How planet-specific production works
Space Age recipes look like ordinary recipes with unfamiliar ingredients. They are not, and treating them that way produces plans that are wrong by a factor of three before you notice.
Speciality buildings are not assemblers
Each planet's signature recipes run in a building that exists only there, and those buildings are considerably faster than a general assembler.
| Planet | Building | Speed | Module slots | Recipes |
|---|---|---|---|---|
| Vulcanus | foundry | 4 | 4 | 20 |
| Fulgora | electromagnetic plant | 2 | 5 | 11 |
| Gleba | biochamber | 2 | 4 | 23 |
| Aquilo | cryogenic plant | 2 | 8 | 15 |
The size of the error is easy to show. Casting 20 iron plates per second on Vulcanus needs 8.0 foundries. A planner that assumed a tier 3 assembler at speed 1.25 would report 25.6 — over three times as many buildings, and a power budget to match.
The question that actually matters: what has to be shipped in
Machine counts are the easy half. The half that decides whether a base is viable is which inputs the planet cannot make. Rocket capacity is finite and continuous demand is expensive, so a recipe needing one imported ingredient at a high rate can be less practical than one needing three at low rates.
Our planner splits every input into locally produced and imported, using each planet's own resources and recipe set rather than a hand-written list. That split is the reason the tool exists: a general ratio calculator will happily expand a chain across planets that cannot be built in one place.
What each planet actually depends on
The count of recipes needing imports says how dependent a planet is. What it depends on is more useful, because rocket capacity is spent on specific materials, not on a percentage. Counting across each planet's own speciality recipes:
| Planet | Speciality recipes | Most frequently imported |
|---|---|---|
| Vulcanus | 20 | lubricant (4), tungsten carbide (3), electronic circuit (1), refined concrete (1) |
| Fulgora | 11 | holmium plate (4), processing unit (3), plastic bar (3), accumulator (2) |
| Gleba | 23 | iron plate (1), electronic circuit (1), landfill (1), biter egg (1) |
| Aquilo | 15 | superconductor (5), quantum processor (5), fluoroketone cold (4), tungsten plate (4) |
Two patterns stand out. Aquilo's dependencies are almost entirely manufactured goods — superconductors, quantum processors, plates of metals it has no ore for. Nothing there can be substituted by finding a local patch; every one of them is a continuous rocket lane from a planet that makes it.
Vulcanus is the opposite shape. Its most common import is lubricant, a fluid, and beyond that the list thins out quickly. A planet that mostly needs one fluid is a different logistics problem from one that needs five different finished components, even if the recipe counts were identical.
Fulgora sits in an awkward middle. It imports holmium plate and processing units repeatedly — but holmium is something Fulgora itself produces from scrap. The dependency is on processing capacity arriving in the right form, not on the raw material being absent. That distinction does not show up in a percentage, and it changes what you build first.
Spoilage has no equivalent on Nauvis
Gleba materials decay on a timer. This breaks two habits that serve well everywhere else. Buffering stops being insurance and becomes loss, since a full chest of nutrients is a chest of spoilage in a few minutes. And belt length between production and consumption becomes a design constraint rather than a layout convenience.
The shortest shelf lives are around a minute, which is not long enough to cross a large base. Plans that work on paper can fail here for a reason a machine count does not capture: the arithmetic assumes the material is still there when it arrives.
The spread is wider than most planning accounts for. Across the 13 items that decay, shelf life runs from 1 minutes (copper bacteria) to 126 minutes (raw fish) — a factor of roughly 126. Treating them as one category is the first mistake.
The practical consequence is that three habits which serve well everywhere else become actively harmful here:
- Buffering. A full chest is normally insurance against a hiccup upstream. With a one-minute item it is a chest that will be full of spoilage before anything downstream asks for it. Size buffers by how fast they drain, not by how much they hold.
- Long belt runs. Distance is usually free once the belt is built. Here it is a tax paid in freshness on every item, and the tax rate depends on which item is on the belt.
