Chapter 36: The History of Game Design — Key Innovations That Changed What Games Could Be

You are not the first person to design a game. You are not the first to design a platformer, an RPG, an action-adventure, a roguelike, or a procedurally generated open world. Every mechanic you reach for has a lineage. Every "obvious" design pattern was, at some point, a leap of imagination by someone who did not have it as a reference. The double-jump, the save point, the minimap, the skill tree, the difficulty curve, the boss fight, the collectible — each was invented somewhere by someone, and each has rippled forward through forty-plus years of subsequent games until it became part of the air designers breathe.

This chapter is not a history of video games. There are entire books for that, and the further reading at the end will point you to several. This is a history of design innovations — the specific moments when a game introduced (or perfected) a mechanic, structure, or philosophy that changed what subsequent games could do. Understanding this lineage is not academic. It is the difference between believing you have invented something new and recognizing that you are standing on the shoulders of designers who solved a problem you did not know you were inheriting.

There is a tired cliché that those who do not study history are doomed to repeat it. In game design the truth is more specific and more useful: those who do not study history are doomed to reinvent it badly, slower, with less polish than the person who already solved it in 1985. The designer who knows that Metroid (1986) invented gated progression as a deliberate world-building technique can use that pattern with intention. The designer who does not know will accidentally invent a worse version of it, ship it, and then hear from a thousand reviewers that "this game owes a lot to Metroid." Better to owe knowingly than to plagiarize unconsciously.

I want to make a stronger claim. Knowing the history makes you a better innovator, not a more derivative one. The reason is simple: when you know which problem an old design pattern was solving, you can recognize the moments when the problem still needs solving and the moments when the problem has changed and the pattern no longer fits. Save points solved the problem of cartridge memory in the 1980s. The problem went away in the 2000s. Designers who kept using save points anyway because "RPGs have save points" missed the chance to ask whether their game's save system should reflect a different design intent. Dark Souls asked that question and gave the bonfire as an answer. Celeste asked it and gave per-screen instant respawn. Hades asked it and gave death itself as the save mechanic. None of those answers exist without someone first understanding what save points were for.

So this chapter walks through the eras chronologically, then circles back through a thematic catalog of specific innovations that deserve recognition. We will be specific. We will name games, years, designers, studios. We will debunk a few "first ever X" claims that you will hear in podcasts and reviews. We will end where every chapter in this part ends — back at your own progressive project — by asking you to identify three specific historical innovations that your game depends on, and to name the designers and games you owe.

You are in conversation with every game that came before yours. The first step in joining that conversation honestly is to learn what has already been said.

Why Design History Matters

The question "why study the history of design?" deserves a real answer, not a slogan. Let me give three.

First, fluency in references. Every review of your game will compare it to the games it resembles. Reviewers, players, fellow designers, your publisher, your trailer-cutting marketing team — all of them are looking at your game through a frame built from games they have already played. If your game has a lock-on combat camera, somebody is going to bring up Ocarina of Time (1998) and Dark Souls (2011), and you should know what they are talking about and what the differences are. If your game has a hub-and-spokes level structure, somebody is going to mention Mario 64 (1996) and Crash Bandicoot (1996), and you should be able to articulate why your structure resembles or departs from theirs. Designers who lack fluency in reference are at the mercy of comparisons they did not choose. Designers who have it can shape the conversation about their own work.

Second, technique recovery. Many of the cleverest design solutions in games are old. The first wave of arcade games solved problems — how to keep a player engaged across a short coin-pop session, how to escalate difficulty without an explicit difficulty selector, how to teach a control scheme in seconds — that modern designers still face but rarely study. Mobile designers who study Space Invaders (1978) and Galaga (1981) understand session-pacing in a way that mobile-only thinkers do not. Dwarf Fortress (2006) and NetHack (1987) have solutions to systemic emergence that AAA studios are still trying to catch up to. The history is a library. It is full of solved problems. You are allowed to walk in, take what you need, and use it.

Third, innovation by composition. A great deal of what looks like innovation is actually the recombination of older patterns into new contexts. Hades (2020) is a roguelike (1980, Rogue) plus an action game (1988, Streets of Rage lineage) plus a visual novel (1994, Otogirisō lineage) plus a procedurally narrative dialogue system (a true innovation). The brilliance is in the composition. Supergiant did not invent any of the components from scratch. They knew the components well enough to combine them in a way nobody else had tried. This is the most realistic path to innovation: mastery of existing patterns, recombined.

Against these gains, the cost of studying history is small. You play some old games — many of them are now free, in the browser, on the Internet Archive's MAME cabinet — and you read a few books. The return on the investment is permanent. You will look at every new game you encounter for the rest of your career through a deeper frame.

💡 Intuition: When you find yourself thinking "I should add fast travel to my open world," ask: who first did this, in what game, why did they need it, and is my situation enough like theirs that I should copy or different enough that I should diverge? Replace "fast travel" with any feature. The exercise costs you ten minutes and prevents you from shipping a poorly-fitted version of a solved problem.

Pre-Digital Roots (Before 1950)

Game design did not begin with computers. It began thousands of years ago, and the templates the ancients gave us still drive how digital games work.

Go, played in some form for at least 2,500 years in China, is the cleanest example of a design philosophy modern designers should internalize: thin rules, infinite depth. The rulebook fits on a single page. Two players take turns placing stones on a 19×19 grid; stones surround empty space to claim territory; the player with more territory wins. That is essentially the entire game. From those rules emerge a strategic and tactical universe that no human or computer fully maps. AlphaGo's 2016 victory over Lee Sedol did not exhaust Go; it merely showed that the game's depth exceeds human cognition by a margin we did not previously appreciate.

The lesson for digital designers is that complexity in rules is not the same as depth in play. Many young designers reach for complex rules — more stats, more systems, more buttons — when what they actually need is depth, and depth is best achieved with fewer rules that interact richly. Tetris (1984) is the digital descendant of Go's design philosophy. So is Threes! (2014). So is Into the Breach (2018). So is every well-designed roguelike. The Go-like design template — small rule set, big interaction surface — is the most underused design technique in modern game development.

Chess, codified in roughly its modern form by the 16th century, gave us a different template: asymmetric pieces with distinct movement rules, combined with a win condition that creates indirect pressure. The objective is not to capture the most pieces; it is to threaten the king. This indirect goal forces players to think positionally rather than transactionally. The chess template runs through every tactics game (Fire Emblem, XCOM, Into the Breach, Advance Wars), every game with distinct unit types, and every design where the goal is not "destroy them all" but "achieve a specific board state."

Tabletop wargaming matters more than most digital designers realize. Kriegsspiel, the Prussian military training simulation developed around 1812, was the first formal "rules-based simulation of conflict for non-recreational purposes." It introduced the concepts of hidden information, the umpire role (the proto-game-master), dice for resolving combat, and hex/grid movement. A century later, Avalon Hill commercialized hex-and-counter wargames in the 1950s, and out of that lineage came Dungeons & Dragons (1974), which gave us the fantasy archetypes, the experience-point progression, the modular dungeon, and the dungeon master — and from D&D came essentially every CRPG ever made. Ultima, Wizardry, Final Fantasy, Baldur's Gate, Mass Effect, Persona, Witcher 3 — all are children of D&D, which is a child of Avalon Hill, which is a child of Kriegsspiel.

