Eric Roffman had a Ph.D in theoretical physics, and he wanted to make games. After working on an interactive LaserDisc poker title, Roffman looked for a way to combine his interests in science, games, computer graphics, and film. So in 1990, he started Personal Media Interactive, a company that would develop “projects that looked at the future, or combined gaming with interesting ideas, including a number of games designed to be both intelligent and entertainment.”1
They would make “intellitainment,” as they awkwardly dubbed it, multimedia games for adults.
Intellitainment began and ended with their only title, Millennium Auction. As the title suggests, it’s an auction game, a genre Millennium Auction basically made up. Auctions have an unpredictable, suspenseful rhythm, so Roffman planned a game around them as a way to experiment with a variable, randomized narrative in a speculative setting.1
The game certainly delivers way more intrigue than I expected from a virtual auction with nothing at stake. Chance plays a major part in that, though, and it raises questions about the role of randomness as a narrative tool.
Millennium Auction‘s future is the year 2011 – as envisioned in the early 90s, exaggerating those years’ trends in international politics and communications. The newly formed World Body government has total control of the planet. Everything filters through InfoLink, a virtual reality network that regulates identity and finance. Today, the World Body Auction House is hosting a once-in-a-lifetime sale of the world’s most treasured artwork and historical objects. For once, there’s no mystery to solve. The action never leaves the auction house. You and the other members of high society are here to bid.
So why would you participate in an imaginary auction? You can’t price a non-existent item like the “Astrobot prototype” or real, unsellable artifacts like the bullet that killed JFK, and of course, you’re not motivated by really owning them.
Millennium Auction isn’t about the individual items, despite their elaborate backstories. The game wants you to sell them for a profit, and in an auction house full of cutthroat buyers, that’s not easy. Everyone knows the market value of the twelve items on the auction block, so they rarely go for below asking price.
Instead of looking for a steal, you want to bid strategically. A few collectors will pay premium for a set of three or four pieces, so getting a group with a high total markup is a safe plan. Assuming your competitors don’t get the same idea (they could buy a portion of the same lot you’re pursuing), your purchases can still gain or lose value. Prices fluctuate naturally over time, but they also depend on external circumstances. Throughout the auction house – especially in the office of Zeke the friendly, resourceful janitor – televisions, radios, and newspapers clue you into randomly chosen world events and expert appraisals that might affect your portfolio.
Consider the case of the final ball pitched at the last Major League Baseball game. (The MLB shut down in this timeline.) A newspaper reference to historian confirmed that this ball was the real thing, so I figured it would rise in value and a slightly higher bid would be okay. Shortly after the auction, though, a Japanese baseball league announced their first expansion into America, immediately ruining the novelty of the last American baseball. I didn’t know something like that could happen! I had to eat the cost, like when Nicolas Cage had to forfeit that stolen dinosaur skull. If only I’d sold it sooner. Think of your purchases like investments, and expect some to be a little more volatile.
As the story behind that baseball suggests, these news briefs and the item descriptions hint at a bigger story simmering underneath Millennium Auction. True to PMI’s mission, the game sneaks in a portrait of a digital, globalized future, one that paved over its old institutions after a nationalist trade war nearly engulfed the planet.
Apart from getting the date wrong, this futurism isn’t too unreasonable. It’s also extremely silly. InfoLink’s presentation falls for the cheesiest 90s virtual reality tropes – blobby graphics, a convoluted drag-and-drop interface with “portals,” tunnels of light to switch between screens, and a belabored, breathy tutorial voiceover. Yet these are perfectly good reasons to love Millennium Auction, too, and not merely for kitsch. The Cold War ended during Millennium Auction’s development, and the game tries its best to anticipate the fears and possibilities of the coming connected world. It encapsulates the dorkiness of millennium-era yearning, taking earnest interest in where we’re all going and expressing it with goofy wireframe models.
That anxiety stays under the surface, for the most part. Millennium Auction keeps its focus on the auction – and on the dynamic stories within there.
The other bidders have psychological profiles that affect their strategy. InfoLink contains dossiers on each person’s background, thrown together from news clippings. The suggestions about their personalities vary in subtlety: business mogul Rasheed Ahmad seems preoccupied with paternalism and legacy, while artist Tory Swift directly monologues to a reporter about her concern for the environment. The best profiles don’t immediately reveal how their subject might behave. Young business heir Randall Prescott Smith is apparently impatient and isolated, and he runs a disreputable art collecting organization that may have ill intentions. It’s intriguing to picture how that might figure into his auction style.
