Pixel Cloud Lab

Cinevva's analyzer read this game's design doc and its source, then wrote the brief below. "Make your own" hands it to the game creator as your opening message, so you start from a real spec rather than a one-line idea.

Create a first-person 3D educational exploration game titled “EARTH IN CHANGE: PIXEL CLOUD LAB.” Begin with an on-screen modal introduction over the lab: “You are a Junior Earth-System Scientist.” Explain that the player must collect materials, assemble a cloud-in-a-bottle, run three trials, record measurements, and connect aerosols to condensation, the water cycle, and ecosystems. Clearly warn that this is a simplified model and does not directly measure PM2.5 or rainfall. Include an ENTER LAB button.

Use an FPS camera in a futuristic enclosed science laboratory. The player starts near the front of the room and can walk with WASD or arrow keys, look with the mouse, interact with E, run trials with R, and open the notebook with N. Show a small circular crosshair, a cyan-bordered objective panel at upper left, controls help at upper right, and contextual prompts near the bottom center. The lab has dark teal walls, floor, ceiling, fog, cyan/blue glowing trim, amber lights, benches, shelves, cabinets, monitors, computers, desks, and scientific display boards. Use a low-poly/pixel-tech visual style with a very dark navy background, luminous cyan, lime-green, and amber accents, monospace/arcade typography, hard rectangular panel borders, and emissive materials. The room should contain a central experiment table and wall boards labeled EARTH SYSTEM, WATER CYCLE, PARTICLES, and WILDFIRE. The visible board text includes concepts such as atmosphere/smoke clouds, hydrosphere/rain water, biosphere/forest life, geosphere/soil peat, evaporation → cloud, condensation → rain, runoff → soil/rivers, smoke, aerosols, CCN, and wildfire smoke references.

Implement the gameplay as a staged objective sequence. Stage 1 says “Find every material in the lab” and shows a checklist. Place these eight interactable objects around the room, each with a visible uppercase floating label: Transparent 2-L bottle, Bottle cap, Experimental liquid, Hand/bicycle pump, Laboratory lamp, LUX sensor, Aerosol particle sample, and Scientist notebook. They are arranged mainly on/around the central table and nearby benches. When the player approaches an item and presses E, mark it collected, hide it, check it off, and show a short completion toast. Once all eight are collected, advance to Stage 2.

Stage 2 says to assemble components at the central table by pressing E. Perform these seven assembly steps in order, one interaction per press: Bottle, Experimental liquid, Bottle cap, Pump, Light source, LUX sensor, Aerosol sample. Display a checklist of these steps in the objective panel and toast each “ASSEMBLED” item. After all steps, create a visible assembled rig on the central table: a transparent bottle with blue liquid, a cyan LUX sensor, an amber light source, and a label indicating “LIGHT SOURCE → BOTTLE → LUX SENSOR.” Advance to the trial stage and show “CLOUD-IN-A-BOTTLE READY.”

Stage 3–5 are three timed trials run from the central table. Trial 1 is “NO AEROSOL,” trial 2 is “AEROSOL,” and trial 3 is “HIGHER AEROSOL.” Start a trial by pressing R, or E while interacting with the table. While active, show a cloud inside/above the bottle as a cluster of many small semi-transparent white/blue glowing spheres. Animate its scale subtly and vary its visual strength: weakest for No Aerosol, stronger for Aerosol, strongest for Higher Aerosol. Use cyan, amber, then purple accent lighting for the three trials. Run each trial for roughly 8, 10, and 12 seconds respectively, then hide the cloud and open a result overlay. The result overlay should explain, respectively: baseline under the same setup; aerosols can provide surfaces where water vapor condenses; and higher aerosol loading can change droplet number and size depending on atmospheric conditions. Include three explanatory cards covering atmospheric effects, water-cycle connection, and biotic/abiotic effects, plus the warning “LAB MODEL ≠ DIRECT REAL-WORLD MEASUREMENT.” Provide a button to enter results in the notebook.

The notebook is a large cyan-bordered overlay titled SCIENTIST NOTEBOOK. It explains the independent variable as aerosol particle condition (0 No Aerosol, 1 Aerosol, 2 Higher Aerosol), and the dependent variables as cloud density, cloud duration, condensation/precipitation response, and LUX/light transmission. Explicitly warn that the visible cloud is condensed water droplets, not PM2.5; LUX measures light intensity, not PM2.5 concentration; cloud formation is an observable condensation indicator; and the experiment does not directly measure rainfall. Provide an editable table with rows No Aerosol, Aerosol, and Higher Aerosol and columns Condition, Trial, LUX, Cloud Duration (s), and Density /5. Add SAVE DATA & CALCULATE, OPEN CENTRAL KALIMANTAN EVIDENCE, and OPEN EARTH-SYSTEM BOARD buttons. Saving computes/display averages for LUX, cloud duration, and cloud density plus a slope calculated from Higher minus No Aerosol divided by two. Draw a dark graph titled “AVERAGE CLOUD DENSITY / 5” with three labeled points/conditions and provide a textarea where the player can type an interpretation. Warn not to treat the lab model as proof of a real-world precipitation relationship.

The Central Kalimantan evidence screen is a research-context overlay, not a fabricated data simulator. Include cards about the 2015 fire season and NASA-observed smoke, Palangkaraya aerosol optical-depth/AERONET observations, and the fact that comparable reliable annual PM2.5 data is unavailable for years without suitable measurements. Show a 2015–2026 wildfire-hotspot graph whose annual values are intentionally blank and prominently state “NO COMPARABLE RELIABLE ANNUAL DATA AVAILABLE.” Explain that satellite hotspots are possible active-fire detections and are not automatically individual fire counts or PM2.5. Cite the displayed source categories: NASA FIRMS, NASA Earth Observatory, AERONET, and BMKG.

The Earth-System Connection Board should present clickable nodes: Human land use, Fire risk, Wildfire, Atmospheric aerosols, Cloud processes, Possible precipitation changes, Water availability, and Vegetation / ecosystems. Each click appends that node to a visible chain in the order selected. Include a caution that atmospheric effects depend on moisture, temperature, cloud type, aerosol properties, and other conditions. Preserve the game’s educationally careful language and do not add combat, enemies, health, scoring, inventory capacity, or direct PM2.5/rainfall measurement.