The pad decides which games exist

It is tempting to treat a controller as plumbing. The game is the idea, and the pad is how the idea reaches the hands. The pad is upstream of the idea. A cluster of buttons, two sticks, and a pair of triggers decides which actions are cheap to attempt, which games a stranger can learn in a minute, and which fantasies are uncomfortable to hold for an hour.

That is why so many games share a shape. The face buttons invite a jump, a strike, a dodge, and a tool. A design that needs six of those at once is fighting the object in the player's hands. The interesting work is not to ignore the limit. It is to notice which game the limit is quietly volunteering to be, and to decide whether that is the game you meant.

Yes, the controller draws a boundary around what gets designed. That cluster of buttons, sticks, triggers, and bumpers is not a neutral pipe between player and game. It decides which games are cheap to build, quick to explain, and comfortable to play, and those three pressures quietly steer what anyone bothers to attempt.

Arcade cabinets escaped the problem by building the input to fit one game. Pong ran on dials. Dig Dug needed nothing more than a joystick. Robotron handed the player two sticks so it could throw chaos at them from every direction. Marble Madness used a trackball, because rolling a marble by rolling a ball is an idea that needs no explanation at all.

Consoles could not work that way. One controller had to serve every game on the machine. The Atari 2600 joystick was an early attempt and the Atari 5200 phone pad a less graceful one, but the standard that stuck came from the NES. Its pad was superb at pushing a character around a 2D plane, which is what Mario and Zelda ask for. It was correspondingly poor at everything else. Marble Madness lost something in translation, the Robotron sequel Smash TV never sat right on it, and anything resembling twin-stick play had to wait for hardware with more inputs to spare.

Each addition to the standard controller answered a problem games were already straining against. The pad grew where designers kept running out of room.

Three-dimensional space was the first hard stop. A d-pad has eight directions, and 3D movement wants everything in between as well. Early PlayStation games such as Tomb Raider carry that awkwardness openly, moving a character through depth with habits inherited from flat games. Nintendo reset the standard again with the N64 analog stick, and steering Mario through a 3D world stopped being a negotiation.

Sony followed with the Dual Analog controller and then the DualShock, which brought a second stick. For a while nobody knew what the extra stick was universally for, beyond curiosities like swinging a net in Ape Escape. Then developers worked out that it could drive the camera, and the console first-person shooter became viable. Medal of Honor Underground and Alien Resurrection point that direction, though players took time to adapt: a review from 2000 treated the twin-stick layout in Alien Resurrection as very nearly unplayable.

The modern pad is the fossil record of that argument. Every part of it corresponds to a demand games kept making: move a body through space, aim a view or a crosshair, and manage both at once precisely enough for combat to function. Even now that last part leans on the game meeting the hardware halfway, through generous auto-aim and enemies slow enough for a thumb on a stick to track.

Plenty of interactions never sit comfortably on a pad, and they tend to share a shape: they want either instant access to any point on screen or fine, small movements of the hand.

Real-time strategy is the clearest case of the first. The genre depends on reaching any part of the map immediately, which a stick-driven cursor cannot do, and console versions have historically felt like wading. The exceptions confirm the diagnosis. Overlord and Pikmin put the player inside a character who walks around the space, which drags the genre back toward what the pad is genuinely good at.

Small, precise hand movement is the second case. Drawing on a whiteboard with an analog stick, as Duke Nukem Forever asks, exposes how blunt the instrument is at that scale. Now imagine Trauma Center on an Xbox pad. A game built on quick, exact surgical incisions would stop reading as surgery and start reading as slapstick.

Trauma Center wants a more direct instrument, which is why it lives on the Wii remote and the DS touchscreen. Both devices are full of games that would be diminished or impossible on a pad, because the interaction itself is inseparable from the hardware performing it.

Direct input does not port pad games to a new surface. It produces a different category of game, because the core action is the physical gesture rather than a symbol standing in for one.

The DS touchscreen gave us Elite Beat Agents, Kirby Canvas Curse, Cooking Mama, and Drawn to Life. The touch devices that followed produced Fruit Ninja, Flight Control, Blek, Eliss, and Cut the Rope. None of those are pad games with a finger substituted for a stick. Their verbs are touching, slicing, dragging, drawing, and moving pieces of the screen around.

The Wii found a different flavor of directness. Its games have the player hold a sword, swing a racket, aim a flashlight, or take a drink by moving the remote roughly the way the character moves. The link between the body and the screen is short enough that the idea arrives before the rules do.

That is the real advantage. The player is not decoding an arbitrary mapping first; they are doing something that already resembles the intended action. A touchscreen can make slicing fruit or pulling a lever self-evident. A motion controller can make tennis playable before anyone has memorized a button diagram.

