← all stories

Era 1 · Arcades, first consoles and mainframes

1985–1999sold to Yovelle Renaissance Corp. (January 1996), then IDT…

GEnie

In 1985 a General Electric subsidiary decided to earn extra money on mainframes that sat idle in the evenings, following the example of CompuServe but aiming to undercut its price and beat it on games. Out of this grew RoundTables, the first mass-market multiplayer games (Air Warrior, GemStone), and subscribers around the world. In 1994 GEnie crashed from its peak of 400,000 subscribers, and GE, which had never let the service grow, sold it in 1996 to a small company that resold it to the next one. The service shut down at the end of 1999, officially citing the Y2K problem.

Founders Bill Louden (sole founder, previously a product manager at CompuServe: forums, games, e-mail, CB Simulator)
Domains —
online-servicedial-upidle-capacityhourly-billinggames-retentionprice-war

In brief

GEnie (General Electric Network for Information Exchange) was a side business of General Electric: in October 1985 GE decided to earn extra money on mainframes that sat idle in the evenings, copying a CompuServe trick from six years earlier, but from the start it made multiplayer games not a decoration but a paid lever of growth. Air Warrior from the independent studio Kesmai, a flight simulator carrying an hourly surcharge of $10–11/hour, became one of the first graphical MMOGs in history, 9–10 years before Ultima Online. GemStone, MultiPlayer BattleTech, and Stellar Emperor (a port of the cult CompuServe game MegaWars III) kept players on the line on hourly and per-minute billing until, in the early 1990s, competition among GEnie, AOL, and Prodigy forced even game rates down. By 1994 subscribers collapsed from a peak of 400,000 to 75,000 in half a year, and in 1996 GE sold the service to Yovelle, which sold it to the telecom company IDT. The games themselves outlived GEnie: by the mid-to-late 1990s Air Warrior, Dragon's Gate, GemStone, and CyberStrike had moved to AOL and the web.

How it started (the founder)

GEnie's sole founder was Bill Louden, a product manager at CompuServe who had been responsible there for games, forums, and chat. In 1985 he moved to General Electric Information Services and applied the CompuServe trick of selling the evening hours of idle mainframes, but on GEIS's worldwide network rather than just a domestic one. The games on GEnie were made not by GE itself but by the independent studio Kesmai: Kelton Flinn and John Taylor, classmates at the University of Virginia, who back in 1977–1979 had written a text-based combat game called Air on the UVA mainframe, long before GEnie and with no connection to the university system PLATO. Kesmai sold the same engine to several services at once, GEnie and CompuServe in parallel, which means GEnie got its key game franchise not as an exclusive but as one of several clients of an outside developer.

Year-by-year timeline

Lesser-known but significant facts

  1. Air Warrior beat Ultima Online by ten years, and it grew not out of GEnie but out of a student project on a different mainframe. Kesmai made its first combat game, Air, back in 1977 on the University of Virginia mainframe, 8 years before GEnie and 20 years before Ultima Online (1997).
  2. GEnie's game rates were falling under competitive pressure as early as 1992–1993, earlier than people usually think of the "flat-rate war" of the mid-1990s: CyberStrike came out cheaper than Kesmai's earlier games right away because of competition among GEnie, AOL, and Prodigy.
  3. Stellar Emperor on GEnie was a direct port of the cult CompuServe game MegaWars III. The game was made by Kesmai, and at CompuServe Louden himself had been responsible for it; after moving to GE, he brought the same engine from the same developer to GEnie instead of commissioning a new game estimate.
  4. GemStone outlived GEnie by almost 30 years. Born on GEnie in 1988 with per-minute billing, by the mid-1990s it was generating 1.4 million hours of play a month on AOL and holding 2,000–2,500 players, and it is still alive today as GemStone IV on Simutronics' own service.
  5. Not a single source gives the share of games in GEnie's revenue, even though both Wikipedia and the press of the era describe GEnie as a service "known mostly for its popular on-line, multiplayer games" [not found].

Legend vs. the record

The first growth lever

GEnie did not make games itself. It bought ready-made engines from independent studios (Air Warrior and Stellar Emperor from Kesmai, GemStone from Simutronics, Dragon's Gate as a reworking of someone else's MUD, Aradath) and turned multiplayer games into the stickiest line of the service and the one that earned the most by the hour. The logic is the same as in the price war against CompuServe: GEnie won not on product quality but because it found a genre people were willing to pay a lot for, and for a long time, on top of the base rate. Air Warrior held "hundreds of users" at once in a shared arena at $10–11/hour, while GemStone and Dragon's Gate kept players paying by the minute and by the hour for years, until competitors began undercutting even game rates in the early 1990s. GEnie's regular content moved to a flat rate in 1989, but the games kept charging by the minute for several more years: gamers remained the audience most willing to pay even after the rest of GEnie became "unlimited."

The paired story

GEnie and PLATO are parallel but unconnected stories of the same discovery: multiplayer games keep people on the line better than anything else. On the university system PLATO, from 1973 on, students wrote games for free on infrastructure built for teaching (see the PLATO dossier). On GEnie a decade later the same genre became a paid business: the hourly and per-minute surcharge for Air Warrior and MultiPlayer BattleTech (Kesmai) and GemStone (Simutronics) was a direct lever of growth for the service itself. There is no direct personal link between the two stories. The authors of Air Warrior, Kelton Flinn and John Taylor, came not out of PLATO but out of the separate mainframe culture of the University of Virginia, where back in 1977 they wrote the text game Air, and not a single source confirms that they ever used PLATO. What ties these two dossiers together is not a baton passed between particular people but a shared plot of the era before home internet: the university mainframe (free, on a grant) and the commercial dial-up service (paid, by the minute on the meter) discovered the same economics of engagement independently of each other. GEnie simply turned it into an hourly bill, while PLATO kept it free.

Parallels today (projects from the catalog)

What a builder can take from this in 2026

  1. You do not have to build the game engine yourself: you can buy the best one on the market and get an exclusive growth lever. GEnie did not develop Air Warrior and GemStone itself; it made deals with Kesmai and Simutronics, which sold their games to several services at once.
  2. A service's highest-paying audience is not always its largest. GEnie's games charged by the hour or by the minute even after the rest of the service moved to a flat rate in 1989: gamers were willing to pay more, and for longer, than the average subscriber.
  3. Competitive pressure on price starts earlier than it seems in hindsight. People usually think of the "flat-rate war" as a mid-1990s story (AOL, 1996), but GEnie was already cutting its game prices in 1992–1993 under pressure from AOL and Prodigy.
  4. A product bought from an outside supplier can outlive the platform that sold it. GEnie shut down in 1999, while its games, Air Warrior, GemStone, and Dragon's Gate, were already living on AOL and the web regardless of GEnie's fate.
  5. Parallel stories do not have to be personally connected to teach the same lesson. GEnie and PLATO share no founders and never crossed paths, but both discovered that multiplayer games were the best retention tool on mainframes before the internet.

Discrepancies and what we could not verify

Sources (primary first)

Secondary (context, cross-checking):

Internet History Keeper

Every historical dossier here is free and open, and it will stay that way. If you want the series to keep going (new dossiers, checks against primary sources, the English edition), support it with a subscription for $9.90 a month. It is support, not access, and you can cancel at any time.

Become a keeper — $9.90 a month →

The same thing, about today

The same breakdowns, but of projects launching right now: what the product is, where the first users came from, how they charge. The card is free, the full dossier is $5 (the dossier itself is written in Russian).

Browse the dossier catalog →