- Batching. Producing in bursts and letting stock accumulate works when stock keeps. When it does not, a burst produces a spike of material that decays together, and the consumer sees famine between bursts rather than a steady supply.
None of this is visible in a ratio. A production line can be correctly sized, correctly fed, and still deliver nothing usable because the material aged out between the two ends. That is why the planner reports shelf life alongside machine counts rather than treating it as a footnote.
Finished goods versus fluids: two different logistics problems
Aquilo imports superconductors, quantum processors and tungsten plates — manufactured components that each require a full production chain elsewhere. Vulcanus imports lubricant, a single fluid. The import count alone makes Aquilo look four times as dependent, but the more important difference is what the dependency demands of the rocket network.
A fluid travels in barrels. One rocket carries a finite number of barrels, each holding a fixed volume of fluid. A continuous demand for lubricant at a foundry means a steady stream of barrel rockets that must be unbarrelled on arrival. The logistics are repetitive and predictable: one fluid, one recipe, one rate. Building the lubricant supply is a question of throughput, not of variety.
Aquilo's demands are the opposite. Superconductors need holmium plates, copper plates, plastic and light oil. Quantum processors need a separate chain of their own. Each imported component represents a production line on another planet that must be built, powered and maintained before a single item reaches the rocket. The rocket capacity is spent on items that already consumed resources elsewhere; the cost per rocket is not just the slot but the entire upstream factory.
This is why "number of imports" is a misleading metric on its own. A planet that needs one fluid at a high rate has a logistics problem (run enough barrel rockets). A planet that needs five different manufactured items at low rates has a capacity problem (build five off-planet production lines). The first is solved with more rockets. The second is solved with more factory, and the factory has to exist before the planet can scale.
How spoilage shapes Gleba's layout
The spoilage list spans a 126-fold range, from copper bacteria at one minute to raw fish at over two hours. A production line on Gleba cannot be laid out the way a Nauvis line would be, because distance between producer and consumer consumes shelf life. The layout has to be planned around how long each material lasts.
The shortest-lived materials — copper and iron bacteria at 60 seconds — set the tightest constraint. A belt crossing a large base can take longer than that end to end, especially if it weaves around obstacles. The practical rule is to place bacteria consumers adjacent to producers, with no buffer chest between them. A chest of bacteria is a chest of ore in 60 seconds, and the ore it becomes is not what the downstream biochamber expects.
| Material | Shelf life | Spoils to | Layout implication |
|---|---|---|---|
| Copper / iron bacteria | 1 minute | Copper / iron ore | Producer adjacent to consumer; no buffer |
| Yumako mash | 3 minutes | Spoilage | Short belt runs only |
| Jelly | 4 minutes | Spoilage | Compact block layout |
| Nutrients | 5 minutes | Spoilage | Can cross a small base |
| Agricultural science pack | 60 minutes | Spoilage | Can be buffered modestly |
| Bioflux | 120 minutes | Spoilage | Approaches non-spoiling for layout purposes |
The medium-life materials — yumako mash at three minutes, jelly at four, nutrients at five — allow short belt runs but not a bus that crosses the base. These are the materials that determine the overall footprint of a Gleba block: every machine that touches them has to be within roughly a minute's belt travel of every other. That constraint is tight enough to force a compact, stacked layout rather than the spread-out designs that work on Nauvis.
At the long end, agricultural science packs last 60 minutes and bioflux lasts 120. For these, buffering works the way it does on Nauvis: a chest of science packs bought an hour of shelf life, which is long enough to absorb production hiccups. The mistake is treating all Gleba materials as if they shared that property. A buffer sized for science packs is catastrophic for bacteria.
There is also a conversion loop worth designing around. Several materials spoil into spoilage, and spoilage can be converted back into nutrients. A line that produces yumako mash and does not consume all of it does not lose the excess; the mash spoils to spoilage, which feeds nutrients, which feeds biochambers. Building this loop deliberately — routing spoilage from short-life materials back into nutrient production — turns spoilage from pure loss into a secondary input stream. It does not eliminate the timing constraint, but it reduces the cost of getting the layout slightly wrong.
