Game atoms, mechanics representation, micro-mechanics and emergence
BSc (Hons) Game Design & Development
Describe it in three words, without mentioning anything you see.
actions, resources, goals, rules and feedback.
atoms → mechanics → dynamics → aesthetics, and find where the chain breaks.
separate the system from its surface, and use latency, forgiveness and polish on purpose.
with loops and simple rules that combine into play you didn't script.
The five types, and how to find them
Pac-Man in five steps
Feel, forgiveness, combos, mastery
Same code, different feel; schemas
Complexity from simple rules
Atom audit, then design a rule
Mechanics produce dynamics that players feel as aesthetics. Designers build M→A; players meet A first.
One coin, five layers of feedback. Nested loops: seconds inside minutes inside weeks.
Affordances and learned conventions; Norman's mapping: controls that match results.
What is a mechanic made of? Why does the same mechanic feel different in two games? And how do small mechanics add up to play nobody wrote?
The building blocks of gameplay
A game atom is the smallest, indivisible unit of meaningful interaction. Split it further and it stops meaning anything to the player.
GDnD wiki · Game Atoms and Micro-MechanicsMove, jump, attack, build, vote.
Dark Souls: dodge, parryHealth, ammo, gold, time, trust.
Hades: Obols, DarknessWin conditions and objectives.
Among Us: finish tasksConstraints and cause-and-effect.
Fortnite: the storm closes inSound, animation, numbers, rumble.
BotW: the weapon-break shatterEvery mechanic is a structured interaction between atoms. Change one atom and you change the mechanic.
List the actions, resources, goals, rules and feedback, then ask how each pair interacts.
Draw how actions lead to outcomes and how resources flow in and out.
Track the sequence of decisions a player makes, and the atoms used at each stage.
Play with a notebook. Change your behaviour on purpose and watch what the rules do.
Mapping Pac-Man, from a joystick to a feeling
Steps 2–4 are Week 2's MDA. Step 1 is what's new: MDA tells you the layers; atoms tell you what the bottom layer is made of.
GDnD wiki · MDA FrameworkAction · moveResource · pelletsResource · power pelletsResource · livesGoal · clear the mazeRule · ghost touch = lose a lifeRule · wallsFeedback · wakka sound, score, blue ghosts
| Mechanic | Atoms | What happens |
|---|---|---|
| Movement | movewallsanimation | Steer through the maze; walls block you |
| Collection | movepelletsound + score | Moving over a pellet eats it |
| Avoidance | moveghost = deathsurvive | Stay away from the ghosts |
| Power-up | power pelletghosts vulnerableblue ghosts | For a few seconds, the hunted hunts |
| Life / failure | livesloss on contactdeath jingle | Lose them all and the game ends |
| Dynamic | Emerges from |
|---|---|
| Risk–reward | Collection + avoidance: chase the pellets near a ghost? |
| Strategic timing | Power-up + avoidance: save the power pellet until the ghosts close in |
| Spatial awareness | Movement + maze: plan safe routes, predict the ghosts |
| Mechanic / dynamic | Aesthetic |
|---|---|
| Collecting pellets | Satisfaction, rhythm, flow |
| Avoiding ghosts | Tension, fear, excitement |
| Power pellets | Empowerment, mastery |
| Clearing a maze | Achievement, relief |
"Save the power pellet" is written nowhere in the code. Players discover it.
No risk–reward: the aesthetic is boredom. Fix a rule (ghost speed) or the maze.
No timing dynamic: confusion. Fix a feedback atom.
The rhythm goes: flat. Same rules, lost aesthetic.
The feel of play: split-second exchanges between input and feedback
Macro-mechanics are the systems: combat, traversal, crafting. Micro-mechanics are the feel: the split-second loop of input and response.
Game feel is real-time control of virtual objects in a simulated space, with interactions emphasised by polish. (Swink, 2009)
Together: a rhythmic, tactile flow that rewards precision.
Try jumping late off the edge. Then try pressing jump just before you land.
Something visibly changes within 100 ms of the press. Every frame of lag reads as weight or mush.
The same input always does the same thing. That's how muscle memory forms.
Anticipation and follow-through, hit-pause and shake (sparingly), layered sound.
Coyote time: jump a moment after leaving the ledge. Input buffering: a press just before you can act is remembered. Both make the game match what players meant, not what they did.
Each action triggers the next, through timing or logic.
Inputs, timing and feedback combine into a flowing sequence.
Visuals, sound and score tell you (and everyone watching) it worked.
D C B A S SS SSS
Dope → SSStylish. The meter rises for variety and pressure, and drops if you repeat the same move or get hit. The rank is the mastery, made visible.
Celeste gives each chapter one new element (dream blocks in Chapter 2) and remixes it with everything before.
A plateau is where players get bored, or where they practise for combos. Know which one you built.
