Ask most people what Nintendo is and you get an answer about consoles. That answer covers a minority of the company's existence. Nintendo was founded in Kyoto in 1889, which puts its start closer to the American Civil War than to the first home console. For the better part of a century it sold playing cards, then a scattered set of unrelated products, most of which failed. The video game business did not arrive until the company was nearly ninety years old. Understanding what came before is the only way to understand why the games worked when they finally came.

The original product was hanafuda, a Japanese flower card game played with small painted cards. Fusajiro Yamauchi started making them by hand and built a steady trade. Western style playing cards had been banned in Japan for long stretches because of gambling, and hanafuda occupied a gray zone that gamblers used freely. Nintendo cards ended up in parlors run by organized crime, where a fresh deck was opened for every game and thrown away after. That habit made the business profitable in a way a card company normally is not. The company's first real customers were not children.

Hiroshi Yamauchi took over in 1949 at twenty two years old, after his grandfather had a stroke. He fired the existing managers, including family members, and ran the company with a hard hand for the next five decades. In 1959 he struck a licensing deal with Disney and put familiar characters on card decks aimed at families. The cards reached toy shops for the first time. Sales jumped and the company listed on the stock exchange. Then Yamauchi did the math and saw a ceiling, because a household buys one deck of cards and keeps it for years. He decided the company had to become something else.

What followed was a decade of expensive failure. Nintendo ran a taxi company, and it went well until labor disputes made it unmanageable. It operated short stay hotels, a business Yamauchi later spoke about with visible embarrassment. It sold instant rice, competing against food companies with far more scale. It sold a vacuum cleaner. None of these ventures had anything to do with the others, and nearly all of them lost money. By the late 1960s the company was in trouble and the stock had fallen hard.

The turn came from a maintenance engineer. Gunpei Yokoi worked on the machines that made the cards, and Yamauchi noticed him messing with an extending grabber arm he had built for fun. Yamauchi told him to turn it into a product for the holiday season. The Ultra Hand sold well over a million units and rescued the year. Yokoi was moved into product development and spent the next two decades designing toys and then electronic devices. His stated philosophy was to use mature, cheap technology in a new way rather than chase the newest parts, and that idea shaped everything the company later shipped.

The electronics arrived gradually. Nintendo built light gun games and put a laser clay shooting system into bowling alleys during the 1970s. It licensed console technology and released the Color TV-Game series in Japan. Yokoi designed the Game and Watch in 1980, a handheld built around calculator screens, which sold in enormous numbers and introduced the directional pad. In 1981 a young artist named Shigeru Miyamoto was handed a failing arcade cabinet and told to salvage it, and he turned it into Donkey Kong. The Famicom followed in 1983 in Japan. Every one of those steps used the toy company's instincts rather than a computer company's.

The American launch is the clearest proof of the point. The video game business in the United States collapsed in 1983, and retailers wanted nothing to do with the category. Nintendo did not sell the machine as a console. It sold it as a toy, with a robot accessory and a light gun in the box, and placed it in the toy section of stores. The company also controlled which games could be published on the system after watching a flood of poor titles help sink the previous generation. Both decisions came out of a company that had spent decades learning how toys are sold. A pure electronics firm would likely have made neither.

The pattern holds if you keep watching. The handhelds, the odd controllers, and the motion hardware all follow Yokoi's rule of using proven parts in unexpected ways rather than winning a specifications race. The company has lost that fight when it entered it and won repeatedly when it refused. That approach did not come from software. It came from a hundred years of selling cards, toys, and a few things that failed badly. The failures were not a detour before the real business started. They were the training.