Kotaku is proud to present an excerpt from Nintendo Magic: Winning the Videogame Wars, by Japanese reporter Osamu Inoue. In this section, an insider’s view of how the DS and Wii came to be.
“What’s keeping people from touching game machines? What’s making them run away?”
Their discussions started there. Recent game systems had button-encrusted controllers that were too complicated. Software that used complicated technology was becoming more common, and the gap between experienced players and beginners was growing wider all the time. It scared people off-or worse, made them actively dislike videogames. Their discussions grew to encompass the themes of games.
While adventure games like Mario were fine, was that really enough? What if games included themes that related to the lives of ordinary people? Would people who considered gaming a waste of time embrace it then? As they converged on the company’s new direction, the idea came to Miyamoto: dedicate one of the displays to touch control.
One screen would be used for intuitive, approachable controls, and the other would be the main display. The system would be easy for anyone to control, and it would allow the development of new kinds of games.
Miyamoto immediately set to work to come up with prototype hardware that would resemble, in outward form, the “Game & Watch” series of electronic games that had set Nintendo down its electronics path in the early 1980s, yet represent a radically different concept.
In the summer of 2003 Miyamoto showed Iwata a new PDA. It was called a PocketPC and had started to gain popularity among businessmen with its stylus-based input system.
The small device allowed users to manage their contacts, email and calendar using either an on-screen keyboard or handwriting recognition. Miyamoto had stuck a strip of tape across the middle of the screen, dividing it into two halves-a kind of makeshift dual screen.
Yet there on the screen was Mario. He dropped down from the top of the display; touching him with the pen made him jump into the air. It was an uncomplicated piece of software.
“Hey, this is great!”
Iwata and Miyamoto beamed at the simple controls for the simple game. It was the moment that crystallised the basic design for the Nintendo DS.
It was only a few months earlier that Sony Computer Entertainment-the company that dominated the home console marketplace-had announced that they were getting into portable game systems as well.
“We’d like to welcome the newest member of the PlayStation family: The PSP. It’s a portable PlayStation.”
Held annually in the United States, E3-the Electronic Entertainment Expo-is the largest videogame industry event in the world. In May 2003, SCE president Ken Kutaragi had spoken those words, and taken a tiny optical disc called a UMD, or Universal Media Disc, from his pocket. The PSP would be the first portable game system to use an optical drive. It would carry two powerful CPUs, with processing power roughly equivalent to the PS2.
Kutaragi promised to “show the system itself at next year’s E3, and have it on sale worldwide by the end of next year.” Nintendo’s stock took a major blow at the introduction of this so-called “21st Century Walkman” that could not only play games but also music and movies. Investors couldn’t help but have visions of the past when Sony stole huge chunks of Nintendo’s home console business. The pressure was on.
But Iwata was unmoved; while Sony chased processing power, Nintendo would head in another direction entirely.
In August 2003, once the basic design for the DS was decided, Iwata launched a small counterattack at a corporate strategy briefing attended by journalists and analysts. “I can’t reveal any details,” he said, “but Nintendo is heading for growth in a new direction. We’re working on something completely new that anyone might pick up.”
Early the next year, in January 2004, Nintendo revealed that they would be launching a twin-screened portable system at the end of that year-the same time as the PSP. In May they demonstrated the hardware before an excited audience at E3. It was the first time the public saw the fruits of Nintendo’s efforts to increase the gaming population, and the media reported on the new system with breathless anticipation.
“This direction is the right one.” Iwata’s increasing confidence was being reflected in the development of another strategic product.
Mum Has To Like It: The Development of the Wii
In the first half of 2003, when Iwata and Miyamoto were dreaming up designs for the Nintendo’s new dual-screened portable system, they were also deep in discussions with the head of integrated research, Takeda, about a new home game system.
Takeda was Nintendo’s hardware pro, responsible for the development of every home game system from the NES to the GameCube – but this time, Iwata had given him new orders.
“Takeda, listen – this time we can’t just focus on making a more powerful system.”
