Puzzle Design

Ten Principles for Good Puzzles §

• From Schell
• Make the goal easier to understand. Unless the figuring out what to do is fun, a confusing puzzle is not fun. View the puzzle in the lens of goals.
• Make it easy to get started. Make it toy-like ,easy to manipulate.
• Give a sense of progress. A puzzle demands an answer but involves manipulating something so you can feel yourself getting closer to the solution
• Give a sense of solvability. Convince the player they can solve the puzzle. Convince them there is a solution.
• Increase difficulty gradually. Give player control over the order of steps. Make the steps difficult to perform
• Parallelism lets the player rest - Parallelism gives the player the opportunity to do something easier.
• Pyramid structures extend interest. - Pyramid
• Hints extend interest - A well-timed hint, especially for a difficult part of the puzzle can re-ignite player’s interest.
• Give the answer. The Eureka moments are triggered by seeing the answer to the puzzle which allows us to reexamine the puzzle and how it hints at the answer. Additionally, if players are stumped, consider giving them the answer.
• Perceptual Shifts are a Double Edged Sword. - problems where you either get it or don’t are problematic. When the player can shift their perspective, the puzzle is pleasurable. Otherwise, they get nothing but frustration.

Grant’s Puzzle Design Lessons §

• From Elyot Grant - 30 Puzzle Design Lessons

• Eureka moments are the atoms of puzzles. A Puzzle is anything that conceals a Eureka moment.

  • A Eureka moment is characterized by a sudden, pleasureful, fluent, confident feeling of understanding.
  • A puzzle designer identifies powerful eureka moments, creating maximally impactful setups for them.

• Puzzles are isomorphic to humor.

  • Eureka moments induce positive affect and reduce stress. A powerful Eureka moment motivates the player to keep playing.
  • This is similar to humor.

• Maximize the amount of sparkle. The sparkle pertains to how good the Eureka moment is. We need lively answers and clever clues.

• Avoid the chaff. Avoid the fill or filler stuff that no one likes.

  • Clues that feels forced and natural or unintuitive.
  • Clues or fill that are meaningless.
  • Clues that are too obscure or reliant on very specific knowledge

• Surprise is the key for sparkle. The solution should be unexpected — the last thing that the solver expects.

• Interesting truths are the root of surprise. The puzzle sets discovering the truth as its challenge. Its value then becomes the ability to convey the truth to the solver.

  • Generally more unexpected = more dopaminergic. The human brain has evolved to find novel and unexpected things interesting to remember them strongly and to seek more of them.
  • Puzzle designers exploit the natural human instinct to discover.

• Eureka is not Fiero. The pleasure is not necessarily in overcoming the challenge of the game, it could simply be about discovering the interesting thing about the game.

  • The Fiero approach is to create a challenging puzzle. The player will feel smart after they, themselves overcome the challenge.
  • We can synergistically combine Fiero and Eureka.
  • Only puzzle can confer Eureka. Puzzles are not solely for padding or obstacles in games. They must communicate something of actual substance. Beautiful puzzles convey something interesting.
    • Many people solve puzzles for different reasons. It is important to understand the target audience and their reasons for playing.

• Eureka is shareable.

  • The Eureka moment is not contingent on players solving the puzzle. They can simply be shown the solution and have the same moment.
  • Eureka are not spoilable.

• Create many Eureka moments. Especially useful for escape rooms or sequential puzzles.

• Create a rewarding ending.

  • Caution: The rewarding ending should not make the puzzle trivial to solve.
  • Extraction steps that setup metapuzzles can create more Eureka moments.

• Embed secrets and surprises Secrets or easter eggs that could be missed unless the solver takes a closer look.

  • Even if they may be missable, they provide additional surprise to the solver and encourages them to look more closely for more secrets.
  • They also provide reasons to go back and comb through the game, even after it is finished

• Aporia is a promise of surprise.

  • Aporia is a state of puzzlement where the solver believes that the puzzle is impossible.
  • The obvious things should lead to non-solutions. Promise that there is an interesting trick to the puzzle.
  • In this state, the user anticipates that there will be a big Eureka moment incoming.

