Skip to content

Building an Unreal Engine 5 FPS in 5 Days ​

Unreal Engine 5, Blueprint

With zero prior Unreal Engine experience, this FPS was built from scratch in five days. Time was tight and development was chaotic, but it was a great challenge.

Building an Unreal Engine 5 FPS in 5 Days gameplay screenshot

What was implemented ​

  • Animations & state machine — Idle, Walk, Run, Jump, Fall, and Land states via Blend Space, with a State Alias so Fall/Land transitions trigger automatically based on the Sequence Player rule. Sprint (Shift, reverts to walking on release) and jump (Space) mapped to input.
  • Environment & lighting — A dark, moody, crimson map, with a Post-Process Volume to make emissive materials and glowing elements pop against the darkness.
  • Map flow — Loading Scene → Video Screen → Main Menu → Lab (Level 1) → Egypt (Level 2).
  • Enemy AI (the Ant) — A light source attached to the mesh for visibility in dark maps; blood-splatter particles and spatial audio with randomized pitch on hit; the enemy speeds up when it spots the player or takes damage; on collision it explodes and knocks the player back via Launch Character with randomized force. Health shows on a billboarded World Space UI bar that always faces the camera.
  • Interactive items — Point lights and particle effects on pickups; ammo pickups float up and down via InterpToMovement; health pickups get their own custom movement.
  • Player mechanics & audio — A toggleable flashlight bound to F with a clicking sound; custom Anim Notifies for footsteps (avoiding glitches like a footstep looping mid-air after a jump), blended with a Sound Modulator and random pitch for variation; randomized pitch on rifle fire, dry fire, and background music via Sound Cues.

Cut for time ​

Advanced animation blending for mid-run jumps/landings, landing sound effects, destructible barrels, and physics-based hazards like swaying bridges — backlogged for a future pass.

Retrospective ​

No time went into clean architecture or optimization — the focus was cramming in as many features as possible to see what worked, and it was a lot of fun. Starting late on a Sunday night cost some time that could've gone toward a wider variety of item drops and enemy types, and the recording came out lower quality than expected — worth fixing before the next video devlog.

GitHub · itch.io · Video