--- id: game-design-practice lang: en title: Game design practice description: >- Original exercises about choice, balance and player behaviour, with constraints, deliverables and prototype checks. intro: >- Short exercises about choice, balance and player behaviour. Build a small prototype, play it and explain what changed. durationNote: >- Times are estimates. Testing with another player may require a separate session. provenance: >- Thematic inspiration: Challenges for Game Designers by Brenda Brathwaite and Ian Schreiber. These scenarios, constraints and checks are independently written around broad game-design learning topics. This is not a translation or a sequential retelling of the book exercises. updatedAt: '2026-10-20' sources: - title: 'Challenges for Game Designers: author website' url: 'https://designgames.wordpress.com/' description: The book description and its practice-based approach to game design. - title: 'Challenges for Game Designers: table of contents' url: 'https://designgames.wordpress.com/toc/' description: >- The book learning topics. A source for thematic inspiration, not the text of these exercises. counts: tasks: 14 url: 'https://gamemechanics.pro/en/practice/' markdownUrl: 'https://gamemechanics.pro/content/practice/en/game-design-practice.md' license: CC-BY-4.0 codeLicense: MIT attribution: gamemechanics.pro contributors --- # Game design practice Short exercises about choice, balance and player behaviour. Build a small prototype, play it and explain what changed. Times are estimates. Testing with another player may require a separate session. Thematic inspiration: Challenges for Game Designers by Brenda Brathwaite and Ian Schreiber. These scenarios, constraints and checks are independently written around broad game-design learning topics. This is not a translation or a sequential retelling of the book exercises. ## 01. A current on paper id: paper-current url: https://gamemechanics.pro/en/practice/#paper-current Topic: Prototyping Level: Beginner Estimated duration: 60 min **Goal:** Preserve the key decision of a digital mechanic in a simple tabletop prototype. A boat crosses a river with moving obstacles. The player chooses between a short dangerous route and a longer calm one. Move this situation onto a grid: the boat and current move in steps, while tokens represent the water state. **Materials:** Board and pieces ### Constraints - No screen, timer or facilitator making the player's decisions is needed. - A game lasts no more than ten turns. - Before choosing a route, the player sees at least some consequences of the next step. ### Steps 1. Write down the decision you want the adaptation to preserve. 2. Define the order of boat, current and obstacle movement. 3. Play the short and long routes from the same starting state. Clarify the collision rules. ### Deliverable A board, single-turn rules and a comparison of the two routes. ### Check your work - Another person can play a turn without extra verbal rules from the designer. - The routes differ in risk or cost, not just appearance. - List the feelings of the digital version that the paper prototype cannot test. **Reflect:** Which decision survived the adaptation? What was lost by making time turn-based? ## 02. The last ferry id: last-ferry url: https://gamemechanics.pro/en/practice/#last-ferry Topic: Choice and rules Level: Beginner Estimated duration: 45 min **Goal:** Create a choice whose value depends on the situation. You run a ferry before the harbour closes. Different cargo earns points but takes different amounts of space and time. Design a six-turn game: each turn, the player chooses what to carry, what to leave behind and when to send the ferry. **Materials:** Paper and tokens ### Constraints - Use no more than three actions and two resource types. - All costs and rewards are known before the choice. There are no random events. - The rules must fit on one page. ### Steps 1. Describe the goal, actions, ferry capacity and end of the game. 2. Prepare two starting setups in which different first actions are preferable. 3. Play each setup once. Change one rule and repeat the check. ### Deliverable A rule sheet, two starting setups and a short note about the difficult choice. ### Check your work - Each setup offers at least two legal first actions. - One action does not outperform every other action on every consequence in every setup. - The player can explain what they gave up to carry their chosen cargo. **Reflect:** Which constraint creates an interesting decision? What happens if you remove it? ## 03. One rule, two games id: one-rule-two-worlds url: https://gamemechanics.pro/en/practice/#one-rule-two-worlds Topic: Choice and rules Level: Beginner Estimated duration: 40 min **Goal:** Trace how a small rule change alters player behaviour. Create a small game about collecting samples on an island. In version A, players earn points for delivered samples. In version B, change just one scoring rule to reward a different behaviour, such as distributing finds between participants. **Materials:** Cards and a board ### Constraints - Keep the board, actions, turn count and starting resources identical in both versions. - Change only one reward rule. - Do not add items or abilities. ### Steps 1. Write down the behaviour you want each version to encourage. 2. Play both versions from the same starting setup. 