This week’s prototype explored puzzle mechanics, where the observer defines reality through a colour filter system that controls which cubes exist, how they behave, and whether the player can interact with them.

Some entities, like an autonomous eye, follow their own filter logic, adding asymmetry to how the world responds. Movement is also filter-dependent: switching states while overlapping certain cubes launches the player or locks them in place. The level design draws inspiration from Portal, using a structured progression, setup-payoff interactions, and layered spatial logic, including moments where players must set-up level spaces in advance to trigger delayed, solutions.

Key Design Highlights

• Built rooms with clear goals. Players can identify key objects and exits immediately upon entering rooms.
• Embed incorrect assumptions into the puzzle design, enabling puzzle revelations.
• Combined visibility logic with movement mechanics. Overlapping objects during filter changes trigger state-based motion.
• Structured puzzles to require setup in neutral spaces before activating filters, encouraging forward planning.
• Introduced mechanics in isolation before combining them in more complex, layered challenges.
• Designed non-player perception systems where certain objects interpret reality differently from the player, creating asymmetrical logic.
• Took inspiration from Portal’s approach to puzzle readability and gradual escalation.

What Worked Well

• Staggered introduction of mechanics helped build player understanding and supported the development of more complex challenges.
• The filter system proved flexible and opened up a range of interesting design opportunities.
• Puzzle layouts that encouraged non-linear problem solving through object reuse, intentional destruction, and delayed interactions.
• Designed each level with a specific goal or concept, allowing them to connect into a clear, consistent progression path.

What Could Be Improved

• The interaction system limited player freedom in some situations and occasionally made the intended solution harder to read.
• The additional movement mechanics did not add enough to the puzzle design. They functioned more as platforming segments than integrated systems and would have required further development to support gameplay meaningfully.
• A more focused interaction design could have opened up more possibilities for expression or experimentation within the puzzle space.

This week’s prototype explored rule-constrained navigation through an unknown rule set and customisable control scheme. It is structured as a top-down 2D track sport where players must complete events without committing any suspicious behaviour.

Before each attempt, players assign movement actions from a limited pool, building a control layout that must function across spatial challenges while avoiding hidden rule violations. The rules remain consistent between levels, and any breach is flagged immediately, with the corresponding rule added to an on-screen rulebook. The short-loop structure supports trial-and-error without relying on guesswork, aiming for clarity, fairness, and fast iteration. Design emphasis was placed on visibility, iteration, and adaptation within tight constraints, drawing inspiration from the immediate feedback and active rule-tracking seen in King of the Bridge.

Key Design Highlights

• Built rooms with clear goals. Players can identify key objects and exits immediately upon entering rooms.
• Embed incorrect assumptions into the puzzle design, enabling puzzle revelations.
• Combined visibility logic with movement mechanics. Overlapping objects during filter changes trigger state-based motion.
• Structured puzzles to require setup in neutral spaces before activating filters, encouraging forward planning.
• Introduced mechanics in isolation before combining them in more complex, layered challenges.
• Designed non-player perception systems where certain objects interpret reality differently from the player, creating asymmetrical logic.
• Took inspiration from Portal’s approach to puzzle readability and gradual escalation.

What Worked Well

• Playtesting showed players responded well to the rule discovery format, clearly identifying when and why they failed.
• The gameplay loop encouraged experimentation, supporting an enjoyable trial-and-error experience without confusion or frustration.
• Customisable controls led to varied approaches, with players making deliberate trade-offs in movement costs and control mapping.
• Level layouts pushed efficient planning and adaptation, leading to satisfying problem-solving within limited inputs., allowing them to connect into a clear, consistent progression path.

What Could Be Improved

• Earlier levels could have more tightly controlled rule exposure to better manage learning flow and avoid premature discoveries.
• Additional content and variety, such as more event types, additional challenges, and meaningful rewards, would have reinforced replayability.
• Some systems lacked depth, particularly around challenge incentives and progression pacing, which could be expanded in a future version.

This week’s prototype explored Metroidvania-style progression built around genetically modified animal abilities, created under the randomly assigned theme Meat Hero.

Set in a 2.5D world, the player unlocks five meat-based powers, each derived from modified organisms and altering how they interact with the environment. Abilities are introduced through controlled encounters, requiring players to overcome specific traversal challenges. Level design emphasises gated progression, forced backtracking, and soft-path routing based on the current ability set. The structure supports a linear introduction of mechanics while allowing for optional divergence and layered puzzle spaces. Design focused on teaching through use, encouraging players to understand and apply each new skill as part of a growing toolkit.

Key Design Highlights

• Limited players to two bound abilities at a time, forcing deliberate loadout choices and ability combinations.
• Designed traversal challenges around ability pairings, requiring players to experiment with sequencing and spatial timing.
• Introduced each ability through immediate gated encounters, reinforcing function and enabling meaningful backtracking.
• Used consistent visual cues for ability interactions, including red glow accents and shared material types for readability.
• Built level spaces with gates and diverging paths, supporting both guided progression and optional exploration.
• Balanced puzzle structure to reward mastery of single abilities before layering them into multi-step solutions.

What Worked Well

• The metroidvania structure tested well in playtests, with no major feedback calling backtracking tedious or repetitive.
• Fencing and layout design clearly communicated ability requirements and encouraged creative combinations using only two active powers.
• Despite differing interaction types, all abilities were understood and used effectively without overuse of tutorials.
• Visual design choices, such as consistent glow cues and material types, supported fast environmental readability.

What Could Be Improved

• Earlier playtesting would have helped refine the feel of the movement system; the pulsing motion was often described as imprecise or difficult to control.
• Some mechanics, especially the final ability, were well received but underutilised due to time constraints and limited level space.
• The ability switching system could be further refined for responsiveness and flow, improving how players chain powers in active traversal.