Skip to content

Latest commit

 

History

History
76 lines (65 loc) · 3.03 KB

command_readme.md

File metadata and controls

76 lines (65 loc) · 3.03 KB

COMMAND AND QUANTIFY BETA v1.0

WebXOS 2025 Game Manual

Overview

4-player grid conquest. You (Quantum Commander, green) vs 3 CPUs. Destroy enemy Quantum Cores to win. Expand territory turn-by-turn.

Setup

  • Grid: 16x16 (switchable to 8x8).
  • Players:
    Player Color Type
    0 Green You
    1 Blue CPU Alpha (Tactical)
    2 Yellow CPU Beta (Defensive)
    3 Red CPU Gamma (Aggressive)
  • Start: Each gets Quantum Core in corner + 50 Quantum Units (QU).

Entities

Icon Name Cost Health Movable Notes
⚛️ Quantum Core 250 100 Yes Win by destroying enemies'. 25% self-defend chance.
⚔️ Barracks 100 60 No Enables Army production (up to armies ≤ barracks). 50% self-defend.
🌾 Quantum Farm FREE 40 No +10 QU/round per farm. Instant destroy.
🎯 Army 100 50 Yes Attacks adjacent. 50% win vs other Army.

Your Turn (Green Panel)

  1. Build: Click button → highlights valid spots (adjacent to your territory).
    • Farms: Free, expand safely.
    • Barracks: For armies.
    • Army: Needs barracks capacity.
  2. Move/Attack: Click your movable entity (⚛️/🎯) → highlights adjacent cells.
    • Empty: Move.
    • Enemy: Attack (auto-resolve).
  3. Auto-end: Action completes turn.

QNN Sliders (tune your AI assist):

  • Aggression: Risk-taking.
  • Memory: Learns from history.
  • Speed: Fast vs precise.

Rules

  • Expand: Build only adjacent to your entities.
  • Move: Adjacent only (1 step).
  • Combat:
    Attacker → Defender Outcome
    Any → Core Core 25% destroy attacker
    Any → Barracks 50% defend
    Any → Farm Destroyed
    Army → Army 50% win
  • Income: +10 QU/farm at round end.
  • Elimination: Core destroyed → out. Last standing wins.

CPU Turns

  • Watch panels: "QNN Calculating...".
  • They build/move/attack automatically.

Win

Destroy all 3 enemy Cores. Victory screen shows rankings.

Controls

  • Grid Mode: Dropdown (restarts).
  • Restart: "QUANTUM RESTART" button.
  • Mobile: Touch grid/entities.

QNN Mechanics:

  • Model: TF.js sequential NN (input: 10 state features; hidden: 16 ReLU → 8 ReLU; output: 4 softmax actions: farm/barracks/army/move).
  • Sliders (player-tunable, CPU-fixed profiles):
    Param Effect Player Default CPU Examples
    Aggression Risk/attack bias 50% Alpha:65%, Beta:35%, Gamma:85%
    Memory Depth History learning 50% Alpha:75%, Beta:85%, Gamma:30%
    Speed/Accuracy Fast vs precise 50% Alpha:45%, Beta:60%, Gamma:70%
  • Beta v1.0: Model inits + dummy predict; sliders update params (unused). CPU: rule-based (attack cores → farms → barracks → armies → random).
  • Training: Game JSON (entities/moves) feeds lightweight NN for spatial agents (drones/chess). Browser sims evolve policy via epochs/loss.