Honestly, though, there’s no way to tell. You could eavesdrop on every conversation to inform your choices, but when you’re actually bidding, you can’t filter personality from random AI behavior.
The auctions crackle with drama. Every long pause between bids builds pressure. The audience gasps when the price goes too high, and the top bidders seem to be trying to psyche each other out. How much does character determine their actions? Did that tech genius linger on their last bid because they like to keep a low profile, or did the game arbitrarily wait for a second? The same concerns apply when considering what factors might sway item values: does the player risk putting more thought into them than the game does?
Are we projecting our own expectations onto random occurrences, like those pictures of signposts and faucets that look like faces? If we still enjoy the story, maybe the intention doesn’t matter. The signal and the noise don’t have to be separate. People are erratic, and so are world events and sudden bidding wars.
Semi-randomized storytelling like this gets old after a few iterations when you start to notice the Mad Libs-like patterns, but Millennium Auction‘s novelty only carries enough interest for a single playthrough – short enough not to spoil the effect. And during that one time through, you can get caught up in the psychology of aggressive bidding without feeling like you’re overdoing it.
And you should overthink it! Millennium Auction is best enjoyed when taken too seriously. You get a better mental workout if you totally submerge yourself in the setting, goofiness included. It inspires fun strategies: I would purposely drag out an auction I didn’t intend to win just to tempt my opponents into spending too much.
(The game also allows additional human players in place of the AI characters. That could lead to more cutthroat strategy, but it negates the investigatory, character-driven role-playing parts of the game. It also doesn’t have a good way for multiple people to use the InfoLink without taking turns and waiting.)
Millennium Auction doesn’t go as deep into its alternate future premise as it could. Instead, the game tells a story that blurs the line between world-building and random statistics thanks to the power of suggestion. Like tabletop or live-action role playing, but with asset management as the main source of tension. The approach works well for Millennium Auction‘s smaller, non-violent situation, and you could imagine expanding the idea, maybe into a political thriller that explores the dynamics of a speculative world more fully.
Unsurprisingly, though, that was the end of intellitainment. Millennium Auction was not successful, and the studio disbanded shortly after the game’s release. The story of the game’s troubled production sheds light on the struggles that new developers could encounter in the 90s multimedia landscape, plus some pitfalls unique to this company. It was a miracle that they finished the game at all.
PMI was buoyed by a $2 million investment from Armenian construction magnate Berj Kalayjian, an unusually large budget for a multimedia CD-ROM.2 Eric Roffman’s game concept attracted interest at trade events from big names including Electronic Arts and Bill Gates,1 but after years of production, PMI had essentially nothing to show.3 With his theoretical physics background, he had seemingly limited experience managing a project at this scale. Kalayjian grew restless. As Roffman recalls, “[Kalayjian] felt the project was taking too long and we had a shake-up.”1
In 1992, the company approached Michelle Blank, an emerging-tech-focused strategist, to serve as vice president of business development and work on marketing.2,3 According to Blank, when she interviewed for the position in April, Millennium Auction was barely in pre-production – and they wanted to launch in October.3
I came in interviewing probably about four months before what was being represented as the planned launch date. […] And so I said, “Oh that’s awesome. You must be in beta.” And Eric goes, “Well, no.” I go, “Oh, okay, you’re not in beta and you think you’re gonna launch in four months from now? Well then, you must have like an alpha version.” And he goes, “Well… not really.” I said “Oh, okay, so you can’t show me a beta version, and you can’t show me an alpha version. […] So do you have a detailed, you know, functional requirements – the product specs, you know, for the game […] to be driving the development?” And… they didn’t have that. […] I had first been approached to do business development, and then I quickly realized that there was no product, at this point, to even begin in any kind of business development because we were still in the early stages of product concept development. […] There were a couple of ideas, and they had hired great, you know, 3D animators and artists and all this kind of stuff, but they weren’t going anywhere because there was no well-defined product.3
Blank further characterized the game up to this point as “at best, a one-page description of this game set in the future […] where people are competing with a unified money system for historical art pieces.”3
Still, Blank was interested. Though she initially declined to join the company,4 Blank recognized the growing market for socially oriented multimedia games “targeted to a different population than your typical hardcore gamer.” The Millennium Auction concept, with its focus on art and social interaction, could capitalize on this.3 She came aboard the next year.