“So you’re telling me to go off the tech roadmap?” Takeda asked.
“That’s right. Let’s get off it.”
It went against every piece of received wisdom in the videogame industry.
Computers optimised for playing videogames – and especially home videogame systems – have significant overlap with advancements in general-purpose computing. From the chips that served as their CPUs and graphics processors, to optical media like DVDs, to wireless networking-the list goes on and on. The desperate competition for supremacy in the home console domain led manufacturers to the bleeding edge of technological advancement.
It was common sense, then, that console makers would plan their hardware development along the expected path of technological advancement – the “technology roadmap.”
Nintendo was no exception. They had embodied the state of the art with the 64, and even though its successor, the GameCube, had not been designed solely for raw performance, it was far more powerful than the 64. The GameCube had straightforwardly incorporated advancements in technology made since the 64. The Wii’s debut was fully five years after the GameCube’s. It was only reasonable that it would include five years’ worth of technological advancements.
But now Iwata was telling Takeda to stop planning designs based solely on technological progress. What he was suggesting now was an entirely new approach; it was something that had never been tried before.
Instead of designing a console around fundamental performance, the new system would expressly seek out technology that would endear itself to families -a “Mum has to like it” approach to development.
“Videogames drive mum crazy – she has to pick up the controllers once the kids are done playing, they’ve already got multiple consoles plugged into the TV and she doesn’t want another one. They’re a nuisance, as far as she’s concerned. We realised that if we wanted to grow the gaming population, we had to build a console that no one in the family hated.”
Mom didn’t care about processing speed, so using technology as the starting point was meaningless. So what did mum care about? They would determine the Wii’s specifications to appeal to a mom’s sensibility.
Compared with its rivals, the PS3 and Xbox 360, the Wii is significantly smaller. Viewed end-on, the area of the main unit’s front panel is less than half of either the PS3’s or Xbox’s, and it takes up only one-sixth the volume of the PS3, and one-fifth that of the Xbox.
This was the first thing Iwata had fixated on.
It started during a meeting Iwata had with Miyamoto and Takeda.
They’d gotten as far as “don’t chase performance” and “Mum has to like it” when Iwata said “Hang on just a second,” and suddenly left his office. He returned with several common DVD clamshell cases. He stacked two, then three of them together.”
Our next-generation machine shouldn’t be any bigger than two or three DVD boxes.
“Every manufacturer’s consoles had gotten bigger with each generation, as if to display their increased capabilities. Yet Iwata was now saying their next home console would be smaller even than the original NES.
From the standpoint of appealing to mum, though, Iwata thought it made good sense.
A smaller console would feel less like extra clutter. In order to avoid overheating, the compact design would require the hardware to use less electricity, which would make it more economical to run. If they could control the heat problem, that meant the console wouldn’t need a noisy cooling fan.
Raw computing power was completely antithetical to the triple virtues of compact size, low power consumption, and quiet operation. Nintendo had sworn to increase the gaming population, and it was clear which direction they would choose.
However, hardware engineers who dislike high performance are rare. It would be quite difficult not to care what their rivals were up to. Sony was in the midst of developing their next-generation system, the PS3, whose main selling point would be its supercomputer-like processing speed. Its custom-made Cell Processor utilised the latest in semiconductor miniaturisation technology to fit nine CPUs on a single chip.
Microsoft, too, was working with IBM to increase the capabilities of the PowerPC chips that would power its Xbox 360 system, which would be many times as powerful as its predecessor. The first manufacturer to use the PowerPC CPU in a game console had been none other than Nintendo.
Their rivals were pressing onward. Iwata was convinced that there was no future in pursuing pure computing power, but the group of engineers that Takeda led were bound to be worried.
When Takeda was asked about doubts and concerns he may have had, he answered, “I can’t say we didn’t have any. We’re engineers, after all. We’d think about things like wouldn’t there still be a certain market for a faster machine, or whether it was okay to be taking a gamble like this. When an order comes down to do something different, not having a specific solution, no counterproposal, makes everyone anxious of course.”
