Devlog Archive
[R1] BUILD Build · v0.4.0

Internal build · jam-mode playtest scene

Isolated playtest scene with a 3-day terraforming loop, fuel-gated machines, and the new build-mode reticle. Available to Patrons.

Built-out outpost with wired power and an Air Purifier going down — Day 2.

A small build for Patrons — v0.4.0 — drops today. Not a vertical slice, not a demo for a campaign. It’s a focused playtest scene I built to put the new systems in front of real hands.

What’s in this build

A single isolated scene called _JamPlaytest.unity, scoped to a 3-day terraforming loop. Each day has a target percentage you have to hit before night closes the loop:

  • Day 1 — Heat the planet to 30%. Place power generators, then heat generators.
  • Day 2 — Pressure climbs to 55% + first Oxygenation. Atmospheric machines come online. Power network has to actually carry load by now.
  • Day 3 — Restoration to 100%. Beacon goes up. Cinematic plays.

The scene is intentionally bare — no biomes, no mob pressure, no story. It’s a wind-tunnel for the build-mode UX and the atmospheric pacing.

Build mode placing power poles — blue ghosts are valid placements, orange are blocked.

What I’m trying to learn

Three things, in order:

  1. Does the build-mode reticle feel right? It’s a new component (BuildModeReticle.cs) that surfaces “where is your placement going to land” without losing the top-down framing. The previous version used a ghost mesh hovering at cursor; this one adds a small ground-projected reticle that’s readable even when buildings stack.
  2. Are the fuel-gated machines fun, or do they just feel like a chore? Power generators consume fuel from a shared pool; if the pool empties, the network goes dark. There’s a FuelConsumer.cs per machine and a JamFuelHUD.cs showing the budget at the top of the screen.
  3. Does the 3-day rhythm work? I’m specifically testing whether 20 minutes of session time with two clear “punctuation marks” (Day 1 success, Day 2 success) is more satisfying than the 5-minute moment-to-moment of an extraction game with no pacing layer.

Known issues

  • Wires don’t always render the right color when a generator runs out of fuel mid-frame. Cosmetic, not functional — the network state itself is correct.
  • The Beacon broadcast cinematic is placeholder. It’s a fade + text card right now. The R8 work has the multi-stage broadcast designed but not fully wired.
  • Saves are scene-local. Quitting the playtest scene drops the run state. Saves come in a future build.

How to grab it

Patreon supporters at the build tier and above get the v0.4.0 zip in the post locked to that tier. PC-only for now (Windows + Mac), no Linux build until I can test on a real Linux box.

How to give feedback

There’s a Tally form linked from the build’s main menu. Three questions:

  1. What did you spend the most time on in Day 1? (looking for friction)
  2. Did Day 2’s pressure feel like a step up or a wall?
  3. Anything that felt fake / placeholder you’d want polished first?

Three minutes total. The data shapes what gets prioritised in R7 and R10.

What this build doesn’t tell you

It doesn’t tell you what the real game feels like. It’s a wind tunnel. There’s no biome restoration, no apex fauna, no co-op, no narrative beats around the Beacon. All of that lives in the rest of the roadmap.

But if the rhythm in this scene clicks, the rest of the architecture is already lined up to extend it. If the rhythm doesn’t click, that’s an R7-or-earlier problem and I’d rather know now.

If you find a bug — or just have a strong opinion — drop a note in the form or grab the next devlog. Both work.