Exploration-Driven Blueprint Acquisition

[bright axle]

💡 Suggestion: Exploration-Driven Progression

Hello Dev Team,

I love the core loop, but the current progression feels detached from the game's theme.

🛑 The Core Problem

The XP/Skill Point leveling system feels arbitrary and detracts from the 'scavenge-and-restore' atmosphere. Earning points from recycling trash and spending them on a menu to unlock key items feels abstracted from the core loop of exploration, repairing, and caretaking.

✅ Proposed Solution: Discovery & Repair

I propose removing the linear XP/Skill Point system for new capabilities. Instead, progression should be tied directly to environmental discovery and core objectives. Players would find capabilities, not simply buy them.

🔨 Concrete Examples: How to Acquire Skills

• Crafting Blueprints (e.g., Generator): Acquired only after discovering a broken version in the environment and successfully repairing/dismantling it. This reinforces the theme of repairing what's broken.
• Enhancer Modules/Skills (e.g., Weapon/Caretaker Attachments): Found as physical loot in hidden caches, successful repairs, or as drops from dangerous rogue machines. This makes exploration much more rewarding.
• Key Tools (e.g., Dismantle/Repair Tools): Unlocked via mandatory quest steps that require fixing a core system on the ship or a facility. Tool acquisition feels earned and necessary.

🌟 The Benefit

This change would make progression feel more organic and immersive. The reward for deep exploration is new, tangible capabilities making discovery a much more exciting and meaningful moment.

I think this small change would greatly enhance the atmosphere and reward loop!

[lucid chasm]

I think this would be good in addition to, not replacing the core xp/skill system. They've leant hard into the idea of functioning as a robot and 'upgrading' yourself and I think it would change the game fundamentally if they moved away from that.

I think the addition of finding things that have been 'cobbled together' in the wild that you could reproduce would be interesting though, such as a combined wind/sun generator that some robots made out of spare parts when they started running out, etc?

[bright axle]

That's a great point about the developers leaning hard into the robot/upgrade concept. I agree that keeping the core loop feeling consistent is crucial.

However, my suggestion isn't to stop the player from "upgrading" or finding new parts, but to change how those upgrades are acquired to better fit the robot theme.

Why XP is Thematic Conflict

I think the current XP system is exactly why the mechanic feels disjointed from the "robot" concept. Robots don't "grow" from experience; they adapt by adding or changing physical components.

• Blueprints/Skills: A caretaker robot should either already have the data for basic blueprints or find the necessary data/schematics in old facility hard drives, archives, or scanning broken equipment. Earning generic XP to unlock a complex engineering blueprint feels arbitrary.
• Caretaker Enhancers: These should be physical, swappable modules (like your suggestion about "cobbled together" parts). The robot is upgrading its hardware by physically installing a found part, not leveling up its software through a generic RPG system.

Leveraging the "Cobbled Together" Idea

I really like the idea of finding "cobbled together" designs! This fits perfectly with the proposed system:

Instead of being rewarded for level milestones, the player discovers unique, high-value blueprints or modules by exploring locations where other robots have attempted jury-rigged solutions (e.g., finding the unique combined wind/solar generator blueprint in an old emergency shack).

This approach maintains the focus on upgrading the robot while making progression feel like exploration, discovery, and physical augmentation—the way a robot should evolve.

[lucid chasm]

I think you're being a little bit too literal about growth. The XP system is more of a metaphor about the player/robot processing and integrating new data, similarly to how we train LLMs on things so they can develop.

Again, I do think your suggestions have merit though and your additional idea of finding 'parts' or physical things you can integrate into your robot are a great idea.

[lyric robin]

i will admit i prefer the more organic method of levelling and learning, based on what you do instead of "magically" acquiring new knowledge

[lucid chasm]

The way it's designed does give us a little more freedom - if it was designed around acquiring items and upgrading your robot that way, we would be roadlined into going to specific ways to get specific upgrades. The way it's designed currently you can go where you want and do what you want and still get XP to level your skills. (Disregarding the storyline)

[bright axle]

You raise a critical point about roadlining—forcing players to a single spot for essential gear. Player freedom must be maintained while improving immersion.

1. Thematic Redundancy for Discovery

To prevent roadlining, essential Schematics must have multiple, logical locations across the map.

• Example: Schematics could be found in a high-security archive, a distant console, or scavenged from a difficult Rogue Unit. This lets the player choose a path based on their playstyle (stealth, puzzle, or combat).
• Enhancer Modules (physical loot) should be high-value, rare rewards for completing difficult objectives or defeating high-tier enemies, rather than being tied to one specific map location.

This ensures discovery is exciting, but never mandatory along a single route.

2. Renaming XP for Immersion (The Capacity Model)

We keep the freedom of XP gain but re-contextualize the spending to fit the robot theme.

• XP becomes Operational Data (OP Data): Data is collected from any task (recycling/combat). The robot is collecting data on how the world operates, like how LLMs are fed data.
• Skill Points become Processing Units (P-Units): P-Units are spent on Core Processing Upgrades (CPU Tier), the literal cost to upgrade the robot's processor to use the additional hardware.
• The System: You must find the Schematic and have the necessary CPU Tier (unlocked by spending P-Units) to use it. This links progression to actual hardware limitations, not abstract growth.

🔄 The Combined Result

This system maintains the best features of both ideas:

• Freedom is Maintained: Earn OP Data anywhere for Capacity Upgrades.
• Roadlining is Avoided: Crucial Schematics are acquired via multiple paths.
• Immersion is Improved: You find the knowledge (Schematic) and upgrade the hardware (CPU) to utilize it.

Discovery remains exciting without forcing linear movement.

[lucid chasm]

I have to ask, are you using AI just to help with formatting and the like because I don't want to have a discussion with an LLM so if these aren't your actual ideas, I'd rather just drop it.

If they are your ideas and you're just using an LLM for coherence, that's completely fine.

[bright axle]

These are my ideas, but yes, I am using AI to help make the idea more comprehensive and help with formatting to bring to the forefront the important points

[lucid chasm]

Regarding point 1 : I think we run into risks of linearity no matter how you design it, since the game is handcrafted. If it were more open world or proc gen, it would be possible to have more 'random drops' that you could use as upgrades etc - but since we have a finite amount of enemies, it would essentially amount to the same thing, which leads me to point 2 : this is basically what we have already with extra steps.

I think (I could be wrong) you're coming at it from a point of immersion and realism which is totally fine - it's just that sometimes developers need to adhere to some level of balanced gameplay in order to deliver a product that's actually fun and playable.

Anyway, I'll leave it for other people to have input, but I think your ideas have merit in ways, just not as a total overhaul (in my opinion).

[bright axle]

I understand your point. But I believe that the Enhancers aspect, as the most disconnected part of the leveling system, should be changed to modules that can be built and swapped as needed.