A standard pad is an abstract device, and abstraction has to be taught. Swinging a tennis racket becomes pressing the green button, and every game using a pad has to explain its own vocabulary before play can start. The cost of that lesson shapes what gets made.

The first effect is gravitational. Designers drift toward genres whose control language players already carry. Once aim, fire, crouch, reload, sprint, and melee sit in the same places across dozens of shooters, the shooter becomes the cheapest game to hand someone. Hardware, genre, and muscle memory reinforce each other.

Those inherited expectations are also a resource. Portal borrows the entire grammar of a first-person shooter to deliver a puzzle game with no shooting in it. Looking, moving, and aiming come free, so the game can spend all its teaching on portals rather than on the interface.

The second effect is a ceiling. Abstract controls can only carry deep mechanical sophistication across a handful of actions. Push past that and every new capability needs its own tutorial, and the player ends up maintaining a mental list of special-case commands instead of playing.

Because a pad supports few actions well, most games commit to one or two and build everything on them. That constraint is not automatically a loss; a tight set of verbs is often what makes a game legible.

Call of Duty grants enormous precision in aiming and firing, yet the notorious cemetery scene reduces mourning to holding a single button. The fix is not making the player manually steer a hand onto a coffin. There is genuine elegance in a game that knows which single mechanic it is about.

The catch is repetition. Those one or two interactions get performed thousands of times before the credits, so they have to survive the grind. Punching, slashing, shooting, driving, and jumping became staples because they repeat well, vary easily, and stay expressive under pressure from timing, movement, and difficulty.

Other games want a character who does many small things. Max in Life is Strange can play a guitar, water a plant, or take a photo. Routing all of that through contextual button prompts flattens it: the tactility, the challenge, and any sense of mastery drain out of actions that are only ever a prompt.

Monument Valley and The Room show the alternative on a touchscreen. Poking, pulling, prodding, and twisting the world is satisfying because the player is handling the object itself instead of picking an entry from the pad's short menu.

VR controllers are interesting precisely because they are good at the things a pad is worst at, and weak at the things a pad does effortlessly. They are a trade, not an upgrade.

Oculus Touch, the Vive controller, and PlayStation Move support interactions that would be miserable on a pad. Giant Cop lets the player pluck criminals off the street. I Expect You to Die and Job Simulator let them reach out, grab, pull, and prod objects with no contextual prompt and no memorized layout in between.

The limits show up immediately in movement. These devices excel at hands, objects, and manipulation within arm's reach, and are poorly suited to walking a character across a large space. That is why so many games built around them keep the player seated or confined to a small tracked area. The hardware's strengths and weaknesses push the design there.

Every controller is good at something and bad at something else, and the designer's job is to build with the grain. That is not the same as obeying convention. Some of the most interesting work comes from asking what a familiar device could do that nobody has asked of it.

A reinvented control scheme can do more than feel novel. It can carry the theme, generate real skill, and change what the game is about.

Skate throws out trick combos and has the player flick the right stick to ollie and kick. The choice supports the grounded tone of the series, because the player and the skater are both executing a tricky physical motion. It also creates skill worth having: advanced tricks demand practiced input rather than a remembered sequence.

Others push further in the same direction. Grow Home maps the triggers to a left and a right hand for climbing. The Fight Night games move punches onto the analog stick. Katamari Damacy uses both sticks together to imitate shoving a huge ball with two hands. Brothers gives the player one character per stick, then uses that arrangement of the hands to tell a story about siblings.

Brothers is the clearest evidence that input can hold meaning. The controller is not a convenience layer sitting outside the work; it can be part of the theme, the emotion, and the structure. It is also why a touchscreen port with a virtual joystick can damage a game so badly: the original was built around a relationship with the hardware that the port discards.

Controllers do dictate what is possible, and abstract pads narrow the field further by rewarding familiar mappings, repeatable verbs, and genres that already fit. None of that is a cage. Designers keep finding new play by reinventing the wheel, or the analog stick.

If a game genuinely does not suit a pad, something else may: a motion-sensing remote, a multi-touch screen, Valve's mouse-like Steam Controller, or whatever the next platform holder decides to experiment with. There is no best controller in the abstract. There is only the question of what this particular game needs the player to do.

So start with the hands. Which action will be repeated for hours? Which one has to feel direct? Which deserves a skill ceiling? Which should shrink to a prompt because the game is really about something else? Answer those and the controller stops being a late-stage detail and becomes part of the design.

The answers are hard to reason about on paper, though. Whether an action survives its ten-thousandth repetition is something you find out by getting the build running and playing it. Your own hands are the test.

In the Flockbay app, name four buttons and nothing else, then describe the game those four can support. Play it. If you immediately want a fifth button, either the action is wrong or it can live on a direction you already have. Add the button only after the four have failed in your hands.

A fighting game template makes the limit obvious. The AI fighting game maker will happily invent more moves than the hands can hold. Refuse the extras.