3. Compare routes and decisions. If behaviour stays the same, adjust the reward. ### Deliverable Two rule sets with the difference highlighted and a turn-by-turn table of player decisions. ### Check your work - Exactly one rule has changed. - Expected and observed behaviour are recorded separately. - Explain why the reward changes decisions, or why it had no effect. **Reflect:** Which incentive mattered most: points, a limited resource or another player's actions? ## 04. Three minutes at the museum id: three-minute-museum url: https://gamemechanics.pro/en/practice/#three-minute-museum Topic: Audience and interface Level: Beginner Estimated duration: 60 min **Goal:** Connect audience constraints to rules and the first interaction. Make a game for a museum visitor who has three minutes. They see the screen for the first time, need not read a long tutorial and may leave at any moment. The game idea is to reconnect an object with its use. **Materials:** Paper screens ### Constraints - The initial screen makes the first action visible. - A session fits within three minutes and can restart immediately. - Key instructions work without sound and do not depend solely on colour. ### Steps 1. Describe the visitor's situation without inventing verified audience research. 2. Draw the initial screen, play screen and ending. 3. Check the paper scenario with someone who has not seen the project. Note where help was needed. ### Deliverable Three screens, session rules and a record of a new player's first attempt. ### Check your work - The player can identify the first action without a hint from the designer. - The result of an action is visible on screen. - Ending and restarting need no explanation. **Reflect:** What did the short session force you to remove? What did the player gain? ## 05. Remove the market id: remove-the-market url: https://gamemechanics.pro/en/practice/#remove-the-market Topic: Iteration Level: Beginner Estimated duration: 50 min **Goal:** Check what a mechanic contributes and whether it is worth keeping. Take a small game of your own that has several mechanics. Choose a supporting system, such as resource trading. Make a version without it while preserving the game's goal. Check which decisions disappear and which rules become simpler. **Materials:** An existing small prototype ### Constraints - Use your own prototype or a game you have permission to modify. - Remove one system and its associated exceptions. - Do not replace it with another large mechanic. ### Steps 1. Write down what the chosen system should offer the player. 2. Prepare the reduced version and check that the game can end. 3. Compare explanation time and decisions in both versions. Decide what to restore, change or leave removed. ### Deliverable Two rule sets and a justified decision about the mechanic's future. ### Check your work - The reduced version contains no references to removed resources or actions. - The goal remains reachable. - Name the useful choice that was lost or explain why there was none. **Reflect:** Which complexity did the designer notice? Which did the player actually notice? ## 06. The signal room id: signal-room url: https://gamemechanics.pro/en/practice/#signal-room Topic: Puzzles Level: Intermediate Estimated duration: 60 min **Goal:** Build a puzzle around a consistent rule and a player discovery. A control panel has four levers and three lamps. Invent a rule that connects lever positions to the lamps. The player must produce a target signal. Create three levels: discovering the rule, checking it and using it in an unexpected combination. **Materials:** Paper prototype ### Constraints - The same rule applies at every level. - Available information allows a solution without random trial and error. - The third level is harder because it combines conditions, not because it introduces a hidden rule. ### Steps 1. List the panel states and confirm the target states are reachable. 2. Prepare three levels and a solution for each. 3. Give them to someone who has not seen the rule. Record where they get stuck and revise the clues. ### Deliverable Three panel diagrams, the rule, solutions and a note from the new-player check. ### Check your work - Each level has at least one known solution. - Clues agree with the rule. - Identify the specific discovery the player needs to make. **Reflect:** Was the player struggling to understand the rule or to apply a rule they already understood? ## 07. The weather window id: weather-window url: https://gamemechanics.pro/en/practice/#weather-window Topic: Chance and balance Level: Intermediate Estimated duration: 50 min **Goal:** Make a random outcome part of a decision rather than just a surprise. An expedition must send fragile cargo today or wait for another weather window. Today's forecast gives known probabilities for three outcomes. Design the cost of waiting, rewards and losses, then compare the choices of a cautious player and a risk-taking player. **Materials:** Event cards ### Constraints - Probabilities, rewards and losses are visible to the player. - A ten-card deck can represent the outcomes. - Waiting has a cost; it does not grant a free retry. ### Steps 1. Build a deck with the intended proportions of outcomes. 2. Calculate the expected result of sending: sum each outcome's result multiplied by its probability. 