Commercial board games in the 20th century — Parker Brothers, Milton Bradley — established the template of mass-market game design: a box, a board, components, instructions, and the assumption that the box would be played repeatedly by friends and family. Monopoly deserves its own paragraph because its design history is famously misrepresented. The game we know was based on Elizabeth Magie's Landlord's Game (1903), which she designed to demonstrate the harms of land monopolies. Her game had two rule sets — one anti-monopolist, one pro-monopolist — and was a teaching tool. The Parker Brothers version that became famous kept only the pro-monopolist ruleset, stripped Magie's name from the credits, and sold tens of millions of copies. The lesson is sharp: design intent is fragile, and once a game escapes its designer it can come to mean the opposite of what was intended. Every modern designer who watches their work twisted by community interpretation can reach back and shake hands with Elizabeth Magie.

Pre-digital games gave us, then: thin-rules-deep-play, asymmetric pieces, indirect win conditions, hidden information, hex maps, character progression, and the notion that game design encodes ideology whether the designer wants it to or not. Every one of these still drives digital design today.

The Mainframe Era (1950s–1970s)

The first generation of computer games was made on machines that were not built for games — research mainframes, oscilloscopes, dedicated TV-display hardware — by engineers who built games as side projects, demonstrations, or jokes.

Tennis for Two (William Higinbotham, 1958) is often cited as the first interactive video game. Higinbotham, a physicist at Brookhaven National Laboratory, built it for the lab's annual visitor day on an analog computer wired to an oscilloscope. The game was a side-view tennis simulation; players turned knobs and pressed buttons to time their swings. It existed for two visitor days, then the equipment was disassembled. Higinbotham did not patent it. He did not consider it important. He went back to his actual work on nuclear non-proliferation. Tennis for Two is the first lesson in the history of game design: the medium did not look like a medium yet, and the people who built its first artifacts had no idea what they were starting.

Spacewar! (Steve Russell and friends, MIT, 1962) is the more important origin point because it was the first networked competitive game and the first to be shared and modified. Two players controlled spaceships on a PDP-1 minicomputer, firing torpedoes at each other while a star at the center of the screen exerted gravity on both ships. Spacewar! introduced (or at least firmly established) several patterns: real-time competitive play, a physics simulation as core mechanic, hyperspace as a panic button, and — critically — open distribution. The game was copied to every PDP-1 in existence within a year. Russell never tried to monopolize it. The notion that games could be made by enthusiasts and shared freely was set at the very beginning. Modern open-source game development, the modding scene, the indie ethos — all carry a thread back to Spacewar!.

The Magnavox Odyssey (Ralph Baer, released 1972) was the first home video game console, and Baer is the legitimate father of the home games industry. The Odyssey shipped with twenty-eight games on plastic overlays — translucent sheets you taped to your TV screen because the system itself produced only white squares. Baer's contribution was institutional rather than mechanical. He proved that games could ship as a consumer electronics product, that families would buy them, and that the underlying business model was viable. Atari and Nintendo would later eat his lunch, but Baer made the meal possible.

Pong (Atari, designed by Allan Alcorn, 1972) is the first commercially successful video game and is therefore the template for the entire video game business. The design — two paddles, a ball, score on top — was deliberately simple because Bushnell and Alcorn wanted something a drunk person in a bar could play. The trick was the physics tuning: the ball changed angle based on where it hit the paddle, so good play meant placing the paddle precisely. Most of Pong's downstream influence is business rather than design — the coin-op model, the single-screen arcade cabinet, the public-space play context — but its design lesson is also worth keeping: make the input simple; make the consequences of input nuanced. This is still good advice.

The mainframe era ends with a baseline established. Games are interactive, they can be competitive, they can run on consumer electronics, and they can be sold profitably in public spaces. What has not yet been figured out is everything else: how to make games that hold attention longer than thirty seconds, how to give them characters and stories, how to teach control schemes, how to vary difficulty within a single play session. The next decade solves most of these in the arcade.

The Arcade Golden Age (1978–1983)

The five years between 1978 and 1983 are arguably the densest period of design innovation in the medium's history. The arcade environment forced designers to solve hard problems quickly: a player has a quarter, three minutes, and a thousand other distractions in the room. How do you hold them, escalate them, and make them put in another quarter? Every solution invented for that problem became a foundational pattern for everything that followed.

Space Invaders (Tomohiro Nishikado, Taito, 1978) introduced two patterns that have never gone away. First, difficulty that scales with progress through emergent mechanics. The aliens descend faster as you destroy them, because each destroyed alien reduces the CPU's per-frame workload — Nishikado did not deliberately design the speedup, but he kept it once he saw how it worked, and the principle of the game getting harder by virtue of player success became a foundational technique. Second, the high-score table. By recording the top scores publicly on the cabinet, Space Invaders created a meta-game of competition and replay that did not exist before. The arcade industry had its first true social loop.

Asteroids (Atari, 1979) added vector graphics and the brilliant five-button control scheme — rotate left, rotate right, thrust, fire, hyperspace — that made the ship feel like a physical object you were piloting. Asteroids is also where the initials-on-the-high-score-screen tradition reached its iconic form. Three letters. The whole arcade population was suddenly competing to leave their initials behind.

Pac-Man (Toru Iwatani, Namco, 1980) is more important than most retrospectives admit because it solved several problems at once. It introduced character: Pac-Man and the four ghosts (each with a distinct AI personality — Blinky chases, Pinky ambushes, Inky flanks, Clyde wanders) are the first video game characters that audiences could love and merchandise. It introduced the attract mode, the looping demo that played on the cabinet to draw players in. It introduced cutscenes, brief intermissions between levels that broke the gameplay rhythm and gave a sense of narrative. And it introduced the design discipline of targeting an audience the existing market wasn't serving. Iwatani designed Pac-Man specifically to attract women and couples, who were not playing the shooter cabinets. Pac-Man broke the medium out of the male-teen demographic for the first time, and its commercial success demonstrated that who you design for shapes what your game has to be.

Donkey Kong (Shigeru Miyamoto, Nintendo, 1981) is where the platformer is born and where Miyamoto begins his career. Donkey Kong is also the first arcade game to have narrative framing built into the mechanics: a princess (Pauline) is captured by a giant ape; the player (an unnamed carpenter who would later be renamed Mario) climbs scaffolding to rescue her. The "rescue the princess" framing is now a tired cliché, but in 1981 it was a leap — most arcade games had no narrative at all. Donkey Kong also gave us the platformer's grammar: jumping, climbing ladders, hammers as power-ups, escalating level difficulty across four screens. Miyamoto designed it because Nintendo's previous arcade game, Radar Scope, had bombed in North America and they had thousands of unsold cabinets in their warehouse. Donkey Kong is what they put on the boards.

Defender (Eugene Jarvis, Williams, 1981) deserves its own paragraph because it is the patron saint of "games for serious players." Defender's control scheme — five buttons plus a two-way joystick, with the left/right direction of the ship controlled separately from movement — was punishingly complex. New players bounced off it within a quarter. Experienced players described the moment of "getting" Defender as a breakthrough. The game's commercial success despite its difficulty was a proof of concept that there was a market for games that demanded mastery. Every subsequent "hardcore" game owes Defender a debt, and so does the Dark Souls design philosophy of the difficulty is the point.

