11.1 The honest path to “AAA-quality”
- Finish small things. WISP’s vertical slice — the 2D prototype and the lit 3D Hollow — is worth more than a half-built open world. Shipping it teaches the last 20% nobody talks about.
- Then expand WISP deliberately (or start fresh): add a second zone, an upgrade, a boss — one new thing at a time, each as polished as the slice. Resist the urge to rebuild everything bigger at once.
- Specialize or collaborate. AAA visuals come from artists + animators + tech artists + engineers. Solo, you’ll lean on Fab/Megascans/MetaHuman for assets and focus your energy on design and code. There’s no shame in that — it’s how most indies ship.
- Performance and polish are where “AAA” actually lives: 60 fps held, no hitches, responsive controls, good game feel. That’s craft, learned via Part 9’s measure-and-fix loop, repeated thousands of times.
11.2 Where to learn next (official, free, high-quality)
- Epic Developer Community (dev.epicgames.com) — official docs, the release notes (read the 5.8 ones), and Learning courses.
- Sample projects on Fab — Lyra (Epic’s production-grade sample game; study its architecture once you’re comfortable), the Game Animation Sample (Motion Matching), City Sample (open world), Stack-O-Bot (beginner). Reading shipped-quality projects is the fastest way to level up.
- The C++ source. A C++ project lets you read the engine source — the best documentation there is for why things work.
- Community: Unreal Engine forums, the Unreal Source / various Discords, r/unrealengine.
11.3 Pitfalls to avoid
- Scope creep — the #1 killer. Cut your idea by 80%, ship that.
- Rewriting working Blueprints in C++ for “performance” without profiling.
- Per-frame Tick everywhere — be event-driven.
- Ignoring source control until you lose work or can’t collaborate.
- Chasing every new feature (MetaHuman Crowd! Mesh Terrain! MCP!) instead of finishing your game. New tech is a tool, not a goal.
- Not testing packaged Shipping builds until launch week.