richard Garriott.

But most of all it was Warren Spector who had good reason to be pleased with the compromise he had fashioned. The Ultima sub-series that was born of it, dubbed Worlds of Ultima, would run for only two games, but would nevertheless afford him his first chance at Origin to fully exercise his creative muscles; both games would be at bottom his babies, taking place in settings created by him and enacting stories outlined by him. These projects would be, as Spector happily admits today, “B” projects at Origin, playing second fiddle in terms of internal resources and marketing priority alike to the mainline Ultima games and to Wing Commander. Yet, as many a Hollywood director will tell you, smaller budgets and the reduced scrutiny that goes along with them are often anything but a bad thing; they often lend themselves to better, more daring creative work. “I actually liked being a ‘B’ guy,” remembers Spector. “The guys spending tons of money have all the pressure. I was spending so little [that] no one really paid much attention to what I was doing, so I got to try all sorts of crazy things.”

Those crazy things could only have come from this particular Origin employee. Spector was almost, as he liked to put it, the proud holder of a PhD in film studies. Over thirty years old in a company full of twenty-somethings, he came to Origin with a far more varied cultural palette than was the norm there, and worked gently but persistently to separate his peers from their own exclusive diets of epic fantasy and space opera. He had a special love for the adventure fiction of the late nineteenth and early twentieth centuries, and this love came to inform Worlds of Ultima to as great a degree as Lord of the Rings did the mainline Ultima games or Stars Wars did Wing Commander. Spector’s favored inspirations even had the additional advantage of being out of copyright, meaning he could plunder as much as he wanted without worrying about any lawyers coming to call.

[...] Unfortunately, gamers of the early 1990s were rather less blown away. Released in October of 1990, The Savage Empire was greeted with a collective shrug which encompassed nonplussed reviews — Computer Gaming World‘s reviewer bizarrely labeled it a “caricature” of Ultima — and lousy sales. With the release of Martian Dreams in May of 1991, Origin re-branded the series Ultima Worlds of Adventure — not that that was an improvement in anything other than word count — but the results were the same. CRPG fans’ huge preference for epic fantasy was well-established by this point; pulpy tales of adventure and Victorian steampunk just didn’t seem to be on the radar of Origin’s fan base. A pity, especially considering that in terms of genre too these games can be read as harbingers of trends to come. In the realm of tabletop RPGs, “pulp” games similar in spirit to The Savage Empire have become a welcome alternative to fantasy and science fiction since that game’s release. Steampunk, meanwhile, was just getting off the ground as a literary sub-genre of its own at the time that Martian Dreams was published; steampunk’s founding text, the novel The Difference Engine by William Gibson and Bruce Sterling, was published less than a year before the game.

For all that the games were thus ahead of their time in more ways than one, Worlds of Ultima provided a sobering lesson for Origin’s marketers and accountants by becoming the first games they’d ever released with the Ultima name on the box which didn’t become major hits. The name alone, it seemed, wasn’t — or was no longer — enough; the first chink in the series’s armor had been opened up. One could of course argue that these games should never have been released as Ultimas at all, that we should have been spared all the plot contortions around the Avatar and that they should have been allowed simply to stand on their own. Yet it’s hard to believe that such a move would have improved sales any. There just wasn’t really a place in the games industry of the early 1990s for these strange beasts that weren’t quite adventure games and weren’t quite CRPGs as most people thought of them. Players of the two genres had sorted themselves into fairly distinct groups by this point, and Origin dropped Worlds of Ultima smack dab into the void in between them. Nor did the lack of audiovisual flash help; while both games do a nice job of conveying the desired atmosphere with the tools at their disposal, they were hardly audiovisual standouts even in their day. At the Summer Consumer Electronics Show in June of 1991, Martian Dreams shared Origin’s booth with Wing Commander II and early previews of Ultima VII and Strike Commander. It’s hard to imagine it not getting lost in that crowd in the bling-obsessed early 1990s.

So, Origin wrote off their Worlds of Ultima series as a failed experiment. They elected to stop, as Spector puts it, “going to weird places that Warren wants to do games about.” A projected third game, which was to have taken place in Arthurian England, was cancelled early in pre-production. The setting may sound like a more natural one for Ultima fans, but, in light of the way that Arthurian games have disappointed their publishers time and time again, one has to doubt whether the commercial results would have been much better.
​