Analogue Control | Code Like It's 198x

A digital joystick is four switches. Push it and the program learns which way, and nothing else. An analogue control also reports how far: a knob turned a little or a lot, a stick nudged or pushed to its stop. That one extra number changed what games could ask of the player, from putting a bat exactly in the path of a ball to letting Mario tiptoe past a sleeping enemy. Atari put an analogue stick on every 5200 controller in 1982, and players fought with it because it did not spring back to the centre. Fourteen years later the Nintendo 64 put a small thumb stick in the middle of its pad, and within two years Sega and Sony had followed.

Fast facts

The first analogue controls: knobs. The earliest home games “were dedicated ping pong machines controlled with twisting knobs” (Edge, 1996); see paddle controller.
How a stick measures distance: usually a potentiometer on each axis, a variable resistor whose setting the machine times or measures.
Atari 5200 (1982): a two-potentiometer stick that did not self-centre, “an incredibly precise 360° analog joystick” in Atari’s advertising and the 5200’s “Achilles’ heel” in Electronic Games.
Vectrex (1982): a self-centring analogue stick, read by the same converter that drew the picture.
Nintendo 64 (1996): “press up fully and Mario will run, press it halfway and he will walk” (Edge, February 1996). Inside, the stick counts pulses from optical encoders instead of reading potentiometers.
Followers: Sega’s analogue pad with NiGHTS (1996); Sony’s Dual Analog (Japan, April 1997; Britain, September 1997, about £25) and DualShock (Japan, November 1997).

Knobs and paddles

The first analogue controls in the home were knobs. Edge’s 1996 history of controllers noted that the first home systems “were dedicated ping pong machines controlled with twisting knobs”, and the paddle controller entry follows that line from the Magnavox Odyssey and Pong to the Atari VCS and the home computers. A paddle tells the program where the player wants the bat to be, not which way it is moving. A joystick built the same way does that in two directions at once: a potentiometer on each axis, which, in Edge’s words, “changes the amount of current running from the stick to the computer, enabling the computer to know exactly how far and in what direction the joystick has been moved.” PC flight sticks worked like this from the 1980s onwards.

The Atari 5200: analogue, but no spring

Atari’s 5200 reached American homes at the end of 1982 with an analogue stick on every controller. The service manual describes “two pots set in the bottom controller case”, moved by two sliding plates under the stick, and wired to POKEY, the “Pot Keyboard” chip that reads them. The same controller was meant to do every job. Electronic Fun with Computers & Games found in its first hands-on review, in November 1982, that the 5200’s “universal controllers” were “designed to combine a paddle, joystick, and keypad into one handy box”, so Super Breakout used the stick in place of a paddle. The reviewer preferred the old VCS paddles.

Atari’s January 1983 advertisement called the controller “the most advanced in the world. With an incredibly precise 360° analog joystick.” Players were less sure, because the stick had no centring spring. Let go and it stayed where it was. Electronic Fun warned that VCS owners “may find it hard to get used to the new Atari joysticks, which are the non-centering variety”, and Electronic Games called it “the 5200’s Achilles’ heel: its poor, non-centering joystick controller”. Robotron: 2084 needed two sticks at once, and Electronic Games thought they “would have rendered Robotron unplayable” without the stand Atari put in the box; even then, players had to set both sticks “back to center to avoid a fatal first step” between rounds. A reader wrote in to say that a Wico replacement “overcomes the major problem with the Atari 5200 non-centering. It makes Pac-Man playable again”. When Electronic Games reported the console’s end in November 1984, it noted that Wico, High Score and Coin Controls had all made “good, self-centering 5200 replacement controllers”, and that Atari’s plan “to set a new state of the art by providing an analog joystick” had foundered on sticks that “proved frustratingly difficult”. Edge still remembered it in 1996: “If the Atari 5200 was the first massive commercial control failure, it was certainly not the last.”

The Vectrex: the same idea, centred

GCE’s Vectrex, launched the same autumn, got the spring right. Its service manual lists a “Self-Centering Joystick” with two 10K potentiometers beside the four buttons. The console has no separate chip to read it. The processor feeds values to the digital-to-analogue converter that also steers the electron beam, and compares each voltage with the stick’s position “until the MPU generated voltage is equal to the joystick voltage”, a successive-approximation measurement done in software.

The Nintendo 64

Edge’s report from Nintendo’s Shoshinkai show, published in February 1996, described what the thumb stick at the centre of the new controller did for Super Mario 64:

Nintendo technicians claim the pad can detect a full range of 360 directions, affording even the most basic of software a degree of directional control above and beyond the eight-way limit of a PlayStation or Saturn controller. The pad also adds a new dimension of velocity since it is able to detect degrees of movement in particular directions. Mario 64 makes use of these features extensively: press up fully and Mario will run, press it halfway and he will walk, press it a little and in some circumstances he will crawl or tip-toe his way forward.

Shigeru Miyamoto told the magazine that “in the past there have been lots of games that could not have been developed because of a lack of analog control. In fact, the consumer has often had to buy custom controllers as accessories, so companies have been afraid of supporting such peripherals.” Putting the stick in the standard pad meant every owner had one. Edge’s Wave Race 64 review found the jet skis took “quite a bit of getting used to”, and then “an intuitive simplicity rarely found in videogames”. Asked in January 1997 whether players were really pressing “halfway for a gentle turn” or still making “the ‘tap, tap, tap’ turns that were required with a traditional eight-way joypad”, Miyamoto said the pad “has been accepted with open arms” and that “most people manage to master it after about two hours’ playing time”.

