Joukowsky Institute for Archaeology & the Ancient World
Box 1837 / 60 George Street
Providence, RI 02912
Telephone: (401) 863-3188
Fax: (401) 863-9423
Be it a quick round of solitaire or a seemingly never-ending session of Monopoly, games have always been about entertainment. Yet, up until the last century, games were usually a physical object with which one interacted directly. Regardless of whether you move the red jack to the black queen, or roll the dice and move the little dog to Community Chest, you're still directly interacting with the game. However, come the 20th century, this started to change. With the advent of video games, one could no longer draw a card from the top of the deck by picking it up, nor take out a mortgage on Park Place by flipping over the card. The computer kept track of the state of the game, and to modify that state, an input device, or, more specifically, a game controller, was required.
How does one define a game controller? It could simply be defined as an input device for a machine on which you play a game. No, a game controller is more specific than that. It is an input device used for interacting with a computer on which a game is played. Even more specifically, that computer is generally, although not exclusively, a dedicated machine to playing games, regardless of whether it hooks up to a home television, or to an arcade machine. It is the medium through which one interacts with the machine. However, in this project, the scope of the game controller is limited to those which hook up to a home gaming console. The game controller means more than just an input device, though. Icons of game controllers are often used to represent video gaming as a whole, much more often than icons of gaming consoles or other representations. While gaming may be about the games, the controller is the necessary link between the game and the player, and as such, can be just as important as the game. A game will be much less enjoyable on a controller that is hard to use or uncomfortable to hold.
As a game controller is the necessary link between the gamer and the game, the controller also needs to live up to its name and control the game. To do this, there needs to be a mapping from the input on the controller to the event in the game. Good games are about immersion. For a good controller, this mapping must be able to be made by a game in a manner that is intuitive and natural, at least for people who are familiar with video games. For those who are not, the process of mapping button presses and hand motions to various gameplay elements may seem unnatural, as they are used to a full range of motion. As such, some would say that the ideal game controller would be a full virtual reality environment. However, for home entertainment purpose, this is likely, although not necessarily, a pipe dream. At least for now, we are stuck using something held in the hands.
Now, much like all technology, game controllers were not invented fully-functional for all purposes. Just like how the ball mouse had been supplanted, for the most part, by optical mice, so, too, did the game controller evolve. In fact, in the beginning, game controllers were extremely simple, usually consisting of maybe two or three pieces that one could actually interact with. They didn't need to be complex. The games were simple. The very first dedicated game hardware, Tennis for Two, which was built in 1958, used only a knob and a button as input for each player (OSTI, 1981). In fact, Pong didn't even give each player a button, only a knob (referred to as a “paddle”). But as technology marched on, the game controller followed suit, as the simple game controllers just could not provide adequate input for an advanced mapping, much less an adequate experience, for games. It would, however, be a mistake to begin an exploration of game controllers with Pong.
Before Pong, there was the Magnavox Odyssey. The Magnavox Odyssey, which began shipping in 1972, was the world's first home video game console. It cost $100 and could play twelve games. It also featured two paddle controllers. The paddles had two dials: one for vertical motion, and one for horizontal motion (Douglas, 2008). However, as this was the very first home gaming system, it was very basic, and did not yet feature a total disconnect from the more traditional variety of games: the system came with dice, paper cards, and even play money (Gegan, et al., 2003). It did, however, also have a light gun peripheral, called the Shooting Gallery.
The light gun is a very different kind of controller from other game controllers: instead of using it to control something on the screen, it is instead used to shoot at something on the screen. While light guns predate the Magnavox Odyssey, the fact that the very first home game system had a light gun accessory is important. Since then, almost every popular home game system has had a light gun peripheral of some form. Light guns work by having a photosensor inside of the gun that would detect a flash of light from the screen to see if a target was hit. Early light guns could be defeated by pointing the gun at a constant source of light, such as a lightbulb. However, the Magnavox Odyssey did not keep score, making this irrelevant. Needless to say, the Odyssey could not do much compared to modern game consoles.
So we must move on to one of the most well known and iconic classic gaming systems of all time. No, not Pong. The Atari VCS, later known as the Atari 2600. Featuring well known games such as Pitfall!, Breakout, Adventure, and many more, the Atari was a huge hit. The Atari joystick is still well known in today's culture, more than thirty years later. While not the predominant icon of gaming, it is still instantly recognizable as a symbol for gaming, and is still used as such an icon in various places, such as popular technology website Slashdot. Although it introduced the icon of the joystick into modern culture, it had several other controllers, such as a paddle controller (which had a button on the side), a touchpad, which was a device that looked much like a touch-tone telephone, and many third-party peripherals. However, the joystick and the paddle were the main controllers. The joystick could be moved in only four directions and had only one button, and the paddle could only provide one dimensional movement. By today's standards, this would be completely unusable! How would one create a game where the player can both jump and attack? However, a joystick is much more convenient for two-dimensional movement than two separate dials – so how could this be improved?
