In brief
Scott Miller, a Texas programmer and columnist, founded Apogee Software in his parents' house in 1987 and came up with a trick that would later be called "the Apogee model": give away the first third of a game for free over BBSes and charge for the rest by mail. The trick did not work right away (only on the second attempt, after a couple of years of trial), but once it did, Miller became less a developer than a publisher: with a small $2,000 advance he funded three programmers from Softdisk who would make Commander Keen and then leave to found id Software, the creators of Wolfenstein 3D and Doom. Apogee renamed itself 3D Realms, spent 12 years stuck in the development of Duke Nukem Forever, fought Take-Two and Gearbox in court, and changed owners several times; in 2021 Miller bought the Apogee name back.
How it started (the founders)
Scott Miller had been writing games since 1975, as a teenager, on a Wang 2200 in Australia, where his father's work took him, a NASA engineer who took part in the Apollo and Gemini programs. By the mid-1980s he was more of an author than an entrepreneur: he wrote a weekly games column for the Dallas Morning News (1982–1985) and a book on arcade strategies, and on the side he worked at a junior college and in data-processing jobs.
His first attempts to make money from his own games already carried the name "Apogee": he distributed text adventures on a pay-if-you-like basis, and the results were disappointing. The turning point came not with the invention of a model but with a single game: Kingdom of Kroz, a roguelike inspired by Rogue, its name Zork spelled backward, written in Turbo Pascal 3.0 in CGA text mode. The game won silver in the CodeQuest contest of Big Blue Disk, a disk magazine from the publisher Softdisk, and came out through Softdisk with two sequels, while the copyright stayed with Miller. Only after Softdisk released the third game of the trilogy did Miller finally apply the idea he had been nursing: the first episode free, the second and third $7.50 each by mail. It worked immediately and noticeably: checks for $100–500 a day, $80,000–100,000 in total for the first year, even though the same texts and games under the same name had previously brought in almost nothing.
Year-by-year timeline
- 1975: Miller starts writing games as a teenager on a Wang 2200 in Australia fact
- 1982–1985: he writes a games column for the Dallas Morning News and publishes a book on arcade strategies; meanwhile he tries shareware under the Apogee name without success fact
- 1987-11-26: Scott Miller founds Apogee Software in his parents' house in Garland, Texas fact
- 1988-06: Kingdom of Kroz takes silver in the CodeQuest contest and comes out through Softdisk (not yet as an episodic shareware game) fact
- 1989-09: after the third game of the Kroz trilogy comes out through Softdisk, Miller applies the episodic model for the first time: episode 1 free, episodes 2–3 at $7.50 each fact
- 1990-09-28: by letter, Miller reaches the future id Software team through Softdisk and agrees to fund an original game fact
- 1990 (Oct.–Dec.), release 1990-12-14: Commander Keen is developed on a $2,000 advance and weekly $100 checks for pizza; the game comes out on December 14, 1990 fact
- 1991-01: Apogee's first royalty check to the id team, $10,500, decides the team's future fact
- 1991-02-01: the team quits Softdisk and formally founds id Software fact
- 1991: George Broussard (the author of Pharaoh's Tomb) becomes a partner and co-owner of Apogee; Duke Nukem comes out fact
- 1992-05-05: Wolfenstein 3D comes out as a shareware episode; id's advance is $100,000, royalties are raised to 50%, and the trademark is bought for $5,000 fact
- 1993: Doom comes out self-published by id Software: the team takes on distribution itself, parting ways with Apogee amicably fact
- 1994: Apogee launches the 3D Realms sub-brand for 3D games on the Build engine fact
- 1996: Duke Nukem 3D comes out at retail with the participation of FormGen (~3.5 million copies cumulatively over all the years); the company fully rebrands as 3D Realms, and Stargunner is the last release under the Apogee label estimate
- 2009-05: talks with Take-Two over Duke Nukem Forever collapse, and 3D Realms lays off almost the entire team; Take-Two files a lawsuit fact
- 2010-09: the litigation ends in a settlement, and development of Duke Nukem Forever is handed to Gearbox fact
- 2013–2015: a new round of lawsuits, this time between 3D Realms and Gearbox, over royalties and an unauthorized project; the franchise stays with Gearbox for good fact
- 2014: 3D Realms is bought by the Danish holding company SDN Invest and moves to Denmark fact
- 2021-08: Embracer/Saber buy 3D Realms; the same month Miller and a partner buy back the Apogee brand and relaunch it as Apogee Entertainment
Lesser-known but significant facts
- Miller's first attempt to make money from games under the Apogee name failed. Before Kroz he gave away text adventures on a voluntary-payment basis, and it brought in almost no money.