3. Compare sending and waiting at two resource levels. Play through a few outcomes and note separately that this small sample does not establish balance. ### Deliverable The deck, a table of probabilities and consequences, an expected-result calculation and two example decisions. ### Check your work - The probabilities sum to one. - The player knows what they can gain and lose. - Expected gain and the risk of losing the last resources are considered separately. **Reflect:** Why might a player choose the option with a lower expected reward? ## 08. Forecast and surprise id: forecast-and-surprise url: https://gamemechanics.pro/en/practice/#forecast-and-surprise Topic: Chance and balance Level: Intermediate Estimated duration: 75 min **Goal:** Give the player ways to prepare for a random event. A caravan chooses a road between settlements. In version A, danger is determined only after the route is chosen. In version B, part of the forecast is visible beforehand and preparation costs a resource. Compare the decisions that become possible with a forecast. **Materials:** Cards and a play log ### Constraints - Keep routes, rewards and the event set identical in both versions. - Advance information is incomplete but useful. - Preparation reduces a specific loss and has a clear cost. ### Steps 1. Describe both versions and the preparation options. 2. Compare them using the same event order. 3. Record decisions before events and results afterwards. Adjust the cost of preparation if players always take it or always ignore it. ### Deliverable Two rule sets, a decision log and an explanation of the limits of your conclusions. ### Check your work - The forecast allows a change of action before the event. - Preparation does not automatically cancel every consequence. - Do not present a few games as statistical evidence of balance. **Reflect:** Is the player learning to prepare better or simply memorising the event sequence? ## 09. Three workshop shifts id: three-workshop-shifts url: https://gamemechanics.pro/en/practice/#three-workshop-shifts Topic: Chance and balance Level: Intermediate Estimated duration: 60 min **Goal:** Check progression pacing through action costs and changes to available decisions. The player repairs devices, earns parts and upgrades two tools. Design three stages: the start, a first upgrade and specialization. The numbers are experiment parameters rather than a ready-made balance. **Materials:** A spreadsheet, calculator and paper prototype ### Constraints - Use two tools, one resource and three stages. Specify the cost of one repair at each stage. - At least one upgrade changes an action rule rather than only a reward multiplier. - Record action time separately from resource cost. ### Steps 1. Tabulate upgrade prices, repair times and income; calculate the action count before each purchase. 2. Compare two purchase orders on the same task sequence. Record where each tool is preferable. 3. Add a random reward and compare independent rolls with a predefined sequence; retain attempt-level results. 4. Change one price threshold and repeat the check: which decisions become available earlier or later? ### Deliverable A three-stage table, two comparable purchase logs and an explanation of the pacing change after one edit. ### Check your work - Action counts account for rounding up to a whole repair. - Each tool has at least one situation in which it is useful. - The rule-changing upgrade changes a player decision alongside the displayed number. - An average reward expectation is not presented as a guarantee for a short sequence. **Reflect:** Where does the player repeat a solved task without a new choice? How can you shorten that section while keeping the next goal understandable? ## 10. The shared workshop id: shared-workshop url: https://gamemechanics.pro/en/practice/#shared-workshop Topic: Strategy Level: Intermediate Estimated duration: 60 min **Goal:** Create a strategic decision that considers an opponent's actions. Two craftspeople share a set of machines to fulfil orders. An occupied machine is unavailable to the opponent until the next round. Design a choice between a quick simple order and preparation for a valuable order that a competitor can disrupt. **Materials:** Two players and tokens ### Constraints - There is no randomness or hidden information. - Use no more than three machines and three order types. - The game ends after six rounds. ### Steps 1. Describe available actions and when machines become free. 2. Create two situations in which the preferable action depends on the opponent's response. 3. Swap players' positions and check the effect of going first. ### Deliverable Rules, two analysed game states and observations about turn order. ### Check your work - Explain how one player's action changes the other's options. - At least one decision depends on the opponent's expected response. - Check whether the first player receives a persistent advantage. **Reflect:** Does a player win through planning or through the privilege of moving first? ## 11. Rhythm without rushing id: rhythm-without-hurry url: https://gamemechanics.pro/en/practice/#rhythm-without-hurry Topic: Motor skill Level: Beginner Estimated duration: 45 min **Goal:** Separate reaction accuracy, rule comprehension and accessible controls. The player signals ships passing a lighthouse. Acting within a short timing window earns a bonus. Design a standard mode and one with a wider window, keeping the same goal and event sequence. **Materials:** Paper scenario or simple prototype ### Constraints - Use one main input and no complicated button combinations. - Give clear feedback for early and late actions. - Colour must not be the only way to understand a signal. ### Steps 1. Draw a timeline showing the successful window and feedback. 