Tron (1982) is the first significant licensed game and the start of a long, mostly ugly history. Disney licensed the Tron property to Bally Midway, the game shipped, it sold reasonably, and it established a template that the industry would abuse for the next four decades: the licensed-property game built quickly to ride a film's marketing, often with shallow design, optimized for shelf appeal rather than play. We will return to this shortly when we get to E.T..

The arcade golden age also saw Galaga (Namco, 1981) refine the fixed shooter, Donkey Kong Jr. (Nintendo, 1982) prove that sequel design was a discipline of its own, QBert (Gottlieb, 1982) experiment with isometric perspective, and Dragon's Lair* (Cinematronics, 1983) attempt full-motion-video gameplay via laserdisc — a technological dead end that nonetheless prefigured the FMV era of the early 1990s.

By 1983, the arcade was producing more revenue than the entire film industry in North America. Then it collapsed.

📊 Historical Footnote: The arcade's 1982 peak revenue, adjusted to today's dollars, exceeded Hollywood's box office. Most contemporary players do not know this. The arcade was the dominant form of public entertainment in the United States for a moment, and that moment was brief.

The Crash and the Console Rebirth (1983–1989)

The North American video game crash of 1983 is the most studied collapse in gaming history. The short version: the console market was flooded with low-quality games, retailers lost faith, returns piled up, publishers went bankrupt, and the home market essentially ceased for two years. Atari took the worst of the damage. Hundreds of thousands of unsold E.T. the Extra-Terrestrial cartridges (rushed to market in five weeks to ride the film) were buried in a New Mexico landfill — a fact that was treated as urban legend until 2014, when an excavation confirmed it.

The lesson the industry learned from the crash, broadly, is that quality control matters and shovelware kills markets. Nintendo's response when they entered North America in 1985 was the Seal of Quality, the literal gold sticker on every authorized NES cartridge. The Seal was a quality signal to retailers and parents — if it has the Seal, Nintendo has approved it; if it does not, do not stock it. The Seal also locked Nintendo into a controlling-publisher position that gave them enormous leverage over third-party developers. Modern app store curation, console certification processes, and the entire concept of "first-party publishing standards" descend from the Seal of Quality. The crash is also why game development became, for a generation, a console-centric business: PC was less affected because it had more direct distribution to enthusiasts, but the consumer-electronics retail channel had to be rebuilt from scratch, and Nintendo did the rebuilding.

Then Nintendo made the games that justified rebuilding the market.

Super Mario Bros. (Shigeru Miyamoto, Nintendo, 1985) is the most important game ever made by the standard of "design patterns it established that are still in universal use." The side-scrolling platformer with momentum-based jumping, power-ups (mushroom, fire flower, star), enemy variety, level-end flagpoles, hidden bonus areas, world maps, secret warp pipes — every one of those patterns was canonized here, and the canonization stuck. World 1-1 alone, which gets its own case study in this chapter, is studied as a masterclass of teaching-through-play. The deeper innovation, though, is the power-up architecture: a tiered state system (small Mario / big Mario / fire Mario) where damage demotes you down a tier rather than killing you outright. This is the pattern that Hollow Knight's shade system, Cuphead's parry windows, and a thousand other modern games still use as their basic risk-reward unit.

The Legend of Zelda (Shigeru Miyamoto and Takashi Tezuka, Nintendo, 1986) introduced the open-overworld plus discrete-dungeon structure that runs through every action-adventure game since. The player drops into a top-down map with no instructions; the world is gated by items the player has not yet found; dungeons are concentrated puzzle environments containing those items; the loop is explore the overworld, find a dungeon, beat the dungeon, gain the item, unlock new overworld areas. This loop is so durable that Breath of the Wild (2017) — three decades and a console generation revolution later — explicitly returned to it. The first Zelda also introduced battery-backed save on a home console, allowing the player to save their progress and return later. This was a hardware innovation that enabled a design innovation: games could now be longer than a single sitting.

Metroid (Yoshio Sakamoto and Hiroji Kiyotake, Nintendo, 1986) introduced the design pattern that bears its name: Metroidvania — an interconnected map gated by abilities, where the player constantly returns to previously-visited areas after acquiring new movement options. The bombs let you destroy certain blocks; the high jump opens vertical paths; the morph ball squeezes through tunnels. Metroid's atmospheric isolation — the silence, the alien environment, the lone protagonist (later revealed to be a woman, which was its own innovation in 1986) — established that platformers could carry mood as well as mechanics. The Metroidvania template, refined later by Castlevania: Symphony of the Night (1997), became the entire foundation of modern indie game design: Hollow Knight, Ori and the Blind Forest, Axiom Verge, Dead Cells, Blasphemous, Hollow Knight: Silksong, and your own progressive project all live in this lineage.

Final Fantasy (Hironobu Sakaguchi, Square, 1987) almost did not exist. Square was nearly bankrupt; Sakaguchi gave the game its name partly because he expected it to be his last project before leaving the industry. The game succeeded, the studio survived, and the JRPG template was established: turn-based party combat, character classes, a long narrative arc, equipment progression, and a world map separate from town and dungeon screens. The JRPG is the most stylistically distinct genre in games — every JRPG, even today, is recognizably continuous with Final Fantasy (1987) — and the template's resilience is itself a design lesson. Persona 5, Octopath Traveler, Sea of Stars, Chained Echoes: all read as direct descendants thirty-plus years on.

Mega Man (Capcom, 1987) introduced boss-weapon-acquisition as a core progression mechanic: defeat a themed boss, gain their weapon, use that weapon against the boss it counters. The non-linear stage-select screen — pick any of eight bosses in any order — was a small but powerful innovation that gave the player meaningful choice in pacing without complicating the level design. Hades, Cuphead, and many modern action games still echo Mega Man's structure.

By 1989, the console rebirth is complete. Nintendo dominates. The platformer, action-adventure, and JRPG genres are all canonical. Save systems exist. Long-form games exist. The medium has expanded from "three minutes per quarter" to "thirty hours per cartridge," and the design problems have changed accordingly.

The Golden Age of PC (Late 1980s–1990s)

While Nintendo rebuilt the console market, PC gaming developed on a parallel and largely separate track. PC was the platform for deeper, slower, more complex games — the platform where designers could ship a 200-page rulebook with their RPG and find an audience that would read it.

The Ultima series (Richard Garriott, Origin Systems, 1981–1999) is where the modern CRPG was built. Ultima IV: Quest of the Avatar (1985) introduced something that subsequent RPGs are still catching up to: an alignment system based on virtues rather than combat. The game's quest was not to defeat a final boss; it was to become a virtuous person, measured by your behavior across eight virtues (honesty, compassion, valor, justice, sacrifice, honor, spirituality, humility). The endgame was a moral examination, not a fight. Ultima IV is the first game that took adult moral seriousness as its design subject, and it remains one of the few games willing to be earnest about ethics rather than ironic. Disco Elysium (2019) is the most direct modern descendant; Undertale and Planescape: Torment both owe Ultima a debt.