Along comes the Nintendo Entertainment System, bringing the gaming industry out of the near-fatal video game crash of 1983. Although stores were hesitant to stock the NES due to the fact that a deluge of video game systems and low quality games had saturated the market, and demand was low, Nintendo managed to sneak into toy stores through one of the most clever marketing ploys the gaming industry has ever seen, the Robotic Operating Buddy, or R.O.B. While he only worked with two games, R.O.B., a novelty accessory for the NES that looked like a small robot that would interact with games, managed to get retailers to stock the NES, as it was not marketed solely as a game console (GameSpy, 2003). Although short-lived, R.O.B. did his part by getting the NES, with the very first gamepad, into the homes of children in the United States.
The gamepad was a new approach to game controllers, from which (almost) all modern game controllers descended. It had a new approach to what a game controller should look like: instead of something awkward like a bulky joystick, it was held comfortably in two hands while the thumbs did everything. No more holding a joystick in one hand and the whole controller in the other while somehow trying to press a button at the right time. Rather, just move your thumb a centimeter or two and hit a button. Furthermore, the NES controller had two or four buttons. This allowed, for the first time, separate actions to be performed at a point in the game. A player could now launch a fireball or jump, not just one or the other. It also had the start button, generally used for pausing or starting a game, and the select button, which was typically used for changing modes. The Japanese model of the NES, called the Famicom (short for Family Computer), however, had two controllers hardwired, and only the first one had start and select buttons. The second controller had a small microphone in its place (Nutt, Turner, 2003).
Furthermore, the NES controller introduced to the home game system market something which is employed on all modern game controllers, the D-Pad. Short for directional pad, the D-Pad was a new approach to two-dimensional input. Instead of a large joystick which required a whole hand to operate, it was a simple “plus” shape on which the player would simply press the direction they wished to input. While not original to the NES, the D-Pad was the first implementation on a home game system. The D-Pad is originally from another Nintendo product, called the Game & Watch. Invented by Gunpei Yokoi, Game & Watch was a series of simple portable games. Each Game & Watch device would only play a game or two, but was also small and battery powered, so it was easy to take anywhere. Conceived while Yokoi was watching a man fiddle with a calculator on a train, he realized that a joystick would be too large and cumbersome for a portable game system (Crigger, 2007). His solution was the D-Pad, and it has been used on almost every game controller since. Despite its simplicity, some were not satisfied, and a third party did release a joystick version of the NES controller, called the NES Advantage. However, the joystick quickly fell out of use in home gaming systems, and was replaced by the D-Pad almost entirely. The Sega Master System, which came out shortly after the NES, had a controller that was very similar to the NES's controller.
While the NES did have other controllers, such as the NES Zapper light gun, the gamepad was the real contribution to game controllers. Although blocky and not very comfortable to hold, it was a significant step forward. While the implementation might not have been perfect, the fundamental design of the NES controller has been used in virtually all important game controllers since. The layout set the groundwork for the gamepad, and the D-Pad only helped. Each generation of game controller was almost completely different, and better, than the last. However, with the NES controller, what may have been seen as the ideal fundamental game controller design had been found, and upon which could only be incrementally improved. It was good for the gamer, and good for the game.
Next up to the plate was the Sega Genesis. At first, adding only one main button, it wasn't a huge improvement. Later models added another row of buttons, for a total of six main buttons. The extra three buttons were rarely used. The Sega Genesis was not the star of the fourth generation of game controllers. Rather, that title is reserved for the direct successor to the Nintendo Entertainment System, the Super Nintendo Entertainment System. While adding two main buttons and rounded edges were vast improvements, they were not the biggest improvement. The button configuration may have been the configuration that is essentially still in use today, but the SNES controller had a larger contribution, the shoulder buttons.
Again, Nintendo had invented something that is still in use on game controllers today. The shoulder buttons, generally referred to as L and R, for left and right, respectively, were buttons added to the top corners of the controller. Unlike the buttons on the face of the controller, these buttons are meant to be pressed with the index finger. This, again, allowed for more freedom for the player and for more complex games, not just through extra buttons, but through simultaneous input. Although certainly not as important as the D-Pad, the shoulder buttons would prove useful also for one-directional input, usually for such things as switching menus or rotating a camera. This was a step that allowed for mappings that could not be created with just a D-Pad and a few face buttons.