2. Set parameters for both modes. 3. Check the sequence in a simple prototype or act it out as a scenario. State separately that reaction quality has not been tested without a working timer. ### Deliverable Two timelines, timing-window parameters and a list of feedback signals. ### Check your work - The player understands why an action counts or fails. - The wider window preserves a meaningful action. - Distinguish control difficulty from decision difficulty. **Reflect:** Which part of the challenge does a wider window change? Which part does it preserve? ## 12. Light for two id: light-for-two url: https://gamemechanics.pro/en/practice/#light-for-two Topic: Story through rules Level: Intermediate Estimated duration: 60 min **Goal:** Express a story's theme through actions and consequences. A winter station's energy supply decreases each turn. The player can light the living area or maintain communication with a neighbouring station. Create a short game about interdependence in which meaning emerges from energy allocation and the neighbours' responses. **Materials:** State cards ### Constraints - Use no more than five introductory sentences. - No dialogue directly states the moral. - At least one consequence arrives a turn later rather than immediately. ### Steps 1. State the theme in one sentence. 2. Describe the actions and delayed consequences. 3. Run the game and ask the player to explain what it is about in their own words. Compare the answer with your intention. ### Deliverable Rules, a consequence map and a record of the player's interpretation. ### Check your work - At least one repeated player action relates to the theme. - Consequences follow from rules rather than an arbitrary designer decision. - Keep your intended interpretation separate from the player's answer. **Reflect:** If you remove the introduction, what part of the story remains understandable? ## 13. Two dispatch windows id: two-windows url: https://gamemechanics.pro/en/practice/#two-windows Topic: Multiplayer Level: Intermediate Estimated duration: 75 min **Goal:** Make cooperation necessary by distributing information and actions. Two dispatchers guide a repair team through a station. One sees the schedule of safe passages; the other knows the tool supply and door conditions. Design a shared goal and communication rules so that each person's decisions affect the result. **Materials:** Two players and route cards ### Constraints - Players have different information and available actions. - One participant cannot directly perform the other's actions. - The first version needs no network: use two paper information sets. ### Steps 1. Divide information and actions between the roles. 2. Describe permitted communication, winning and losing. 3. Play a game and swap roles. Note moments when one player made every decision while the other merely carried them out. ### Deliverable Two card sets, shared rules and notes on each role's independent decisions. ### Check your work - Each role makes at least one decision affecting the shared goal. - Both players understand what they may communicate and do. - Identify a reason to cooperate and check whether one participant dominates. **Reflect:** Which information needs to stay separate to keep both roles involved? ## 14. Maintenance patrol id: maintenance-patrol url: https://gamemechanics.pro/en/practice/#maintenance-patrol Topic: NPC behavior Level: Intermediate Estimated duration: 90 min **Goal:** Connect perception, memory, action selection and route following. A robot patrols a station, detects a fault and moves to repair it. A door may close between it and the target. Store the robot’s known target position separately from the actual world state. **Materials:** A small scene or step-by-step model with an event log ### Constraints - Use one map, one robot and one fault. Specify the graph and edge costs explicitly. - The robot learns through observations only; its memory expires. - Every action change logs time, observation and reason. ### Steps 1. Define patrol, move-to-fault, repair and search states, including transition conditions. 2. Separate observations from memory. Losing sight must not instantly reveal the target’s new position. 3. Connect action selection to path search and movement; recompute after a closing door or return an explicit failure. 4. Repeat these scenarios: visible target, lost target, closing door and unreachable target. Preserve identical starting conditions. ### Deliverable A runnable scene or reproducible step-by-step model, map, transition table and four logs explaining decisions. ### Check your work - Identical starting conditions and observations produce identical decisions. - After losing sight, the robot uses the last known position until memory expires. - A closed door causes neither wall traversal nor endless path recomputation each frame. - Target reached, no route and start equal to goal are distinguishable results. **Reflect:** Which wrong decision came from memory, which from priorities and which from navigation? Change only one of these layers. ## Sources - [Challenges for Game Designers: author website](https://designgames.wordpress.com/): The book description and its practice-based approach to game design. - [Challenges for Game Designers: table of contents](https://designgames.wordpress.com/toc/): The book learning topics. A source for thematic inspiration, not the text of these exercises.