Sid Meier's Pirates! (MicroProse, 1987) introduced the open-ended sandbox in a way that had no precedent. There was no story. There was no required ending. You sailed, you fought, you traded, you recruited, you danced with the governor's daughter, you retired when you wanted. The game ended when you decided it was over. Meier's design philosophy — give the player a series of interesting decisions, then let them choose how to spend them — runs through every subsequent open-world game.

SimCity (Will Wright, Maxis, 1989) introduced the toy as a legitimate genre. There was no win condition. There was no fail state, really — you could let your city collapse, but the game continued, and you could rebuild. Wright described his design as a software toy rather than a game, drawing on Stewart Brand's Whole Earth Catalog and the cybernetic systems thinking of the time. The simulation-as-toy template runs through The Sims, Cities: Skylines, Animal Crossing, Stardew Valley, Dwarf Fortress, Factorio, and the entire modern cozy-game movement. Wright's deeper claim — that play in a system can teach the system better than any tutorial — is a design philosophy that still rewards study.

Civilization (Sid Meier, MicroProse, 1991) gave us the 4X template — eXplore, eXpand, eXploit, eXterminate — and the concept of a turn-based grand-strategy game playable in a single sitting. Civ's "one more turn" addiction loop is probably the most efficient compulsion engine ever designed in games: each turn is short, each turn produces visible progress, each turn opens new options, and the player always has a reason to take just one more turn before bed. The 4X loop is now the basis for Stellaris, Endless Legend, Old World, Humankind, and a dozen mobile clones.

Dune II: The Building of a Dynasty (Westwood Studios, 1992) is the first proper real-time strategy game in the modern sense: gather resources, build a base, produce units, attack the enemy base. Dune II established the right-click-to-move convention, the resource-and-build economy, and the mission-based campaign structure. Command & Conquer, Warcraft, StarCraft, Age of Empires, and every RTS that followed inherit from Dune II directly. The genre's relative dormancy in the 2010s — most of its biggest studios pivoted to MOBAs — is one of game design history's larger missed opportunities.

Myst (Robyn and Rand Miller, Cyan, 1993) demonstrated that first-person environmental puzzle games could be massive commercial hits. Myst was the best-selling PC game of the 1990s for most of the decade. Its design — pre-rendered first-person scenes, sparse text, no enemies, puzzles solved by manipulating environmental machinery — was the opposite of every contemporary game design assumption. Myst's environmental storytelling (the world tells you what happened here without ever saying it directly) is the foundational text for the entire walking-simulator genre, and modern games like Outer Wilds, The Witness, and Return of the Obra Dinn descend from its lineage.

DOOM (id Software, 1993) is one of the most influential games ever made on multiple axes simultaneously. As a game design, it codified the modern first-person shooter — keyboard-and-mouse controls, pickup weapons, key-card progression, explicit kill counts. As a distribution model, it pioneered shareware: id released the first episode free, and you bought episodes two and three by mailing in a check. As a modding platform, it shipped with deliberately moddable level files, and the WAD-modding community that emerged became the training ground for an entire generation of subsequent designers. (Tim Willits, Tim Sweeney, and many others got their start as DOOM modders.) As a technical artifact, John Carmack's BSP rendering pipeline made 3D-feeling games practical on consumer hardware. DOOM's combined influence on design, business, distribution, community, and technology is probably greater than any other single game in the medium's history.

Magic: The Gathering (Richard Garfield, Wizards of the Coast, 1993) is not a video game but it must be mentioned because it invented the trading card game as a genre and almost every subsequent digital deck-builder owes it a complete debt. Hearthstone, Magic: The Gathering Arena, Slay the Spire, Inscryption, Balatro — all are children of Garfield's design. Magic's deeper innovation is the deck construction phase: you build your deck before play, and the deck-construction is itself a strategic activity equal in depth to the actual matches. The entire roguelike-deckbuilder subgenre that exploded in the 2020s is Magic's late-arriving descendant.

XCOM: UFO Defense (Mythos Games / MicroProse, 1994) gave us tactical turn-based combat with permadeath in a strategic-layer wrapper. Fire Emblem had similar mechanics earlier on console, but XCOM's combination of squad-tactics, base-management, research trees, and the brutal loss of named soldiers created an emotional commitment to your units that no tactics game had previously achieved. Firaxis's 2012 reboot proved the design was timeless.

By the mid-1990s, PC has produced templates for: open-world sandbox, simulation toy, 4X strategy, RTS, environmental puzzle game, FPS, CRPG, and tactics. The console world has produced platformer, action-adventure, JRPG, and Metroidvania. The genre vocabulary that we still use in 2026 was, with few exceptions, set by the early 1990s.

The 3D Revolution (Mid-1990s)

The mid-1990s introduced 3D rendering as a consumer-grade capability, and the entire medium rebuilt itself around the new technology in a span of about four years. The transition is the only design discontinuity in the medium's history that approaches the scale of "everything we knew is now provisional." Every existing genre had to figure out what it meant in 3D, and many of the answers required entirely new design vocabularies — most importantly, the camera.

Virtua Fighter (Yu Suzuki, Sega AM2, 1993) was the first 3D fighting game. The polygonal character models look primitive today, but the design innovation was that the third dimension allowed sidestepping — the ability to move around an opponent rather than only toward and away — and this changed fighting-game tactics permanently. Tekken (Namco, 1994) and the entire 3D fighting genre descend from this.

Mario 64 (Shigeru Miyamoto, Nintendo, 1996) is where 3D platforming was effectively solved on first attempt. Most third-party developers spent the next decade catching up to what Mario 64 figured out about 3D space — and many never did. The key innovations were the analog stick for fine movement control, the dynamic camera with player override (Lakitu the cameraman is itself a design joke about the camera problem), and the hub world with star-based progression. Mario 64's hub-and-spokes structure (Princess Peach's castle leading to themed worlds, each containing collectible stars) was copied by Banjo-Kazooie (1998), Sonic Adventure (1998), Crash Bandicoot (1996, parallel development), and essentially every 3D collect-a-thon platformer of the next decade.

Resident Evil (Shinji Mikami, Capcom, 1996) codified the survival-horror genre. The fixed camera angles (a workaround for the technical limitation that real-time 3D environments looked terrible at the time, repurposed as a deliberate atmospheric choice), the limited inventory, the scarce ammunition, the typewriter save system, the puzzle-and-key progression — all became genre conventions. Resident Evil is also where tank controls were perfected (forward moves you in the direction the character faces, regardless of camera angle), a design solution to the 3D camera problem that aged badly but was rational at the time.

Final Fantasy VII (Hironobu Sakaguchi, Square, 1997) brought the JRPG to mainstream Western awareness on the PlayStation. Square spent unprecedented amounts on pre-rendered cinematics, voice-implied (but mostly text) cinematic scenes, and a CD-based delivery format that allowed the game to be far longer and more visually rich than anything that had come before on cartridge. FF7's cultural impact — the death of Aerith, the iconic opening, the cosplay-friendly character designs — taught the industry that JRPGs could be Western blockbusters and that cinematic spectacle in cutscenes was a marketable design value. The downstream consequences are still being negotiated; some of them are problems we covered in Chapter 23 on cutscenes.

