Concept artworkClassic Survival
Aim. Move. Hold the ridge.
A wave-based survival loop for testing animated characters, combat feedback, room state, and direct browser controls.
- Primary experiment
- Mouse + keyboard
- Fallbacks
- Gamepad support
Current Lab · Private Alpha
ZombieRidge explores play beyond the traditional controller.
A browser-native 3D prototype lab combining animated Three.js characters, three game variants, and mode-specific keyboard, gamepad, touch, hand, face, and head-tracking inputs.
The current focus is controllerless interaction, AI-assisted game building, companion input, and bounded small-room multiplayer experiments on AWS. Zombie Runner is currently a local course.
- Prototype modes
- Three playable variants
- Rendering
- Three.js in the browser
- Coordination
- AWS-backed room experiments
Prototype Visual Direction
Current builds include Classic Survival, Zombie Slapper, and Zombie Runner. The workshop scaffold and broader public playtest format are still in preparation.
How It Plays
Three prototypes. One controllerless question.
Familiar controls remain available. Each mode gives a different movement or gesture another chance to become part of the game instead of treating the controller as the only valid input surface.
Concept artwork
Concept artworkZombie Slapper
Turn a hand gesture into impact.
A close-range mode for testing optional webcam hand tracking, alternate camera control, and immediate physical feedback.
- Primary experiment
- Browser hand tracking
- Fallbacks
- Keyboard, mouse, touch, or gamepad
Concept artworkZombie Runner
Steer, jump, and dive with movement.
A local movement-first escape course where camera-observed head position can drive steering and actions without removing familiar fallbacks.
- Primary experiment
- Face + head tracking
- Fallbacks
- Touch, keyboard, or gamepad
From Prototype to Game Camp
Build one controllerless mechanic. Deploy it. Play it together.
We are shaping a focused 2–3 hour pilot around a prepared workshop scaffold: inspect the current game, tune one interaction with coding agents, deploy to AWS, and finish with a bounded small-room playtest.
Review the Pilot Direction01
Browser game
Next.js, Three.js, React Three Fiber, animated characters, and browser physics.
02
Identity + rooms
AWS Amplify Gen 2, Amazon Cognito, Amplify Data/AppSync, and DynamoDB.
03
Companion input
AppSync Events carries bounded control signals; raw camera pixels stay on the device.
Prototype Status
Separate what works now from what the pilot is meant to prove.
ZombieRidge is an active prototype lab, not a finished game platform. That boundary makes it useful for hands-on architecture and interaction work: attendees can see the decisions that still need to be made.
Current
Working prototype
Three game variants, authenticated rooms, browser game entry, animated 3D characters, and mode-specific controllerless inputs are implemented.
Validation
Device + room acceptance
Real hand, head, controller, companion-phone, and signed-in multiplayer paths still need a final structured human acceptance pass.
Proposed Next
Controllerless Game Camp
Prepare a clean workshop scaffold, define AWS lab guardrails, and validate the format with a small community pilot before a larger engagement.
Questions about access, camera data, workshops, or architecture?
The FAQ separates implemented behavior from proposals and open acceptance work.
AWS Architecture
Browser-native play with an AWS-backed coordination layer.
Rendering, physics, and camera interpretation stay close to the player. AWS provides identity, durable room state, and bounded companion-input transport for the current code-gated room experiments.
The architecture question is explicit: when concurrency, simulation authority, anti-cheat, or operational scale exceeds this model, evaluate Amazon GameLift Servers or another authoritative-server design instead of stretching AppSync beyond its role.
01
Browser game
Next.js, Three.js, React Three Fiber, animated characters, and browser physics.
02
Identity + rooms
AWS Amplify Gen 2, Amazon Cognito, Amplify Data/AppSync, and DynamoDB.
03
Companion input
AppSync Events carries bounded control signals; raw camera pixels stay on the device.
Controllerless Direction
Camera input is optional, local, and paired with familiar fallbacks.
Hand and face interpretation runs in the browser. The companion path publishes normalized gestures and control fields rather than raw camera pixels, while keyboard, mouse, touch, and controller inputs remain available.
Proposed Community Pilot
Controllerless Game Camp
A proposed 2–3 hour hands-on lab for game builders, designers, and technical creators. The first session would be intentionally small so the scaffold, AWS cost guardrails, room flow, and facilitation model can be tested before any larger engagement.
AWS participation, venue, credits, branding, and co-marketing are discussion points—not confirmed event commitments.
- 01
Play the current prototype and trace one mechanic through the architecture.
- 02
Choose a controllerless input and tune one focused interaction.
- 03
Build with Kiro Crew, Codex, or another coding-agent workflow.
- 04
Deploy to AWS, connect a companion device or room, and run a small playtest.
- 05
Review cost, synchronization, simulation authority, and the next scaling boundary.
ZombieRidge Collaboration
Talk about the prototype or a Controllerless Game Camp pilot.
A short note is enough. Share whether you are exploring a private demo, a community lab, game-development collaboration, or the AWS architecture boundary.
Private prototype demo
Review the three game variants, controllerless inputs, room flow, and current maturity boundary.
Hands-on community lab
Shape a bounded 2–3 hour pilot around one mechanic, AWS deployment, and a small-room playtest.
Architecture working session
Examine companion input, AppSync-backed rooms, cost controls, and the authoritative-server boundary.
Private Demo or Pilot
What would you like to explore?
No account is required. CloudRaven will review the note and follow up directly.