But early in the development stages, Takeda realised that “this was just directing our engineering efforts in another direction. It wasn’t as though we were abandoning engineering altogether.”
It was like designing a laptop computer. To fit the necessary components into such a small enclosure, designs became more complicated and more expensive. To hit the three goals of small size, low power consumption and quiet operation without sacrificing too much processing capability and keeping costs down would take serious engineering acumen.”
It might seem like game console design is a unique field, but if you look at larger trends it has a lot in common with computer or cell phone engineering. You’re on a prescribed playing field. What’s the point of doing the same thing as everybody else?”
Takeda put the challenge to his team of engineers, and they set about designing the CPU, which would be the most power-hungry component in the system.
In order to control costs and avoid chasing technology for its own sake, the next-generation system would be based on an updated, improved version of the GameCube hardware. So long as the basic structure remained the same, it would be compatible with GameCube software. After all, one of the main selling points of the PS2 was that it could play games written for the original PlayStation.
The first goal for the engineering team was to find out by how much they could lower power consumption while preserving GameCube-like levels of performance.
Semiconductors allow circuit pathways to be made narrower so that more of them can be fit on a given chip, thereby increasing processing power. With more circuits in the same area, the amount of current flowing through the chip rises, producing more heat-but Nintendo used this law against itself.
Instead of fitting more, narrower circuit pathways on a chip, they scaled the same chip down to shrink power consumption and heat. At the time, 180 nanometre (that is to say, 180 billionths of a meter) pathways were standard; the GameCube was no exception. But the major semiconductor manufacturers like Intel and IBM were already developing 90 nanometre chips, which would be ready for volume production right about the time the next generation of game consoles would be going on sale.
What if Nintendo’s new system used a 90-nanometre version of the GameCube CPU? Based on their research, the development team found that one third to one fourth the power use of the old chip was a feasible goal. While their rivals were using the new technology to ask “How much can we boost performance?” Nintendo was asking “How much can we shrink power consumption?”
The efforts to lower power use didn’t stop with the CPU-they extended into every aspect of the unit, from the graphics processor to the wireless networking, to overall physical structure. As a result, the team’s confidence that they would be able to fulfil Iwata’s reckless goal of a system “no bigger than two or three DVD boxes” was deepening.
As he watched the ongoing efforts, Iwata’s confidence, too, grew – confidence that they would be able to deliver, in the wake of the DS, a new kind of home console too.
In June 2004-only a month after the DS debuted at E3 and shocked the world with its novel concepts – Iwata announced at a corporate strategy briefing in Japan that Nintendo was in the process of developing a new home console.
“What’s important isn’t next-generation technology, but next-generation game experiences. Computing power isn’t that big of a deal. At this year’s E3, we introduced the DS, with a design unlike anything that’s come before it. With our next home console, we’ll be doing the same thing, and you’ll see it at next year’s E3. Inside the company, it’s been code-named ‘Revolution.’
“Iwata had already shown Nintendo’s first concrete step toward bringing people back to gaming – and now he was calling the next step a “revolution”. The name was meant to evoke the surprise that people felt the first time they played the original NES, with intuitive gameplay à la DS that anyone could engage.
The idea to bring innovative gameplay to users already existed, but it would be another six months before that concept was turned to reality.
The Humble Remote Control
For the Revolution to live up to its codename and revolutionise the game industry by expanding the gaming population, it would need something special.
The DS, with its dual screens and stylus controls, lowered the barrier to videogames. The Revolution needed to do the same thing. The controller would be at the core of its interface, and it could not be less than perfect.
Certain aspects of the controller were decided early in its development: It had to be wireless, and it could not be intimidating.
When Iwata was talking to Miyamoto and company about the circumstances that were leading to gaming’s decline, the first thing that came to the president’s mind was a TV remote control-a piece of technology the entire family used.
People who didn’t play videogames never touched game controllers. The wires that snaked out from the console were nothing but a nuisance to them, and if controllers dared to be left about, they were put away. But the TV remote never bothered anybody. As Iwata considered the difference between the two, he realised the new controller would have to be wireless.