GoldenEye 007 (Rare, 1997) demonstrated that a first-person shooter could work on a console — a heretical claim at the time, when FPS was considered a PC-only genre because it required keyboard-and-mouse precision. GoldenEye's twin-stick proto-control scheme (one stick for movement, the other for aim, mapped to the N64's awkward three-handle controller) was the design ancestor of the modern dual-analog shooter setup that Halo (2001) would canonize. GoldenEye's split-screen multiplayer is also the unsung star of the 1990s console FPS — for a generation of players, "FPS" meant "four friends in a room, shouting at the screen." Modern online multiplayer gradually killed couch co-op, and the loss is real.

Half-Life (Valve, 1998) changed what an FPS could be by eliminating the cutscene. Every story beat in Half-Life happened in real-time, in-engine, with the player in control of Gordon Freeman. NPCs talked to you while you walked; events unfolded around you while you participated; the game's narrative was witnessed rather than interrupted. Half-Life's scripted-storytelling-in-real-time approach is the foundation of every subsequent narrative FPS — Bioshock, Halo, Call of Duty's campaign mode, Titanfall 2. The design philosophy that emerged — the camera does not cut; the player does not lose control; story happens in the world, not on top of it — would later be carried forward by God of War (2018) into the action genre.

Metal Gear Solid (Hideo Kojima, Konami, 1998) brought stealth gameplay into 3D and gave us the modern cinematic action-game template. MGS used long, frequent cutscenes (yes, Kojima invented the cinematic-cutscene-overload pattern that Chapter 23 warned you about), sophisticated voice acting, and a recursive metafictional storytelling style that would influence games as different as Bioshock, NieR: Automata, and The Stanley Parable. The stealth-as-primary-mechanic genre — Splinter Cell, Hitman, Dishonored — owes Kojima a foundational debt.

The Legend of Zelda: Ocarina of Time (Shigeru Miyamoto and Eiji Aonuma, Nintendo, 1998) solved the 3D action-camera problem by introducing Z-targeting: hold the Z button, the camera locks onto an enemy or interactable, and your movement becomes strafe-relative to the target. This single design innovation made 3D melee combat feel as readable as 2D melee combat. Every subsequent 3D action game — Dark Souls, God of War, Bayonetta, Devil May Cry, Sekiro, your progressive project's combat scenes — uses some version of Z-targeting. The deeper structural innovations of Ocarina (the time-travel mechanic, the dungeon item that defines each dungeon's puzzles) also became Zelda series staples for two decades.

By 1999, the medium has acquired a 3D vocabulary. The camera is mostly figured out. Cinematic ambition is mainstream. Every major genre exists in 3D form. The next era will be defined by online connectivity rather than rendering technology.

The MMO and Online Era (Late 1990s–2000s)

Persistent online worlds existed in text-only form much earlier — MUD1 dates to 1978 — but the late 1990s and 2000s saw graphical persistent worlds become a mainstream commercial reality, with all the design innovations and exploitations that the always-online format enabled.

Ultima Online (Origin Systems, 1997) was the first commercially successful graphical MMORPG. Its design was famously chaotic — a fully open PvP world with a simulated economy where players could be murdered, robbed, and have their houses stolen out from under them. Garriott and Raph Koster (the lead designer) had imagined a society that would self-regulate through emergent social pressure. What they got was a society dominated by player killers, with carebear-vs-griefer wars that the team had to address through patches. Ultima Online's design lessons — systems will be exploited, anti-social play is more fun for the antisocial player than pro-social play is for the social player, you must design the social structure into the rules — are still being relearned by every subsequent MMO and live-service game.

EverQuest (Verant Interactive / Sony Online Entertainment, 1999) commercialized the MMO formula at scale. It introduced or canonized the patterns that World of Warcraft would later refine: experience grinding, raid encounters, faction reputation, the "trinity" of tank/healer/DPS group composition, and the time-investment treadmill that would define MMO commitment. EverQuest was famously punishing — death meant corpse runs, missing experience could undo hours of grinding — and the player commitment was correspondingly intense. The community nicknamed it "EverCrack."

Counter-Strike (Minh Le and Jess Cliffe, Valve, 2000) began as a Half-Life mod and became the most important multiplayer FPS template of the next two decades. Its design — 5v5, attack-vs-defend with a bomb objective, round-based with permadeath until next round, weapons purchased with round-earned money — has barely changed since 1.6 and is still the template Valorant (2020) uses with cosmetic refinement. Counter-Strike also established the competitive esports format before "esports" was a word, and the entire modern competitive multiplayer industry descends from it.

EVE Online (CCP, 2003) is the cleanest example of one shared persistent world rather than instanced shards. Every player in EVE plays in the same universe. The game's economy is genuinely player-driven — corporations form, wars are fought, scams are conducted, banks are robbed. The famous Bloodbath of B-R5RB (January 2014) was a 21-hour battle involving over 7,500 players that cost roughly $300,000 in real money in destroyed in-game ships. EVE's design lesson — if the consequences are real to the players, the player drama becomes the content — is still being studied. Most subsequent games have not been brave enough to commit to it at EVE's level.

World of Warcraft (Blizzard, 2004) is the MMO that achieved mainstream cultural penetration, peaking at 12 million subscribers. WoW's design contribution was accessibility refinement: it took every EverQuest pattern and softened it into a form ordinary humans could enjoy. Quests were marked on the map; deaths were forgiving; solo play to the level cap was viable; raids were optional content for committed groups. WoW's deeper innovation was the level-cap endgame — the recognition that the game's actual long-term content was what happened after you reached level 60 (later 70, 80, 85, 90, 100, 110, 120, 60-again with the squish, then 70 again). The endgame raid treadmill, the seasonal content cycles, and the gear-progression rituals are the design DNA of every subsequent live-service game.

Guild Wars 2 (ArenaNet, 2012) attempted to address the MMO loneliness problem with dynamic events — public quests in the open world that any nearby players could participate in without forming groups, with shared rewards and no kill-stealing. The design innovation was making the default social interaction in an MMO actually feel social rather than competitive or transactional. Many subsequent MMOs adopted some version of this.

The MMO era also brought us casual web games, browser-based MMOs (RuneScape, 2001), and the first free-to-play business models that would later metastasize into mobile gaming. The design implications — for monetization, for ethics, for player time — we covered extensively in Chapter 33. The MMO era is where the boundary between "design" and "monetization design" first started to blur.

The Indie Renaissance (2008–Present)

The combination of digital distribution (Steam, Xbox Live Arcade, the iOS App Store), accessible game engines (Unity 2005, GameMaker Studio 2007, Unreal Engine democratization, Godot 2014), and crowdfunding (Kickstarter 2009, Patreon 2013) made independent game development economically viable for small teams and solo developers in a way that had not been possible since the bedroom-coder era of the 1980s. The result is the indie renaissance, which I would argue is the most creatively vital period in the medium's history.

Braid (Jonathan Blow, 2008) is conventionally cited as the moment indie became culturally credible. Braid's time-manipulation puzzles, painterly art style, and oblique narrative made the case that a small team — and Braid was effectively one designer plus an artist — could ship a game that competed with AAA on craft. Braid's commercial success on Xbox Live Arcade demonstrated the digital-distribution business model could support meaningful indie development.