The stick was not what it appeared. Genyo Takeda, whose R&D3 team designed the controller, told Edge in August 1996 that it was “not really analogue by the strictest definition of the term, but that it has been engineered to emulate an analogue joystick”, and the magazine could not confirm rumours that it was optical. It is. Homebrew developers’ documentation describes two slotted optical encoder wheels, whose pulses report movement rather than position; a chip in the controller adds them up from zero at switch-on and sends each axis as a signed byte. So the centre is wherever the stick rested at switch-on, and pressing L, R and Start together resets it to zero. Takeda said R&D3 “tried many different types [of controllers] and prototypes and we threw them away”, among them a motion-sensing wristwatch that got as far as a patent application.

Sega’s NiGHTS pad

Sega’s analogue pad for the Saturn was built for one game. Yuji Naka told Edge that Sonic Team wanted a “smooth and comfortable” feeling in NiGHTS into Dreams’ flying and “could not get this to a satisfactory level with the current Saturn pad”, so the team designed a controller to ship with the game: a “fattened” Saturn pad with an “N64 like analogue thumb-pad” above the digital one. It was not the first analogue controller for a home console, nor Sega’s first: the same Edge issue reported that Sega was “releasing an analogue control stick” for the Saturn, and Sega’s programming manual documents an analogue Mission Stick alongside it. In Britain, Sega Saturn Magazine expected the pad to go on sale with NiGHTS on 5 September 1996, at £24.99 alone or £59.99 with the game.

Edge thought NiGHTS “doesn’t really need analogue control” but liked the pad: “Although not as striking as the N64’s, the design is reasonably comfortable and easy to use”, and it added “a new dimension” to existing games such as Sega Rally. Sega’s manual shows what a program received: in analogue mode the pad sends its buttons, then an unsigned 8-bit reading for each of the stick’s two axes and for each of the L and R shoulder buttons, which were analogue too.

Sony’s pads

Other companies were already selling analogue add-ons for the PlayStation. Namco’s NegCon steered by being twisted in the middle, “an awkward analogue controller operated by twisting it” in Edge’s 1996 caption. Sony’s own two-stick pad, the Dual Analog, came out in Japan in April 1997 with vibration motors. Edge was previewing games that “work with Sony’s new dual analogue pad” that June, and in September it reported that the pad had reached British shops “three months earlier than expected”, for about £25, supported by one PAL game, Porsche Challenge. The Western version had no motors; Sony said “market research revealed that gamers found the analogue feature to be most important”. Sony’s hardware manual gives the pad two sets of button assignments, one for “Analog Controller Mode” and one for “Analog Joystick Mode”. The motors came back in the DualShock, launched in Japan in November 1997, whose entry follows the story from there.

Sources

Electronic Fun with Computers & Games volume 1 number 1, November 1982: Atari 5200 hands-on review (the non-centring sticks, the “universal controllers”, Super Breakout).
Electronic Games volume 1 number 8, October 1982: preview of the new systems, the 5200 and GCE’s Vectrex among them.
Electronic Games volume 1 number 11, January 1983: Atari 5200 advertisement.
Atari, CX5200 Field Service Manual (1983), sections 1 and 6: the game controller, its two potentiometers, and POKEY’s pot lines.
Electronic Games volume 2 number 9, November 1983: the 5200 stick as “a non-centering joystick”.
Electronic Games volume 2 number 12, March 1984: Centipede (5200) review.
Electronic Games volume 2 number 14, September 1984: letters (the Wico joystick).
Electronic Games volume 2 number 16, November 1984: “The Supersystem Bows Out” (the 5200 discontinued, replacement controllers, Atari’s intention) and Robotron: 2084 review.
General Consumer Electronics, Vectrex service manual (1982): controller specification, successive-approximation joystick reading, parts list.
Edge issue 29, February 1996: Shoshinkai report (Super Mario 64 and the stick) and Shigeru Miyamoto interview.
Edge issue 35, August 1996: game controllers feature (knob-controlled ping pong machines, potentiometers, the 5200, Genyo Takeda on the Nintendo 64 controller, Sega’s analogue stick, the NegCon); NiGHTS preview and Yuji Naka interview.
Sega Saturn Magazine issue 10, August 1996: news (the analogue pad’s UK release and prices) and NiGHTS preview.
Edge issue 36, September 1996: NiGHTS review.
Edge issue 39, December 1996: Wave Race 64 review.
Edge issue 41, January 1997: Shigeru Miyamoto interview.
Edge issue 46, June 1997: Gex: Enter the Gecko prescreen.
Edge issue 49, September 1997: “PlayStation analogue joypad sneaks into UK”.
Sega, SMPC User’s Manual (1997, Japanese): the multi controller’s analogue mode data format; the analogue Mission Stick.
Sony Computer Entertainment, PlayStation Hardware (1998), appendix A: the analog controller’s two modes.
Wikipedia, “Dual Analog Controller” and “DualShock” (web, secondary): Japanese release dates of April and November 1997.
N64brew wiki, “Controller” and “Joybus Protocol” (web, secondary): optical encoders, relative counting from zero, signed axis bytes, the L+R+Start reset. https://n64brew.dev/wiki/Controller