As part of the reorganization, Blank took on a larger creative and project management role in the company.3 She approached it, appropriately, like a business developer. Blank led an effort to rework Roffman’s original concept, which stuck closer to what she labeled “purely esoteric classical art.” In an effort to widen the game’s marketable appeal, the team emphasized the title’s “social multiplayer gaming” aspects, as well as introducing lighter pop culture-related auction items, like Bill Clinton’s saxophone.3 Roffman says humor was always meant to be part of the game,1 but according to Blank, he bristled at the new direction.
Roffman was dissatisfied with the changes to his company and the project. He quit shortly later.3 At some point in this process, PMI was rebranded Eidolon. (Programmer Jim Berrettini suggested the new name to evoke the illusory nature of their game’s characters.)4 Despite only receiving special thanks in the final product, Blank became, as Robert McNatt put it in Crain’s New York Business, “the de facto chief executive.”2 She effectively took control of the company to get development back on track. With the new leadership, Roffman’s futurist vision seems to have taken a back seat to actually getting the game done.
“I don’t want you to think that I came in and all of the sudden we had this like, fully fleshed-out functional specifications. No!” Blank said. “We certainly had a framework that didn’t exist before upon which we could now put flesh on.”3
A rowdy, freeform year of development followed. The expanded (but still small) Eidolon team huddled together at a private house in the Riverdale neighborhood of the Bronx, “chain smok[ing] and trad[ing] bad jokes.”2 Because of their intimacy, the team could constantly bounce ideas off each other and come up with characters and game ideas organically. The game was developed more-or-less on the fly from there.3
(As a testament to their collaborative environment, one person, David Rosenbloom, received shared credit as writer, programming lead, game designer, creative director, project manager, and original music composer.)
On top of all this, Eidolon had always planned to self-publish,1 a steep challenge for such an peculiar game meant for an atypical audience. The company pursued an aggressive retail marketing strategy based on eye-catching, over-the-top box art and highly visible placement on store shelves; Blank recalls seeing copies in a store window on Fifth Avenue in New York.3 To take some of the work off Eidolon’s shoulders,2 Blank courted companies like Broderbund before partnering with Electronic Arts to handle the game’s distribution.3
Somehow, they got the game out by fall 1994. In an interview with Newsday, Eidolon marketing director Lonnie Stein tossed around a Millennium Heist concept for a sequel,4 but it probably didn’t go further than throwing around an idea. “Everybody… it was a project of love,” Blank said, “and it was sad when people realized, you know, there was not funding for the company to move forward.”3
Despite the growing audience for games during the multimedia era, Blank thinks intellitainment may have debuted at the wrong time. “If you actually look at where the point of the market development was, you were really still getting most people who are hardcore gamers or into technology, right?” she said, contrasting that with the popularity of social board games today. “So [the concept] was misaligned with who was buying video games at this time. […] I totally think that there is a market […] beyond the more violent, hardcore video games that I think are really not elevating anyone’s cognitive capabilities or social skills. But we’re talking more than twenty years ago, right? I mean, it was, it was just… there weren’t enough computers with CD-ROM drives sitting in families.”3
Eidolon is the quintessential struggling 90s CD-ROM developer story, and the messy production of Millennium Auction serves as a good lesson about running loose with big ideas. Roffman had the freedom to explore new modes of storytelling and maybe invented a new genre, but the pursuit of those ideas can get sidetracked when they run up against the realities of managing and marketing a large project.
Clicking the Credits menu button while holding and quickly releasing the Control key will show a picture of the Eidolon development team in spacesuits while Nuria, your virtual guide in Millennium Auction, sings an extremely creepy version of “Daisy Bell” a la HAL 9000.
1. Roffman, E. (2017, April 28). Email interview.
2. McNatt, R. (1994, February 14). Multimedia phantom courts adults. Crain’s New York Business, 10(7), 3.
3. Blank, M. (2017, May 11). Phone interview.
4. Fetherston, D. (1994, April 4). A game for the new millennium creators hope their multimedia software will be a hit with adults. Newsday. C03.