With the end of the 16-bit era, some would say the golden era of gaming had ceased, even though the next generation introduced something almost entirely new to games. Although there were a few tastes of it in the 16-bit era, they were few and weren't yet that visually impressive. Yes, this was the beginning of 3D games. Early entries in the market, such as the Sega Saturn, did not make a lasting dent in the history of controllers. The Sega Saturn controller was merely a six-button Sega Genesis controller with shoulder buttons. The first PlayStation controller was basically a Super Nintendo controller with two pairs of shoulder buttons and ergonomic handles. Rather, the big contributions to controllers in this generation came, once more, from Nintendo's entry.
The Nintendo 64 controller had an interesting setup. Instead of being held with either hand on each side, it had a “trident” setup, whereby you would need three hands to be able to reach all of the buttons. Only two of the three prongs were meant to be held at a time, and as such, there were effectively three shoulder buttons, with the middle one being the Z button. It did away with the rarely used Select button, leaving a red Start button in the center. It once again had a D-Pad, but it also did away with the last two face buttons the SNES controller had, replacing them instead with the C buttons, which were a sort of a disconnected D-Pad on the opposite side. On the back, it also had an expansion port, which allowed for things to be plugged in such as a memory card. None of these were that big. What was big was what was dead in the center: a thumb-stick. While generally referred to as an analog stick, the N64's thumb-stick wasn't actually analog (Kohler, 2005).
Hearkening back to the joystick, although not to the digital joysticks of most early game consoles, the analog stick allowed not only for varying levels of input, but also input in virtually all directions. It was a critical invention for gaming, especially 3D gaming. For once, a player could tell the game that the character should walk slowly, quickly, or at a medium pace, just with a varying degree of tilt on a single input. No longer was walking limited to left, right, up, down, and possibly binary combinations of two of these. It was free in all directions. Controller manufacturers soon updated their controllers to include analog, as well, with the Sega Saturn's “3D” controller, and Sony's Dual Analog, which included two analog sticks – one on each side of the controller. The second analog stick is usually used for camera control in 3D games. In modern games, the analog stick is used as often, if not more often, than its predecessor, the D-Pad. However, both do exist on modern game controllers.
The N64 introduced something else, too. Although not initially part of the controller, an expansion was released, bundled with the game Star Fox 64. This expansion, called the Rumble Pak, allowed, for the first time, three senses to be engaged in the game. Now, not only could a player see the game and hear the sound, but a player could now feel force feedback from the game as well. This added yet another layer of depth to gaming. While it did not directly affect input, it made the controller something more than just an input device. It made the controller part of the game. Other controllers soon followed suit, with Sony releasing its own version, called the DualShock, which replaced the Dual Analog shortly after the Dual Analog's release.
With the analog stick and rumble, the controller had drastically improved over the previous generation of controllers. Mappings, when done right, could be complex and intuitive – the holy grail of game controls. So the question must be asked again: how could the controller be improved? The next generation of game controllers certainly didn't answer this. Kicking off the sixth generation of video games was the Sega Dreamcast, which would prove to be Sega's last entry into the video game console market. Much like the Sega Saturn's 3D Controller, the Dreamcast contoller was a large circular block. With a layout much like that of the Saturn 3D Controller, albeit with two fewer buttons, the Dreamcast controller was not a huge step. Taking a hint from the N64 controller, it had a slot for memory cards, including the bizarre Visual Memory Unit, which could have a game store a small, different game on it that could be played like a Tamagotchi if removed from the controller. It did, however, have one new feature: analog triggers. Analog triggers are much like shoulder buttons, except generally shaped like the trigger on a gun, and also featuring varying levels of input depending on how hard or how far the trigger is pulled back. However, the Dreamcast did not even last in the sixth generation, and was put out of business by the PlayStation 2, GameCube and Xbox.
Second up in the sixth generation of video game systems was the PlayStation 2. However, the PS2's controller, the DualShock2, made very few changes. Although the DualShock2's buttons were all analog by means of pressure sensitivity (Zdyrko, 2001), this has not made much, if any, impact on gaming. Likewise, the GameCube controller, Nintendo's entry, made few innovations. It was similar to the DualShock, except that it switched the position of the D-Pad and the left analog stick, had a smaller secondary analog stick, called the C-Stick, as it replaced the C buttons on the N64 controller, had two analog triggers and one shoulder button on the right (the Z button), and had an utterly bizarre configuration for the face buttons that made playing some games a chore – although games designed for the GameCube controller generally took advantage of the configuration well. Late to arrive was Microsoft's entry, when they broke into the market with their Xbox, with a hulking beast of a controller that contributed absolutely nothing to controller innovation. The “Duke” controller had too many buttons in odd places and could not be used effectively if the player had hands smaller than a breadbox. It had two analog sticks, albeit one had a location switched with the D-Pad, eight buttons, including two buttons just known as “black” and “white”, analog triggers, slots for two memory cards, and a behemoth Xbox logo in the center, just wasting space. In fact, the controller was so unwieldy that they later released a redesign, called the S controller, that was smaller, lighter, and had a more sensible button layout.
