I’m roughly 180 hours into serious Unity development. This is a status report at that mark: where the hours went, where each project stands, and what comes next.
Where the hours went
The bulk of the 180 went into games — jam entries and a steady stream of prototypes across more genres than I’ll admit to. The library I maintain on the Asset Store took about 40 hours of the total.
That ratio surprises people who find me through the library, so here is how those 40 hours happened: there is no library roadmap. A game needs something, I implement exactly that much, release it, and go back to the game. Every feature in Reactive SO traces back to one specific game that demanded it:
- GPU Sync came from a demo with 10,000 particles where I refused to write bridge code between gameplay state and shaders
- Traits (bitmask tags like Cursed / Poisoned / Shielded) came from a status-board demo — writeup still in my drafts folder
- Snapshot and timeline branching came from a tiny strategy sim with a year seekbar — same drafts folder
- The InputActionAsset scene-transition fix came from a jam battler where input died after leaving the title screen
The feature list is really a list of games. The 40 hours are what those games left behind.
What’s shipped
Things that exist outside my machine at this mark — both games came out of Unity 1 Week Game Jam rounds:
- Two jam games on unityroom — INVEST, a short bullet-hell, and Scoop & Cheers Battler
- Reactive SO on the Asset Store, plus WebPublishTools as a Git-URL package for unityroom-ready WebGL builds
- A public Godot sandbox from a two-week prototype sprint, whose persistence and interaction patterns fed back into the Unity projects
Where the library stands
Close to necessary-and-sufficient for the games I make. The part I’d defend hardest is the monitoring side: even if a future project pulled me onto DOTS, I’d expect the monitoring system to hold up as-is.
Practically, that means I don’t expect the 40 hours to grow much. The library gets a line in the next snapshot only if a game forces one.
What’s in progress
Four things are active right now.
Rendering experiments in my public sandbox — dither transparency in two shader flavors, and a comparison rig for URP’s four rendering paths:
PostDither transparency two ways in hand-written HLSL and Shader GraphA camera-blocking wall should fade, not vanish. I built the same dither transparency effect twice in my Unity sandbox — once in HLSL, once in Shader Graph — and the circular-hole mode is where the two approaches stopped being equivalent.The voxel terrain prototype — the Marching Cubes digger and its editor tooling are the part of the project I open most often:
PostFixing mesh holes in my Marching Cubes voxel terrainI ported a Marching Cubes voxel digger from one Unity prototype into the next and watched it grow holes whenever the player dug at an off-grid height. Here's the algorithm swap that closed them, plus the editor workflow I built to iterate on the fix.AI-related experiments in a fourth sandbox that hasn’t left my machine yet. Too early to say anything useful.
And Blender, for the first time. I modeled a low-poly barrel and put it into my own game. Seeing a mesh I made myself in my own scene was more motivating than the asset deserved.
What’s next
Two priorities.
The multiplayer prototype gets its first real playtest — friends, not me alternating between two editor windows. The server-authoritative rework made it something I can hand to other people, and the payload lessons from the wave-3 MTU postmortem are already folded into its wave config:
PostMigrating damage resolution from client to server authority in Unity NGOA co-op survival prototype carried over a client-trusted damage ServerRpc from a previous project. Notes from moving hit detection, projectile flight, and AoE onto the server, plus the unrelated bugs that surfaced along the way.And the voxel prototype is aiming at an actual release: a simple digging game, probably on itch.io.
What I’d tell past me
Ship the jam game even when the architecture is ugly. Extract a library feature only after the same pain hits twice. And the new one from this stretch: art tools are worth trying, but set a time limit before you start — they don’t have a “done” signal the way code does, and an evening disappears fast in there.
Wrap-up
The next snapshot should have a playtest report and an itch.io link in it. That’s the bar this one sets.