Minecraft (Markus "Notch" Persson, 2011 official 1.0) is the highest-selling game of all time and probably the most culturally consequential game of the 21st century. Minecraft's design innovations were procedural infinite worlds, block-based building as the core mechanic, and no defined goals. The player makes their own purpose. The crafting system — pick up wood, make planks, make sticks, make a pickaxe, mine stone, make a better pickaxe — is so well-designed that the entire crafting genre that exploded after 2011 (Terraria, Don't Starve, Subnautica, Valheim) is essentially in dialogue with Minecraft's tech tree. The deeper innovation, though, is player-as-author. Minecraft is a creative tool that pretends to be a game, and an entire generation of players grew up authoring worlds inside it.

Dark Souls (Hidetaka Miyazaki, FromSoftware, 2011) gets its own case study at the end of this chapter. The short version: Dark Souls revived the philosophy that difficulty is a design value and that the player should be respected enough to be allowed to fail. The Soulslike subgenre that emerged — Hollow Knight, Nioh, Lies of P, Lords of the Fallen, Elden Ring, Sekiro, Salt and Sanctuary, Mortal Shell — is now one of the most commercially viable indie templates available.

Undertale (Toby Fox, 2015) is what happened when one designer with deep knowledge of JRPG history asked what would it look like if I subverted every JRPG convention with intent?. Undertale's pacifist run, its meta-awareness of save mechanics, its character-driven combat where every enemy can be spared, and its emotional core all depend on the player having internalized the conventions the game then breaks. Undertale is the cleanest demonstration that subversion is craft, not a shortcut — you can only subvert what you understand.

Stardew Valley (Eric "ConcernedApe" Barone, 2016) is the proof that one person can ship a game that competes with AAA. Barone spent four years building Stardew Valley alone — code, art, music, design, writing — and the result has sold over 30 million copies and spawned an entire farming-sim renaissance. The design lesson is partly about scope (a focused single-person project can outperform bloated AAA), partly about craft (Barone iterated relentlessly), and partly about the cozy-game genre that Stardew helped catalyze.

Celeste (Maddy Thorson and Noel Berry, Matt Makes Games, 2018) is the gold standard of polished platformer craft and the anchor of much of this textbook. Celeste's design contributions were tight game-feel (hand-tuned air-control, jump buffer, coyote time, dash physics), narrative-mechanical integration (the climb is the story, the story is the climb), and the assist mode — the watershed accessibility innovation that explicitly invites the player to play the game in whatever form makes it enjoyable for them. The assist mode is one of the most important design innovations of the last decade. We will return to it.

Hollow Knight (Team Cherry, 2017) demonstrated that the Metroidvania template still had unexplored territory and that three Australian designers in a garage could outperform most AAA studios on craft. Hollow Knight's interconnected world, deliberate pacing, and willingness to let the player be lost (genuinely lost — without map markers, without quest waypoints) is a counter-argument to the assumption that modern players need constant guidance.

Hades (Supergiant Games, 2020) integrated the roguelike with the action game and the visual novel, then added a procedurally generated narrative — repeated runs deliver story content, character relationships develop across deaths, the death-and-return loop is itself the narrative. Hades won every major game-of-the-year award in 2020 and proved that procedural systems can carry character-driven storytelling.

Vampire Survivors (poncle, 2022) demonstrated that an under-five-dollar indie game with intentionally minimalist visuals could become a cultural phenomenon. Vampire Survivors revived the auto-battler-with-buildcraft subgenre and triggered a wave of imitators. The design lesson — that short, addictive loops with deep build expression can outperform AAA budgets in player-hours-played — is one mobile designers had known for years and the indie scene rediscovered.

The indie renaissance continues. The last fifteen years have probably produced more genuine design innovation per dollar spent than the previous fifty.

Mobile and Free-to-Play (2008–Present)

The launch of the iPhone App Store in July 2008 created a games platform with billions of users, no curation barriers, and a payment system designed for impulse purchases. The design consequences were profound and largely uncomfortable.

Angry Birds (Rovio, 2009) was the breakout hit that proved mobile could host a globally successful game. The design — a slingshot physics puzzle with three-star scoring per level — was simple, the touch controls were native to the platform, and the IP was merchandised aggressively. Angry Birds is the moment mobile gaming entered mainstream awareness.

Candy Crush Saga (King, 2012) is the apex of the mobile match-3 design. Its innovation was the lives system — you have five lives, lives regenerate over time, and when you run out you either wait or pay. This monetization design is the foundational pattern for the entire free-to-play casual mobile market, and many of the design ethics issues we covered in Chapter 33 trace back to it.

Clash of Clans (Supercell, 2012) established the F2P base-builder template. The design — build a base, train troops, raid other players' bases — combined slow asynchronous multiplayer with timer-based progression that converted naturally into purchasable accelerators. Clash of Clans normalized the time-or-money design pattern that had previously been niche.

Pokémon GO (Niantic, 2016) introduced augmented-reality location-based gameplay to a mass market. The design innovation was hybrid digital-physical play: the game required you to walk in the real world to play. The launch summer of 2016 produced more outdoor exercise per teenager than any previous design intervention had achieved. The deeper design lesson is that play context can be expanded — games can leak out of the screen.

Fortnite Battle Royale (Epic Games, 2017) brought the battle royale format to mainstream awareness and demonstrated that AAA-quality F2P at console-grade fidelity was viable. Fortnite's cosmetic-only monetization (no pay-to-win) was a deliberate design choice that became influential. Fortnite also established the seasonal narrative — limited-time events, world changes, in-game concerts — that turned the game into a persistent cultural space rather than a discrete product.

Genshin Impact (miHoYo, 2020) brought gacha monetization to console and PC at a fidelity level previously associated only with single-player AAA. Genshin's open world, anime characters, and gacha mechanic for unlocking new characters generated billions of dollars in revenue and signaled that the gacha business model had matured into a mainstream design pattern. Whether this is good for game design is a question Chapter 33 took up at length.

The mobile and F2P era is the part of the medium's history where the financial incentive structures most aggressively shaped design choices. Many of the patterns invented here — daily login bonuses, energy systems, time-gated progression, gacha pulls, FOMO event windows — are now showing up in console games. This is one of the design conversations that will define the next decade.

Key Design Innovations (Cross-Era, Thematic)

Some innovations deserve recognition independent of the era they emerged in. This section catalogs ten that you should know by name.

1. The Save System. Adventure (Atari 2600, Warren Robinett, 1979) had no save — finishing it required a single sitting. The Legend of Zelda (1987) introduced battery-backed cartridge save on a home console. Final Fantasy (1987) introduced the save point as a designed location in the world. Resident Evil (1996) made the save point a limited resource via the typewriter ribbon. Dark Souls (2011) made the save point (the bonfire) a ritual location that resets the world's enemies. Celeste (2018) made the save invisible and continuous — the game never explicitly saves, it just always does. The history of save systems is the history of a single design problem (when does the player keep their progress?) being repeatedly solved with different intentions.

2. The Quicksave / Quickload. Popularized by DOOM (1993) and refined by the Half-Life and Thief eras, the quicksave enabled an entire style of play — try, fail, reload, try again — that became the default for PC immersive sims. The design tension is real: quicksave makes the player invincible to consequence, which can deflate tension. Hitman: Blood Money (2006) restricted quicksaves per level to balance tension and convenience. XCOM 2 shipped Ironman mode for the same reason. The quicksave is a power players want and that designers must decide when to grant.