Fulgora's scrap loop: a dependency that is not a dependency
Fulgora's import profile lists holmium plates as the most frequent import, appearing in four of its eleven speciality recipes. But Fulgora produces holmium plates from scrap using its electromagnetic plant. The material is present on the planet; what is imported is not the raw holmium but the processing capacity to turn scrap into plates at the rate the recipes demand.
This is a different kind of constraint from Aquilo's. A planet that lacks a material cannot solve the problem by building more machines; the material has to arrive. Fulgora can solve its holmium plate shortage by building more electromagnetic plants and feeding them more scrap, as long as the scrap supply holds. The bottleneck is the recycling chain, not a rocket slot.
The practical build order reflects this. On Aquilo, the first priority is establishing the rocket lanes for components that cannot be made locally. On Fulgora, the first priority is building out the scrap-to-plate recycling chain so that local production can feed the speciality recipes. A general ratio calculator that lists holmium plates as an "import" without noting that the planet can produce them from scrap would recommend shipping in a material that is already underfoot.
Module slots and why the planet buildings diverge
The four speciality buildings differ not just in speed but in module capacity, and the slot count changes how each planet's production scales with beacons and modules.
| Planet | Building | Speed | Module slots | Scaling character |
|---|---|---|---|---|
| Vulcanus | Foundry | 4 | 4 | Fast and module-friendly; beacons compound a high base speed |
| Fulgora | Electromagnetic plant | 2 | 5 | Moderate speed, highest slot count; best beacon target |
| Gleba | Biochamber | 2 | 4 | Same slot count as foundry at half speed |
| Aquilo | Cryogenic plant | 2 | 8 | Double the slots of any other; module scaling is the whole point |
The cryogenic plant's eight module slots are the standout. A building with eight slots carrying four speed and four productivity modules behaves very differently from the same building unmodded, and the machine count for a given Aquilo recipe can drop by more than half once modules are installed. This is also why a naive machine count for Aquilo is more misleading than for other planets: the unmodded count is the worst case, not the operating case.
The foundry goes the other direction. Its base speed of 4 already makes it the fastest speciality building, and four module slots add speed on top of speed. The Vulcanus advantage is raw throughput; the Aquilo advantage is module leverage. A foundry with four speed modules is a very fast machine. A cryogenic plant with eight productivity modules is a machine that does more with less input. The two planets reward different module strategies.
Cross-planet dependencies at a glance
The four planets do not import independently. Several materials appear in multiple planets' import lists, and producing them once on the right planet can serve two destinations simultaneously.
Electronic circuits appear in both Vulcanus and Gleba import profiles, though only once each. Plastic bars appear on both Vulcanus and Fulgora. Tungsten plates appear on Aquilo and are produced on Vulcanus. The supply chains are not isolated planet-to-rocket routes; they form a network where a single production line on Nauvis or Vulcanus can feed several planets' speciality recipes.
This matters for rocket scheduling. If Vulcanus and Fulgora both need plastic bars, and both are served by rockets from Nauvis, the plastic production on Nauvis has to cover both rates simultaneously. A planet planner that looks at one planet at a time will size plastic production for each independently and arrive at a total that is the sum, which is correct, but it will not tell you that both demands can share one production line. The cross-planet view is what turns a set of independent rocket routes into a scheduled network.
The most entangled planet is Aquilo. Its superconductors and quantum processors require holmium (from Fulgora), tungsten (from Vulcanus), and processing capacity that pulls in materials from Nauvis. Building an Aquilo base is not a matter of setting up one planet; it requires the other three to be producing at scale first. The planet is the endgame of the production network, not a standalone colony.
Applies when…
- Recipes are the planet's own speciality recipes. Ordinary crafting steps still run in assemblers wherever they are built.
- The local-versus-import split reflects what a planet can produce, not what you have stockpiled there.
- Import lists are one level deep, naming direct inputs rather than the whole chain behind them.
- Figures assume normal quality equipment and no productivity modules.