• Cultivate trust in the solver

  • Make sure that the solver doesn’t think that they are missing an item to solve the puzzle or worse, that the game has a bug that makes the puzzle unsolvable.
  • One way to cultivate trust is to tell the player that they have all they need to solve the puzzle.
  • Another idea is square dealing - abide by conventions to play fair. But more importantly ,communicate these conventions to the player.
    • Example: no outside knowledge required for an escape room.
  • Avoid cheap tricks. No “Oh really?” moments.
  • No poisoning the well — an unhealthy skepticism in the player’s minds that the designers aren’t playing fair. Avoid the following
    • Unclued searching — invisible platforms, wall pressing, darkness. Make sure that the player cannot start the puzzle without the clue
    • Sudden increase in tenuousness. Do not force the player to do something that they normally didn’t have to do.
    • Sudden or unclued use of new trick or method
    • Inconsistent rules

• Be careful with intransparencies

  • These are aspects of the puzzle that are not explained to the solver —
    • Hidden goals
    • Hidden rules
    • Hidden affordances, interactions or consequences
    • Inability to perceive or interpret the puzzle state.
    • Instructionless puzzles (unless the goal is rule discovery).
  • Intransparencies can be fun but
    • They can cause solvers to become stuck (the unfun kind of stuck)
    • They are Bad for linear games
    • They can feel like an IQ test It feels simply like pattern recognition.
    • They are one-time reveals
    • They are easily spoiled
    • They are terrible for an audience or spectation. The audience may spoil it or the player may have to keep explaining the rules.

• Facilitate experimentation.

  • Intransparencies can be made transparent by experimentation
  • Avoid making experimentation frustrating. Reduce the amount of effort to simply start an experiment.
  • Things that can make experimentation tedious: No undo buttons. Unskippable cutscenes.
  • One possible mitigation: Show the mappings between game objects so that the player knows what certain objects do to other objects (which doors does the button open).

• Get the solver on the right track.

  • Confirm that the solver is on the right track.
  • Realizing you’re on the right track is dopaminergic
  • Be cautious with red herrings
    • Dressings - needless, non-pertinent decorations that are salient enough to be interpreted as pertinent. Make sure to highlight them as simply dressing.
    • Distractors - affordance that are not needed.
    • Fake puzzles - what seem to be puzzles are not actually puzzles. Avoid these.
    • Ghost puzzles - artifacts of old puzzles found in the new puzzle that may be misleading. Avoid these.
    • Obfuscation and misdirection - not necessarily bad especially when they highlight why a solution cannot work.
    • Dead end approaches - approaches that do not work and it can be immediately seen why. Can provide Eureka moments as well.
    • Rabbit Trails - incorrect approaches to solving a puzzle that cannot be proven not to work with reasonable effort. It is not clear if the solution is a dead end.
      • Avoid these because they can be demotivating, especially for novices who do not have a good puzzle sense.
      • Some ways to mitigate this:
        • Seal off rabbit trails entirely
        • Add negative confirmers
        • Playtest.
  • Match pertinence with salience. Things that are important must stick out. Things that stick out must be important.

• Seek depth in systems and rulesets.

  • This approach has several benefits
    • A better conceptual model
    • Knowledge is reusable. Puzzles can build off of any emergent lemmas that arise from the ruleset ¹
    • Consistency and depth
    • Emergent surprises.
    • Better Eureka moments
    • Lowe cost of content creation
  • A useful thing to think about is the capacity of a puzzle’s ruleset -
    • How many puzzles can you create before you start repeating the same tricks over and over.
    • How many of the puzzle type would you want to do in real life ?
    • When you only reach the capacity of the current ruleset, you must expand and add new components to add more capacity
    • Another way: Create variants and hybrids of variants.
      • Change the topology or space of the puzzle
      • Add new constraints
      • Add new clue types
      • Blend two puzzle types
      • Liar (some clues are wrong) , Cipher (some clues are obfuscated), Off-by-one (clues are off by one) clues.
  • The depth of a game has many sources:
    • Inherently deep systems with simple rules.
    • Contrivance
    • Emergent interactions
    • Level architecture. This can add more depth without adding new contrivances.
  • Common puzzle tropes
    • Paths and Connectivity
    • Packing or covering problems
    • Perspective or line of sight
    • Logic gates
    • State manipulation.

• Understand the puzzle’s state space.

  • Solving a puzzle involves navigating its state space.
  • In this sense, creating a puzzle is similar to pruning edges in the state tree until we get an interesting maze.
  • We can add difficulty by having state spaces that
    • Are larger
    • Have bottlenecks that indicate a specific trick for solving
    • Have multiple bottlenecks. This makes the puzzle even more difficult.

