1989: Was this the Year the Beat-‘Em-Up Grew Up?
1989 might be the single best year the beat-’em-up genre ever had. Three studios, working entirely independently of each other, each shipped a genre-defining answer within about ten months — Sega in January, Capcom in November, Konami splitting the difference in between. None of them were comparing notes. All three are still talked about today.
Double Dragon, which had landed in arcades two years earlier, sits somewhere behind at least two of these games specifically. Sega’s Makoto Uchida has said outright that he set out to build “a Double Dragon that was not a Double Dragon,” and Capcom’s Yoshiki Okamoto has cited the arcade version of Double Dragon II directly as his benchmark for Final Fight. Konami’s Teenage Mutant Ninja Turtles is a different case — less a direct response to Double Dragon than a beneficiary of the wider co-op brawler wave it helped kick off. Three studios, then, working from at least two distinct starting points, all landing in the same twelve months.
Sega Went First, and Went Fantasy
Golden Axe hit Japanese arcades on 27 January 1989, which makes it not just the strangest of the three answers but the earliest. Lead designer Makoto Uchida had joined Sega’s AM1 division two years earlier as part of Team Shinobi, and Golden Axe was his second project as lead, coming directly off 1988’s Altered Beast. Both games ran on the same Sega System 16B hardware, which came with a hard technical ceiling: the system could only track and display six characters on screen at once, player and enemy combined. Every fight in Golden Axe had to be designed around that limit rather than against it.
The brief from Sega was straightforward: build something in Double Dragon’s lane. Uchida’s own account of how he approached it is the more interesting story. Rather than compete directly with Technos, who’d already spent years building the Kunio-kun street-fighting lineage that Double Dragon grew out of, he deliberately built the opposite of a street brawler — a Conan the Barbarian-inspired fantasy world, three warriors armed with swords instead of fists, and a shared magic meter that let players trade a limited resource for a screen-clearing spell. The team was tiny, reportedly five people, and the whole thing took about a year. It shouldn’t have worked as a beat-’em-up. It worked immediately, and the magic system in particular was a genuine structural innovation — a resource to manage rather than a power-up to grab, which turned co-op play into something closer to a shared tactical decision than button-mashing.
Capcom Went Weighty, and Almost Went by a Different Name Entirely
Final Fight didn’t arrive until 25 November 1989, ten months after Golden Axe, and it took the opposite approach entirely: stay on the streets, but make every hit feel like it costs something. Designer Akira Nishitani built the combat around the arcade version of Double Dragon II as his benchmark to beat, while producer Yoshiki Okamoto pushed the team to plan the whole project like a film rather than a game — reportedly making the staff watch a run of movies for tone reference, with the 1984 film Streets of Fire cited repeatedly as the visual and atmospheric touchstone. Cody’s design in particular draws directly from that film’s lead.
The game very nearly launched under a different name and inside a different franchise altogether. Capcom’s sales division had originally asked for a straightforward Street Fighter sequel, so early trade-show demos were promoted as exactly that, under the working title Street Fighter ’89 — Ryu and Ken were even considered as the playable leads at one stage. Arcade operators weren’t convinced; feedback came back that the game played nothing like Street Fighter, and Capcom retitled it before release. Even the antagonists got a bit of franchise recycling: the Mad Gear gang borrows its name from an unrelated 1987 Capcom racing game. What emerged — Mike Haggar, Cody, and Guy fighting through Metro City to rescue Haggar’s daughter from the Mad Gear gang — went on to sell over three million units worldwide across the series and effectively became the genre’s most influential template, in Japan and out of it.
Konami Went Cooperative, and Made an Absurd Amount of Money Doing It
Teenage Mutant Ninja Turtles, also 1989, took the least obvious path of the three: a straight licensed tie-in, built around a cartoon that was only two years old at the time. First shown at the AMOA trade show in September, it officially reached arcades the following month — the last of the three to launch, and by some distance the most commercially explosive. Konami built the cabinet around genuine four-player simultaneous co-op using their own “two-and-a-half dimension” scrolling engine, a rarity for the era and expensive to manufacture, then refused to cut corners on the game itself underneath the license. The bet paid off spectacularly: it became the highest-grossing dedicated arcade cabinet in the United States in 1990, and remains, by some measures, Konami’s single highest-grossing arcade release of all time. Turtle puns and all.
The Common Thread
None of these three studios were solving the same problem the same way, and none of them were even aware of what the others were building. And yet each one independently concluded that Double Dragon’s formula was a starting point rather than a finish line — fantasy versus street, weight versus speed, four players versus two, all delivered within the same calendar year. Golden Axe’s own team could see the shape of what was coming even as they shipped it; later retrospectives on the game note how directly its ideas fed forward into Streets of Rage two years later, right down to swapping Golden Axe’s magic spell for a screen-clearing police airstrike.
Streets of Rage would go on to synthesise all three lessons into the genre’s home console peak. But 1989 is where the actual experimentation happened — in three different directions, from three studios who never compared notes, arriving within ten months of each other at three completely different answers to the same brief.
Worth remembering next time someone claims a genre only has one right answer. In 1989, three studios proved it had at least four.
