Repository navigation
Process Make Fun
Run the complete prototype as a player and validate that the core loops are actually fun before moving to v0.5. This is a structured playtest and tuning pass, not a feature sprint. No new features ship here. Things that feel bad get fixed.
Points: 3 (tuning and iteration) Dependencies: All prototype features complete (HUD Pass, Ball Scaling, Progression System, Idle Play, First Partner Unlock, Desktop Experience) Unlocks: v0.4-0.5 Beta (the prototype is declared done and fun is considered proven enough to share)
These are the three things the prototype must deliver. Everything in this pass is tested against them.
The player picks up the paddle, keeps a streak going, and wants to keep going. The difficulty ramp from ball scaling should create tension without feeling unfair. When the streak ends, the player's first thought should be "I need to upgrade", not "that was random."
Player leaves the game running or closes it. When they return, something is waiting: friendship earned, a milestone approaching, a streak that ended naturally. The return moment should feel like a small reward, not a chore to process.
After spending friendship and upgrading, the player feels the change immediately. Paddle is faster, bigger, or the ball starts slower, something concrete shifts. The loop of "struggle → upgrade → feel good → hit a new wall → upgrade again" should complete at least once in a single session.
Run each session with fresh eyes, no dev shortcuts.
Start a fresh save. Play actively for 10 minutes without touching the upgrade shop.
What to observe:
- Is the starting difficulty appropriate? Can a new player keep a 5+ volley streak in the first minute?
- Does the ball scaling feel gradual or sudden?
- Is there a natural moment where the player wants to open the shop?
- Does 10 minutes feel too long, too short, or right?
Start from Session A save. Buy the first item, play until the streak noticeably outpaces the improvement, buy again.
What to observe:
- Does each item feel impactful or marginal?
- Is the friendship curve right? (Too fast = items feel trivial. Too slow = player gives up before buying anything.)
- Do the available item choices feel meaningfully different from each other?
- Is there a moment where the player thinks "I know what I want next"?
Leave the game running for 30 minutes (active window, no input). Return and review.
What to observe:
- Is the auto-play AI competent enough to earn friendship but not so good it trivialises active play?
- Is the friendship earned in 30 minutes idle proportionate to active play?
- Does the gap between idle friendship rate and active friendship rate create a reason to jump in?
Close the game. Wait at least 2 hours. Reopen.
What to observe:
- Does the welcome back screen feel like a gift?
- Is the offline friendship amount satisfying without feeling broken?
- Is the return-to-gameplay flow smooth? (No friction before you see the ball bouncing again.)
These are the numbers most likely to need adjustment. Change one at a time, replay the relevant session.
| Variable | File | What it affects | Warning signs |
|---|---|---|---|
| Ball speed increment per hit |
ball.gd / GameRules
|
Tension ramp, natural streak ceiling | Too fast: streaks always end at ~10. Too slow: ball never feels threatening. |
| Friendship per hit (base) | game.gd |
Time to first item purchase | Too high: items feel trivial. Too low: player quits before buying anything. |
| Item friendship costs | ItemDefinition |
Progression pacing | Too cheap: all items gone in one session. Too expensive: player never progresses. |
| Item effect values |
ItemDefinition / ItemEffect
|
Upgrade impact | Too small: player can't feel the difference. Too large: game becomes trivial fast. |
| Idle friendship rate (% of active) | idle_play.gd |
Reason to play actively | Too high: active play not worth it. Too low: idle feels pointless. |
| Offline earnings cap (hours) | progression_data.gd |
Return reward size | Too high: feels broken. Too low: returning after a day feels bad. |
| Auto-play miss rate | AI constants | Idle streak length | Too accurate: idle earns too much. Too clumsy: idle feels broken. |
The prototype is done when all three play goals are met:
- A player unfamiliar with the game (or you, after a week away) can pick it up and have a satisfying 10-minute active session without guidance
- Leaving the game for 2+ hours and returning produces a moment that feels good, not neutral
- The upgrade loop completes naturally in a 20-minute session (spend friendship on an item, feel improvement, hit a new wall, want to buy again)
If any criterion fails after tuning, identify the root cause and fix it before declaring the prototype done. This is the gate before v0.5.
Idle games live or die on feedback loops. Every player action that matters should produce a visible, audible, or physical response. The make fun pass is the right time to evaluate whether the prototype needs a dedicated juice library from AssetLib.
Candidates:
- Godot Juice or similar tweening/effects library: screen shake on miss, bounce/pop on hit, particle burst on streak milestones, satisfying snap when items drop into slots. GDFXR covers sound; this covers motion and visual punctuation.
The goal is not polish (that is v0.5 work). The goal is confirming that the feedback loop feels like something. If a hit and a miss produce the same visual response, the player has no reason to care about streaks. A small amount of juice during the make fun pass is enough to validate whether the core loop has the right feel before committing to full audio and animation work in Beta.
- Not a bug fix sprint (bugs blocking play goals are in scope; everything else is not)
- Not a polish pass (animations, art, audio quality are v0.5 work)
- Not a balance pass (deep economy tuning is in v0.5, Balance Pass. This is just "is it fun at all?")
Contributing
The game
The Cast
Process
Reference