• Understand your puzzle’s solution path
  • Understand how the player makes progress in the puzzle.
  • State spaces or solution paths that have high breadth to allow for more exploration. Similar to this is having multiple break-ins for the puzzle to make some progress.
  • Linear solution paths are much more difficult as they require a single break-in to solve but they can also lead to an interesting solution flow.
  • Aim to have Gem and Flow type puzzles (see below).

	Few steps	Many steps
Low difficulty	Trivial	Gem
High Difficulty	Flow	Slog

• Speak playfully to the solver.

  • Some way to do this:
    • There should be a sense of playfulness in solving the puzzle
    • Some form of pattern or motif apparent throughout the puzzle
    • Repeated use of a lemma or theorem
    • Chain repetition
    • Variants or twists on the lemma
    • There is setup and payoff
    • Clever surprise or ending.
  • Take the solver on a journey throughout the solution path.

• Design from the front.

  • Advice for writing Plots is surprisingly useful, especially when you think of the solution path as a journey.
  • When designing, simultaneously flip between the perspective of the designer and the perspective of the player. Then do a final solve of the puzzle. Repeat this as many times as needed, as if prototyping.
  • Being good at solving the puzzle helps because you know what can and cannot be deduced from the current setup so far.
  • Some benefits:
    • Controls the shape of the solution path
    • Controls difficulty
    • Controls flow
    • Create playful setups
    • Hide eureka moments
    • Create a deductive solution rather than one that’s reliant on brute force
    • Enforces uniqueness to the solution path

• Make the key move of the puzzle provably unique. The key move is the only way to satisfy the constraints of the puzzle or to break-in.

  • This forces the solver to uncover the eureka moment.
  • It is undesirable if the player can find an exploit around the intended solution
  • Deductive reasoning helps. Being able to prove that the key move is unique may require mathematical knowledge.
  • A deductive flow is not possible if the solution was non-unique.

• Add a theme. Some benefits include

  • Makes it easier to remember the puzzle elements by association
  • Adds aesthetic appeal to the puzzle.
  • Additional sparkle
  • More memorable and identifiable puzzle. It gives the puzzle a hook that is easy to latch onto.
  • Puzzle is easier to describe
  • Puzzle is easier to reason about
  • Avoids blank page syndrome

• Don’t be clever in ways that the solver will never notice. Such themes reduce the sparkle of the puzzle .

  • When the aesthetic or thematic goal of the puzzle is too high, consider compromising with the gameplay needs — the need to make the puzzle fun and surprising.
  • Be careful that the theme of the puzzle does not distract the player from having the eureka moment of understanding what the puzzle is communicating or teaching about .

• The format changes everything. The following influences the player experience

  • The modality of the system (physical, paper, digital)
  • Control systems
  • Text size
  • Undo, restart and savestate systems
  • Enforcement of rules
  • Punishment of mistakes
  • Is note taking allowed in game?

• Notetaking can increase depth

  • Notetaking facilitates exploration of solution spaces that have high breadth or are very complex.
  • Some puzzles are simply not possible without the ability to take notes due to cognitive load.
  • That said, notetaking is a tool and not all games need note-taking. Simple puzzles that players can solve in their head do not require note taking.
  • Notetaking feels like progress

• Beware of the temptations offered by the interactive medium. All these pitfalls are not necessarily bad, but they can make it unruly for the player.

  • Physics-based puzzles - the continuous nature of the physics system makes deduction difficult.
  • Statefulness - it adds complexity that can be hard to reason with.
  • Linearity and Plot-driven puzzle games - it can block player progress and force them to look up the solution instead, bypassing the puzzle’s purpose entirely.
    • Challenging puzzle games tend to be non-linear.

• Break the rules, sometimes perversely.

• Serendipity occurs when you’re actively searching for good ideas.

  • Great puzzles are often discovered not invented.
  • Be actively searching for interesting lemmas from the ruleset. Recognize the potential in every innovation you might see in the course of puzzle design.
  • The key is to listen — notice your Eureka moments and turn them into puzzles.

Links §

• Creativity

Footnotes §

1. In this sense, making a puzzle is a lot like making a new algebra and studying its consequences. ↩