- Apogee was the first shareware game company with its own home BBS: by 1995 Software Creations kept 100+ phone lines for 3,500 distribution points and was part of a network of, at its peak, 5,000+ independent BBSes.
- The "Wolfenstein" trademark was bought by the id team itself, not by Apogee: for $5,000 from a private owner in Michigan in April 1992, after the original rights holder, Muse Software, had ceased to exist; Miller only learned that the rights question was settled and immediately approved the project with a $100,000 advance, so the whole franchise rests on someone else's cheap legal find that Apogee simply funded quickly.
- Miller personally turned away a future competitor. Tim Sweeney sent him ZZT for publication, Miller declined it as too similar to Kroz, and Sweeney copied the model himself and founded Epic MegaGames; Miller later admitted that this had been very smart of him.
- Doom left Apogee not because of a conflict but because of growth. id's business manager Jay Wilbur insisted on self-publishing after concluding that Apogee physically could not handle the volume of buyers.
Legend vs. the record
- Legend: Apogee invented shareware. That is how Miller's model is usually presented in retrospectives on shareware games. The record: the pay-if-you-like concept and the term "shareware" itself appeared 4–5 years earlier. In 1982 Andrew Fluegelman distributed PC-Talk under the name "freeware," Jim Knopf (Jim Button) released PC-File as "user-supported software," and in early 1983 Bob Wallace, releasing PC-Write, introduced the term "shareware" for business software, not games. Verdict: Apogee did not invent shareware; Miller applied an existing model to games and added the episodic hook, which had not worked in games before him.
- Legend: id Software published Commander Keen itself. A natural assumption, given that id later became famous precisely for self-publishing Doom. The record: Commander Keen was published entirely by Apogee. It was Miller who paid the advance, ran the mail-order sales of floppy disks, and collected the money; id moved to its own distribution only with Doom in 1993, three years after Keen, at the initiative of its business manager, who saw that Apogee could not cope with the volume. Verdict: exactly the opposite. Without Apogee's publishing model, id would not have had any starting capital for its first game at all.
- Legend: the first-episode-free model accompanied Apogee from its very founding in 1987. That is what most retrospectives imply, merging the founding of the company and the birth of the model into a single date. The record: Kingdom of Kroz came out through Softdisk in June 1988 as an ordinary disk-magazine game, and only after the third game of the trilogy came out in September 1989 did Miller first apply the scheme of episode 1 free, episodes 2–3 paid. That is, almost two years of experiments with other formats passed between Apogee's founding and the first real use of its signature model. Verdict: the company's founding date and the model's birth date are different events, about two years apart.
- Legend: id Software itself approached Apogee with an offer to publish Commander Keen. A plot that seems logical at first glance: a small team looks for a publisher. The record: the initiative came from Miller himself. He wrote fan letters to John Romero about the Gamer's Edge games from a single address, counting on Softdisk not checking unofficial mail; first he offered to publish levels for his own Pyramids of Egypt, was turned down (the rights belonged to Softdisk), and only then did Romero show him a demo in the spirit of Super Mario Bros. 3, from which Keen was born. Verdict: it was not the team looking for a publisher but the publisher recruiting the team.
The first growth lever
The lever was not the game itself but the switch from a failing honest shareware model to an episodic hook. Before Kroz, Miller had already tried giving his games away on the honor system under the Apogee name, and got only a handful of readers willing to pay. When in September 1989 he first applied the new scheme to Kroz (the entire first episode free on BBSes, the second and third $7.50 each by mail), the difference was not a matter of percentages but of orders of magnitude: from almost zero takings to checks for $100–500 every day and $80,000–100,000 for the first year. The free episode spread itself over BBSes at no marketing cost (anyone who had copied the file from a friend could play), and money was asked only of those who had already invested time and wanted more. The same technique was then scaled up to other people's projects: a $2,000 advance on Commander Keen in 1990 turned into $30,000 in the first two weeks of sales and $60,000 a month by June 1991, against Apogee's turnover of about $7,000 a month before Keen, and $100,000 on Wolfenstein 3D in 1992 raised the bar another twentyfold, to $200,000 a month against $10,000 for Keen. The growth lever was the shift from making games and giving them away for free to funding other people's games through the same funnel and taking a share: the shift from developer to publisher.
The paired story
Apogee and id Software are linked, quite literally, by a single money transaction: without Miller's $2,000 advance and his weekly $100 pizza checks in 1990, Romero, both Carmacks, and Hall would not have had the starting capital to leave Softdisk. Commander Keen was published and funded by Apogee entirely on Miller's model, and it was the first royalty check of $10,500 in January 1991 that persuaded the team to formally found id Software on February 1, 1991. From there the pair diverge while staying tied by the same model: id still made Wolfenstein 3D (1992) for Apogee, but on far better terms (the advance grew to $100,000, royalties to 50%), and with Doom (1993) id took distribution fully into its own hands, because Apogee physically could not handle the volume of buyers. The contrast: Apogee invented the funnel of a free piece leading to a paid continuation and remained an intermediary publisher on it for its entire career, while id pushed the same funnel to its limit and walked away from the intermediary as soon as the model began bringing in more money than the publisher's infrastructure could process.