At first, it looked like seventh generation would be mostly maintaining the status quo for controllers, focusing instead on graphics and power. After all, why mess with perfection? Various small things were changed, such as wireless controllers all around, a more aesthetically pleasing controller for the Xbox 360, with the black and white buttons being changed to shoulder buttons to complement the analog triggers, “home” buttons on controllers to allow the game system to return to a home menu, and, from Sony, what appeared to be a silver banana. Or perhaps a silver boomerang. However, Nintendo hid their controller from the public eye until the Tokyo Game Show in 2005 (Torres, 2005). No one was expecting what they revealed.
Up until TGS2005, Nintendo had kept the controller for the Wii (then codenamed the “Revolution”) so secret that rumors had popped up, including one of the most elaborate hoaxes in the history of video gaming. This hoax, known as the Nintendo On, was an expansive, six minute long video that starts with short chronicling of the history of Nintendo's game systems, and moves on to a hypothetical game system, the eponymous Nintendo On. It is shown with a small visor that detects motion and feeds it back into the game as input. A convincing video, it displayed the Revolution as something entirely new, a virtual reality system. It was shown with feedback given to the player through the visor in the form of 3 dimensional video, akin to Nintendo's failed VirtualBoy, and surround sound through four speakers over the ears. While the Nintendo On was a hoax, created by the 23 year old Pablo Belmonte (Klepek, 2005), it set the stage for something almost as impressive.
At first glance, the Wii's controller looked to be a huge regression in game controllers. It looked like a remote control with a few buttons, a D-Pad and a trigger that probably wasn't even analog. It had an expansion port that one could plug a few things into, like a small handheld object with an analog stick and two more shoulder buttons. It sure did look like Nintendo had messed up. However, that's only at first glance. What you couldn't see was the motion sensors. As foretold by the Nintendo On hoax, the Wii Remote (often affectionately referred to as the Wiimote), did, in fact, have motion-sensing capabilities. The demos from TGS2005 showed off the potential of this novel controller (Torres, 2005), and it looked like the Revolution would in fact live up to its name. Nintendo had once again attempted to take a sledgehammer to the boundary between gameplay and real life. The analog stick attachment, known as the Nunchuk, due to its resemblance to nunchaku, also allowed for more traditional gameplay. The button configuration when using the Nunchuk was different from more traditional video games, but it also allowed for a handedness-independent game controller, as the Wiimote and Nunchuk can be held in either hand. It also allowed for gestures with the controller to be recognized. Nintendo even released a Classic Controller that resembles a Super NES controller with an extra set of shoulder buttons and two analog sticks. However, the Classic Controller still suffers from the ergonomic problems that the Super NES controller did. Sony, copying Nintendo and abandoning their silver boomerang design, decided to release a controller with motion sensors, dubbed the SIXAXIS.
While the Wii drastically outsells other consoles on the market (Matthews, 2008), probably due to its lower price and appeal to a wider market, it's still not clear to what extent the Wiimote will affect later generations of games. There are suspiciously few games that effectively use the motion sensing technology. Even though it may allow for more intuitive gameplay, the transition will take some time. If the NES controller is any indication, though, by the next generation, the Wiimote will have taken over. There are a few hints of the future already, however. Though only hints, the future of game controllers does not appear the status quo that may have been predicted during the sixth generation. Nintendo has announced an attachment to the Wiimote that enhances its motion detection capabilities with more accurate and detailed measurements, called the Wii MotionPlus (Ashcraft, 2008). Recently, a patent for a “transforming” controller for the PS3 surfaced (Gino D., 2008) that can be split into multiple pieces and put back together into various shapes.
The past has demonstrated that the game controller co-evolved with technology. As technology became more complex, the games, too, became more complex, and this required the controller to become more complex as well. Life is not as simple as pressing one button. There must be a range of actions available for there to be immersion. However, as a controller gets too complex, the mapping can become unclear. Nintendo tried to alleviate this with the Wiimote, while Sony and Microsoft just made incremental improvements. However, the Wiimote, being something almost entirely different which may have alienated “core” gamers who were used to the gamepad, did attract a new generation of gamers, dubbed casual gamers (McWhertor, 2008). If we are to believe that a more natural mapping is the essence of a well-designed game controller, then the future of the game controller is not incremental improvements. Only time will tell what it really is.
Posted at Dec 22/2008 11:16AM:
chris witmore: Jeffrey, a detailed history of the game controller. As we discussed before other angles to consider here are issues of feel, experience, aesthetics, and bodily engagement with a particular game. Certainly it the latter, the way in which Wii games demand bodily engagement that places Wii in a completely different league.