Every mechanic has two realities
| System layer | Representation layer | |
|---|---|---|
| What it is | The logic: physics, timers, collision boxes, algorithms. Abstract, mathematical. | What the senses get: visuals, sound, motion, rumble, UI effects. |
| A jump | vy = 12 m/s on press; gravity −30 m/s², doubled while falling | A crouch, a stretch upward, a camera tilt, a whoosh, dust on landing |
| Where problems hide | Wrong numbers: too floaty, too slow | Wrong timing or sound: right numbers that feel wrong |
Together these layers translate code into felt experience.
| Celeste (2018) | Hollow Knight (2017) | |
|---|---|---|
| System | Gravity, variable jump height, a dash | Gravity, variable jump height, a dash |
| Representation | Fast animation, bright palette, punchy click, no lag | Slower wind-up, dark palette, heavy, muffled sound |
| Feels | Light, precise, empowering | Weighty, deliberate, sombre |
| Emotional tone | Urgency, agility, optimism | Resilience, melancholy, endurance |
The representation layer carries the emotion, and it should match the theme (Week 4).
GDnD wiki · CelesteThere is one character, one set of numbers, one input. Only the drawing, the particles and the shake differ. Which one feels heavier?
Knowledge that guides expectation. Red barrels explode; hearts heal.
A cue that suggests an action. A handle says "open".
Brains reuse schemas to save effort: ⚙ means settings, everywhere.
A door with a handle that never opens: broken trust, then frustration.
Dark Souls' mimics: a treasure chest that bites. Once. Then you check every chest.
Complexity from simplicity
Action → system response → feedback → player understanding → next action. Every pass teaches a little.
Predictable outcomes build trust; small variations keep it interesting.
When several loops interact, behaviour appears that no single loop contains. That's emergence.
Mine a block → get a resource → wear the tool down.
Craft better tools → build → explore further → find new materials.
Players build whole cities, working calculators and computers out of redstone, and share them.
Week 3 asked whether small loops feed big ones. Today: what happens when loops feed each other?
GDnD wiki · MinecraftBlinky chases you; Pinky aims 4 tiles ahead; Inky uses Blinky's position; Clyde backs off when close. It looks like teamwork.
Fire spreads and makes updrafts; metal conducts; wind moves things. Players invent solutions.
Hunger, hygiene, social, all decaying at once, add up to life stories.
Economy, war and culture collide into a different history every game.
Players become co-authors: they own what they discover.
GDnD wiki · EmergenceA few rules that combine into huge variation: chess, Minecraft, BotW's chemistry. Replayable; hard to predict and balance.
A designer-set sequence of challenges: adventure games, Portal. Controlled; one way through, eventually exhausted.
Recipe for emergence: clear local rules · systems that act on each other · more than one valid solution · test the possibility space, not just the intended path.
Add one rule to Pac-Man that would create a new dynamic. What might players discover?
2:00Five atoms: actions, resources, goals, rules, feedback. Mechanics are atoms in combination.
Trace the chain atom → mechanic → dynamic → aesthetic, and walk back down it when the feeling's missing.
Feel lives in milliseconds: under 100 ms response, forgiveness you never see, polish used sparingly.
Same system, different surface: the representation layer carries the emotion.
Write rules, not events: interacting loops produce emergence, and players co-author it.
Part A: audit a game. Part B: design a rule and test it.
Minecraft · Breath of the Wild · Fortnite · Dark Souls · League of Legends · Stardew Valley · Hades · The Witcher 3 · Among Us · Super Mario Odyssey
Worked example: Pac-Man, slides 13–15.
Present in one minute: your chain, and the one atom you'd change to make it a different game.
| Atom | In this version |
|---|---|
| Action | Place a mark |
| Resource | Three marks each, no more |
| Goal | Three in a row |
| New rule | Placing a 4th mark removes your oldest |
| Feedback | Your oldest mark fades: it's next to go |
| Dynamics | No draws. You plan two moves ahead: what will vanish as well as what you place. |
| Aesthetics | Tension and real challenge, from a game we'd "solved" at age seven. |
Take a simple game (or your own idea). Write its rules on the rule card: actions, resources, goals, rules, feedback. Add or change one rule.
Paper, dice, coins, cards. Play it at least twice in your group.
Clear? Consistent? Can it be exploited? What dynamic appeared? Was it the one you predicted?
Present (2 min): the rule, the dynamic you predicted, the dynamic you got, and one change you'd make next.
“The atom I'd change in my favourite game is …, and it would make players ….”
One sentence each, before you leave.
Bartlett, F. C. (1932) Remembering: A Study in Experimental and Social Psychology. Cambridge: Cambridge University Press.
Cook, D. (2007) 'The chemistry of game design', Gamasutra, July.
Gardner, M. (1970) 'Mathematical games: the fantastic combinations of John Conway's new solitaire game "life"', Scientific American, 223(4), pp. 120–123.
Hunicke, R., LeBlanc, M. and Zubek, R. (2004) 'MDA: a formal approach to game design and game research', AAAI Workshop on Challenges in Game AI.
Juul, J. (2002) 'The open and the closed: games of emergence and games of progression', in Mäyrä, F. (ed.) Computer Games and Digital Cultures Conference Proceedings. Tampere: Tampere University Press, pp. 323–329.
Pittman, J. (2009) 'The Pac-Man dossier', Gamasutra, February.
Skinner, B. F. (1953) Science and Human Behavior. New York: Macmillan.
Swink, S. (2009) Game Feel: A Game Designer's Guide to Virtual Sensation. Burlington, MA: Morgan Kaufmann.