Then he wondered if people found controllers intimidating because of the way they looked.
Game controllers were constantly getting more complicated; in addition to the standard direction pad and buttons, they were now encrusted with all manner of analogue control sticks and triggers, placed seemingly everywhere. Was that alone enough, perhaps, to drive people away from a videogame?
The new controller had to be simple and approachable. Iwata also felt that, like the DS’s touch screen, it needed to facilitate direct, intuitive controls – and Miyamoto and Takeda agreed.
It was an easy goal to set, but a hard one to reach.
Development on the new controller was in full swing by the middle of 2004. With Iwata focusing on the DS, Miyamoto acted in his place on the project. Takeda’s engineering team searched for sensors that would enable intuitive game control, which Miyamoto’s people used to implement the actual controller.
At first, they started just as Iwata had suggested – by aiming for a simple, TV remote-like controller. But as they did the work of testing the controller prototypes that included the new sensors, the form factors began to take unexpected directions.
“No one liked that one,” recalls Miyamoto with a rueful grin, of a large, disc-shaped example. It had a large star-shaped button in the centre, surrounded by three smaller buttons, and used internal accelerometers that let the player control by tilting it forward, backwards, left or right. The prototype was orange, and its bizarre appearance earned it the nickname “cheddar cheese” from the development team.
It was admittedly simple and easy to understand, but it was also far too garish. Miyamoto’s team met with around 40 game developers within Nintendo bimonthly to hear their opinions. This iteration of the controller met with opposition-they complained that it was totally unsuited to traditional games like Mario and Zelda.
Meanwhile, the search continued for something equivalent to the DS that might involve the idea of touching.
The development team was fixated on a pointing device that would allow direct, obvious interaction with the screen-something like a computer’s mouse. “We might still use it in the future so I can’t say much,” said Miyamoto, but it seems that touch pads not unlike the type used on many laptop computers were seriously considered.
The team went through dozens of prototypes-some were truly fantastic, with pointing devices on the front and traditional two-handed button controls on the back. But it wasn’t direct enough-touching the controller here affected the screen there. It wasn’t intuitive.
But then Takeda arrived with a sensor that saved the day.
“Here, give this a try. It’s really responsive.”
It was near the end of 2004, roughly six months since controller development had begun in earnest, when Takeda demonstrated a new prototype.
It used camera components-specifically, a CMOS imaging sensor of the kind widely used in video cameras. By using the sensor to track the position of two light sources placed near the TV, the controller could calculate the direction in which it was pointing.
A normal video camera captures between 30 and 60 images per second. That was fast enough to fool the human eye, but for the controller to keep up with fast-moving gameplay, standard imaging speeds wouldn’t be fast enough. But since the system Takeda had devised had only to track two discrete points of light, it was able to sample images at over 200 frames per second and send the resulting positional information to the console. A dark horse had appeared in the race to complete the controller.
“It’s a cinch tracking the motion.” “Since you’re pointing it at the TV, it matches up with where you’re looking-it just feels right.” “If we make this the core of the unit along with standard controls, you’ll be able to enjoy traditional games, too.”
Praise rolled in from the development team-and it solved the form factor problem as well.
Since the controller had to be pointed at the TV, it was natural to hold it vertically, like a TV remote. Fitting it with accelerometers allowed it to sense movements like tilting and shaking as well. Careful arrangement of the buttons on the unit meant that it could be held sideways and used as a more traditional game controller. In addition to solving all these problems, the new design had settled into something very close to Iwata’s original ideal of a TV remote.
For the controller development team, it felt like the fog that covered the path ahead had finally lifted.
Thus was the Wii born-with an intuitive controller that you could shake and direct at a desired point on the screen.
The Wii development team’s next challenge would be to turn the system that nobody in the family would hate into the system that everybody in the family would love.
Excerpt reprinted with permission. Nintendo Magic: Winning the Videogame Wars is written by Osamu Inoue and published by Vertical, Inc., a division of Random House. It is available now through Amazon and other retailers.