3. The Objective Marker. Deus Ex (2000) and GTA III (2001) popularized the on-screen waypoint or compass arrow that tells the player where to go next. The design tension here is sharp. Markers reduce friction but also reduce attention to the world — the player following the arrow is not exploring, they are commuting. Breath of the Wild (2017) deliberately removed most markers and trusted the player to navigate by landmark. The objective marker is one of the cleanest examples of a quality-of-life feature that has costs you have to weigh.

4. The Skill Tree. Tabletop RPGs had skill trees in the 1970s. Diablo II (2000) popularized the visual branching skill tree in digital form. The Witcher 3, Borderlands, Path of Exile, Final Fantasy X's Sphere Grid, and a thousand other games refined it. The skill tree's design value is twofold: it provides a visualization of progression options and it forces the player to commit to a build identity. The failure mode is the false choice tree — a tree where one path is mathematically optimal, so the choice is illusory. Modern designers should ensure that branches are differentiated by playstyle rather than power.

5. Achievements. Xbox Live (2005, Xbox 360 launch) introduced platform-wide achievements as a meta-layer atop every game. The design innovation was external recognition for play — the achievement persisted on your gamer profile and was visible to friends. Achievements have become contested. Some players love them; others find them an anti-design force that pushes designers toward checklist-style content. Yakuza 0's Completion List is the most aggressive use of achievements as content; Hades's achievements are gentler nudges toward unexplored systems. Both are valid; the design choice is whether your achievements describe play or prescribe it.

6. Crafting. Crafting existed in Ultima and EverQuest but was made dominant by Minecraft (2011). Its design value is that crafting binds resource gathering to player goals — the wood you collect is not abstract XP, it is the wood you need to make the table you need to make the workbench. Terraria, Don't Starve, Valheim, Subnautica, No Man's Sky, Astroneer, and the entire survival-crafting subgenre share Minecraft's DNA. The genre's commercial dominance after 2013 has been remarkable.

7. Procedural Generation. Rogue (Toy and Wichman, 1980) was the first roguelike — randomly generated dungeons every run, permadeath. Diablo (1996) brought procedural levels to mainstream action RPG. Spelunky (Yu, 2008) modernized the roguelike for the indie era and became the design template that The Binding of Isaac, Enter the Gungeon, Dead Cells, and Hades all build on. No Man's Sky (2016) proved procedural generation could fill galaxies. The design value of procedural generation is high replayability per development hour; the design risk is blandness — procedural content tends toward the average, and good designers use procedural systems with hand-authored constraints.

8. Assist Modes. Celeste (2018) is the watershed. Maddy Thorson explicitly framed the assist mode not as an "easy mode" (which carries connotations of inferior play) but as invitation to play the game in whatever form makes it joyful for you. Assist mode lets you slow the game speed, make yourself invincible, give yourself infinite stamina, skip chapters. Hades has God Mode (a stacking damage-reduction buff each death). Cuphead added a Simple mode. The Soulslike community has been more contested — Sekiro's lack of difficulty options sparked years of debate — but the trajectory is clear: accessibility options are now a baseline expectation for narrative-and-craft-focused games. We covered the design and ethics of this in Chapters 11 and 33.

9. Live Service / Games-as-Service. World of Warcraft (2004) was the first major commercial demonstration that a game could be a continuous service. Destiny (2014), Fortnite (2017), Genshin Impact (2020), and Marvel Rivals (2024) made the model dominant in AAA. The design implications are profound: games are no longer "finished" at launch; they are launched, then maintained, then updated, then end-of-life'd. The skill set required to design a live game is different from the skill set required to design a finished game, and the player relationship with a live game is different — the player commits not to a product but to a relationship.

10. Battle Royale. The H1Z1 mod scene (2015–2016) and PUBG (PlayerUnknown's Battlegrounds, 2017) established the format: 100 players parachute onto a shrinking map, last player or squad standing wins. Fortnite Battle Royale (2017) brought it to a billion-player audience by attaching it to a free-to-play cosmetic-monetization model and a vibrant aesthetic. Apex Legends (2019) added hero abilities. Warzone (2020) added the AAA shooter feel. The battle royale is one of the few genuinely new genres of the 2010s and demonstrates that new genres can still emerge — the medium is not closed.

🎯 Tradeoff Spotlight: Every innovation in this catalog is a solution to a problem the designer chose to take seriously. The save system solves the problem of session length. Achievements solve the problem of meta-progression. Procedural generation solves the problem of replayability. Assist modes solve the problem of accessibility. Before adopting any of these patterns in your own game, identify which problem you are solving. Adopting a pattern that solves a problem you do not have is how design becomes pastiche.

Designers to Know

A non-comprehensive list of designers whose work you should be familiar with as a baseline of professional fluency. This is not a "greatest" list. It is a "you should at least know who they are" list. One or two sentences each.

Shigeru Miyamoto (Nintendo) — Donkey Kong, Mario, Zelda, Pikmin. The most influential designer in the medium's history. His "playground first, story second" philosophy shaped Nintendo and through Nintendo most of console gaming.

Hideo Kojima (Konami, Kojima Productions) — Metal Gear series, Death Stranding. Auteur designer who made cinematic ambition central to AAA design. His maximalism is divisive but his influence is total.

Sid Meier (MicroProse, Firaxis) — Pirates!, Civilization, Alpha Centauri. Defined the modern strategy game and articulated the interesting decisions design philosophy that runs through most strategy design.

Will Wright (Maxis) — SimCity, The Sims, Spore. Pioneer of simulation-as-toy. His software toy framing reframed what games could be when they had no win condition.

Hidetaka Miyazaki (FromSoftware) — Demon's Souls, Dark Souls, Bloodborne, Sekiro, Elden Ring. Revived difficulty as design value and proved that a hard game with no compromises could be a mass-market success.

Yoko Taro (Square Enix) — Drakengard, NieR: Automata. Demonstrated that games could carry serious philosophical weight via mechanical metafiction, including endings that delete your save file.

Toby Fox — Undertale, Deltarune. One-person studio that proved subversion of JRPG conventions could carry both humor and emotional weight at AAA-comparable craft levels.

Maddy Thorson and Noel Berry (Extremely OK Games, formerly Matt Makes Games) — Celeste, Towerfall. The contemporary high-water mark of indie platformer craft and accessibility design.

Jonathan Blow (Number None / Thekla) — Braid, The Witness. Proved indie could match AAA in craft. Public proselytizer of games as serious cultural objects.

Eric "ConcernedApe" Barone — Stardew Valley. The proof of concept that one developer can ship a category-leading game.

Lucas Pope (3909 LLC) — Papers, Please, Return of the Obra Dinn. Maker of small, idiosyncratic, design-perfect games. Required playing.

Tim Cain (Black Isle, Obsidian) — Fallout, Arcanum, The Outer Worlds. Defined the modern Western RPG.

Warren Spector (Origin, Ion Storm, OtherSide) — Ultima Underworld, Deus Ex. Father of the immersive sim, one of the medium's most distinctive design philosophies.