Parallels today (projects from the catalog)
- IRON NEST: Heavy Turret Simulator (
iron-nest-heavy-turret-simulator) is a direct present-day analog of the free first episode: a duo with no publisher and no outside money released the game on Steam and gathered 2,600+ reviews in 2 days with an "Overwhelmingly Positive" rating, only instead of mail-order floppies the funnel is built by free access to a demo and an organic wave of Steam reviews. The difference between the eras: Apogee's funnel stretched over months via BBSes, IRON NEST's over days on a single platform. view this project's dossier → - Evil Trout Inc (
evil-trout-jam-to-steam-remake) is an almost literal repeat of Miller's model: Robin Ward finds a free game with a fan base, made by someone else at a game jam, funds and rewrites it from scratch himself, releases it as a paid game, and shares the revenue with the author. That is exactly the role Apogee played for id: the publisher as a source of capital and distribution for a small team with a ready prototype. The difference: Apogee shared the money through an advance and royalties on a new game, while Evil Trout buys out and rebuilds an existing free game in its entirety. view this project's dossier →
What a builder can take from this in 2026
- Give away exactly enough to hook people, and no more. The difference between Miller's failure (honest honor-system shareware) and the success of Kroz lay not in generosity but in the boundary: the free piece must be self-sufficient entertainment, not a promotional placeholder.
- Don't build everything yourself; become the funnel and the capital for other people's teams. Apogee's real growth came not from Miller's new games but from $2,000 on Commander Keen and $100,000 on Wolfenstein 3D put into other people's hands, against someone else's prototype, for a share of sales.
- When a partner solves a legal problem on its own, pay instantly. id itself found and bought the forgotten "Wolfenstein" trademark for $5,000; Miller did not come up with this find but immediately gave a $100,000 advance once he saw that the road was clear, so the franchise rests on someone else's cheap legal find and a publisher's quick decision.
- When your channel becomes a bottleneck, your best clients will go self-publishing. Doom left Apogee not out of a quarrel but out of plain throughput arithmetic, as soon as sales volume exceeded the capacity of the publisher's BBS infrastructure.
- A working combination of an advance plus a share of the funnel is not a one-off stroke of luck but a reproducible formula. By mid-1993, 7 of the 10 best-selling shareware games belonged to Apogee, not because Miller wrote them all himself but because the formula was repeated with new teams.
Discrepancies and what we could not verify
- The year the episodic model was actually launched. The company's founding date (1987) and the date the episode-1-free scheme was first applied to Kroz (after September 1989) are almost two years apart; most retrospectives merge them into one estimate.
- id's exact royalty rate on Commander Keen: not found; all that is known is that by Wolfenstein 3D (1992) the rate had been raised to 50% estimate.
- Total lifetime sales of Wolfenstein 3D: there is only a figure of more than 250,000 copies by the end of 1995; no later cumulative data could be found estimate.
- The account of why id moved to self-publishing is known only from id's side; no comment from Apogee or Miller himself could be found estimate.
- The Apogee FAQ (rinkworks.com) and legacy.3drealms.com were unavailable (network errors at the time of collection), so they could not be checked directly.
Sources (primary first)
- Scott Miller interview — "20 Years Of Evolution: Scott Miller And 3D Realms," Game Developer
- Commander Keen in Invasion of the Vorticons — Wikipedia (deal details, dates, sums)
- Wolfenstein 3D development history — wolfenstein3d.nl
- "Apogee Software: When the sultans of shareware went retail," Game Developer
- The Digital Antiquarian — "The Shareware Scene, Part 2: The Question of Games"
- 3D Realms — Wikipedia
- Scott Miller (entrepreneur) — Wikipedia)
- Kroz — Wikipedia
Secondary (context, cross-checking):
- Wolfenstein 3D — Wikipedia
- Duke Nukem (video game) — Wikipedia)
- George Broussard — Wikipedia
- Development of Doom — Wikipedia
- Take-Two sues 3D Realms for not finishing Duke Nukem Forever — TechCrunch, 2009
- 3D Realms Responds To Take-Two Lawsuit — Game Developer
- Keen Dreams Development — KeenWiki
- MobyGames — Apogee Software, Ltd. trivia
- What Is Shareware, and Why Was It So Popular in the 1990s? — How-To Geek
- Wikipedia (Jim Knopf, PC-Write): only as a pointer to the history of the term "shareware"