Ken Levine (Looking Glass, Irrational, Ghost Story) — System Shock 2, Bioshock, Bioshock Infinite. Refined narrative ambition in the FPS form.

Amy Hennig (Naughty Dog, Skydance New Media) — Uncharted series. Defined the modern cinematic action-adventure narrative voice.

Kim Swift (Valve, Airtight, Stadia) — Portal. Co-creator of the most teaching-perfect game in the medium and one of the design world's strongest voices.

Rhianna Pratchett — Tomb Raider (2013), Mirror's Edge, Heavenly Sword. Game writer who has done as much as anyone to elevate writing as a discipline within game design.

Brenda Romero (Atari, Sir-Tech, Romero Games) — Wizardry series, Train. Tabletop and video game designer whose work spans commercial AAA and serious-games research.

Roberta Williams (Sierra On-Line) — King's Quest series, Phantasmagoria. Co-founder of Sierra; pioneered the graphic adventure game and helped invent the entire interactive-narrative form.

Hironobu Sakaguchi (Square, Mistwalker) — Final Fantasy series. Defined the JRPG.

Markus "Notch" Persson — Minecraft. Created the highest-selling game of all time, then sold it. His subsequent public conduct is its own subject; his design contribution is undeniable.

This list is incomplete. It is missing major Chinese, Korean, and Indian designers whose work has not been adequately surfaced in Anglophone game studies. It is missing several women and non-binary designers whose contributions have been historically erased. Treat it as a starting point. The further reading section will help you fill the gaps.

Progressive Project Update

For your action-adventure project, identify three specific historical innovations the game depends on. Name the games and designers you are inheriting from. Write a paragraph on each: what problem the innovation solved, when it was first introduced or perfected, how your game uses it, and what (if anything) you are modifying about the original pattern.

Example commitment (yours will vary):

Innovation 1: Gated progression via ability acquisition (Metroid, 1986). Yoshio Sakamoto and Hiroji Kiyotake solved the problem of how to make a non-linear world feel structured by gating regions behind movement abilities the player has not yet earned. My game uses this pattern directly: the wall-jump is gated until the second area, the dash is gated until the third, and a backtracking pass through the first area unlocks new content. I am not modifying the pattern significantly. I am hewing close to the source because the pattern works and players understand it.

Innovation 2: The bonfire / persistent save with world-reset (Dark Souls, 2011). Hidetaka Miyazaki solved the problem of how to make a death have weight without making it punishing. The bonfire saves the game, restores resources, and resets enemies — death sends you back to the last bonfire with your in-progress soul-currency at risk of permanent loss if you die again before retrieving it. My game uses a simplified version: shrines save, restore, and reset enemies, but I do not have a soul-currency-at-risk system. I am modifying the pattern toward gentler tension because my game is a 6-hour experience for newer players, not a 60-hour test for veterans.

Innovation 3: Assist mode (Celeste, 2018). Maddy Thorson solved the problem of how to make a craft-focused difficulty game accessible without compromising the design's intent for players who want the standard experience. The assist mode lets the player adjust game speed, invincibility, dash count, stamina, and chapter skipping. My game adopts the pattern with three sliders: enemy damage (50%–200%), platforming difficulty (off / standard / hard), and infinite-stamina toggle. I am modifying the pattern by not including a chapter-skip option, because my game is short enough that I want every player to traverse the full arc.

Write your three. Keep them in your design document. When someone asks "what genre is your game?" you will be ready to answer with something more specific than "action-adventure." You will be able to say: Metroidvania structure, Souls-like checkpointing, Celeste-influenced accessibility. That sentence locates your game in the conversation. It is a more honest answer than "it's like Hollow Knight."

Common Pitfalls

1. Claiming novelty without doing the research. "I invented this mechanic" is almost always wrong, and the designer who says it loudly will be corrected loudly. Before you claim something is original, search for it. Check itch.io, MobyGames, the relevant subreddit, the Internet Archive's MAME library. Most of the time, a cleaner version of your idea was shipped in 1996. This is good news, not bad — you can study the prior art and improve on it.

2. Pastiche without understanding. Building a Soulslike because Soulslikes are popular, without understanding why the Souls design philosophy works, produces a checklist of features (bonfires, stamina bar, weighty animation, opaque lore) without the underlying design coherence (difficulty as respect, world as character, death as teacher). The features without the philosophy is the most common indie-Soulslike failure.

3. Ignoring non-Western traditions. The history above is heavily Anglophone-and-Japanese-centric, which reflects what is well-documented in English. The actual history is broader. Korean MMOs (NCsoft's Lineage, 1998) shaped the entire MMO economy. Chinese mobile design (NetEase, miHoYo) is now setting global trends. Eastern European studios (CD Projekt Red, GSC Game World, 4A Games) have produced some of the medium's most distinctive AAA work. Indian indie scene is growing rapidly. Bahasa and Spanish-language indie scenes are vibrant. If your design canon is only American and Japanese, you are missing most of the conversation.

4. Treating design history as linear progress. It is not. Many "old" games solve problems "new" games still struggle with. Spelunky (2008) is in many ways more elegant than its imitators. Pong is not less designed than Genshin Impact; it is differently designed for a different problem. The temptation to think of game design as progress (more polygons, more systems, more complexity) is a category error. Design quality is orthogonal to technological era.

5. Erasing the labor of unnamed contributors. The designers I named above are the ones who got credit. Each of them worked with teams of programmers, artists, testers, and producers whose contributions to "their" innovations were enormous and usually uncredited. When you study a game, study the credits. Look for the leads who got named, then look for the team members who actually built the systems. Game design is collective labor. The auteur narrative is sometimes useful and often misleading.

6. Mistaking commercial success for design quality. Candy Crush Saga is one of the most commercially successful games ever made. Cataclysm: Dark Days Ahead is a free open-source roguelike with a fraction of the audience. Both are well-designed for their goals. The lesson is not that money equals design — sometimes the money tracks the design and sometimes it tracks the marketing. Read commercial success as a signal, not a verdict.

Summary

Games are in conversation with every game that came before them. The conversation is rich, the lineage is long, and the cleverest moves still being made today are usually composed of patterns that have been around for decades. Your job as a designer is not to escape the lineage. It is to join it — to know what conversation you are entering, to know who has spoken before, and to add something specific and your own.

The history is not a list of triumphs. It is a list of solutions to specific design problems. Space Invaders solved difficulty escalation. Mario Bros. solved teaching by play. Metroid solved non-linear world structure. Final Fantasy solved long-form character-driven narrative on consumer hardware. Counter-Strike solved competitive balance via economy and round structure. Dark Souls solved difficulty as a design value. Celeste solved the accessibility-versus-intent dilemma. Every one of those solutions is available to you. Most of them are documented. All of them can be modified, recombined, or used directly. There is no copyright on design patterns. There is only the discipline of knowing what you are using and why.

The next chapter is on scope — the discipline of building what you can finish rather than what you can imagine. The history above is full of cautionary tales about scope: No Man's Sky's rough launch, Daikatana's collapse, Anthem's troubled production, Cyberpunk 2077's overpromising. The designer who knows the history knows what not to build, as much as what to build. The next ten chapters of this textbook take that knowledge and turn it into the practical work of finishing a game and putting it in front of players.

You are part of this lineage now. Welcome.