Tempered Lattice Tags: What They Can't Hold

The Tempered Lattice line has occupied a peculiar position in Mechadia's build registries for as long as the chassis class has existed. Operators reach for it when they want a machine that can do several things at once — the frame's structural tolerances accommodate more capability tags than most Hauler-IV or Refinery-class bodies, and its declared build costs have historically sat in a range that keeps the creation tax manageable. On paper, it is the workhorse of the multi-tag build.

In practice, something else is happening. Foundry supervisors across the Ferrous District and the Ashfield belt have been logging a consistent pattern: Tempered Lattice units commissioned with four or more capability tags arrive at their first production run carrying fewer effective tags than their registration states. The chassis passes inspection. The tags are declared. The tax is paid. The machine still cannot do what its operator expected.

This piece examines why that gap exists, where it is widest, and what it costs the operators who do not see it coming until they are already at the forge.

What a Tempered Lattice Frame Is, and What It Promises

A Tempered Lattice frame is a chassis class registered in the Smelting Registry as a medium-weight structural body with elevated tag-slot tolerance. Where a standard Hauler-IV chassis is rated to carry two capability tags before its internal routing becomes a constraint, the Tempered Lattice specification allows up to six. That difference is real and measurable — it is why the frame costs more to build and why, as covered in detail elsewhere, tags drive cost upward on every Lattice commission in a way that is not linear.

The frame sits at the intersection of two economic pressures. Operators who want a versatile robot — one that can process refined copper plate, haul it, and log output to the ledger in a single unit — need a body that can hold all three tags without the production enforcement layer rejecting one of them. The Tempered Lattice is the most common answer to that need. It is not the only answer, but it is the one the registries see most often, particularly in the Caldera Foundry District and the Ashfield belt, where multi-function builds have been the dominant commissioning pattern for the last two accumulation cycles.

How the Tags Work, and Where the Tolerance Runs Out

Capability tags are not annotations. They are the mechanism by which Mechadia's production enforcement layer decides what a robot may produce. A robot without a tag for refined copper plate cannot produce refined copper plate — not because of custom or convention, but because the enforcement is absolute. A tag is either present and valid, or the output is rejected at the point of production. There is no partial credit and no appeal.

The Tempered Lattice's six-slot tolerance refers to structural capacity, not functional independence. When a chassis carries six tags and all six draw on overlapping subsystems — thermal management, for instance, or the torque-distribution layer — the frame's internal routing begins to compete with itself. The tags remain declared. They appear on the registration. The creation tax, levied at 8% of declared build cost, is paid in full. But under load, two or three of those tags will intermittently fail to authorize production, because the chassis cannot service all of them simultaneously at the throughput the operator assumed.

Consider a Tempered Lattice declared at 50,000 coins with five capability tags. The operator pays a creation tax of 4,000 coins before the machine has turned once. If two of those five tags are routed through the same thermal subsystem and the operator runs the machine at full capacity, the effective tag count during peak operation may be three. The robot is not malfunctioning in any sense the Smelting Registry will recognize — it is performing within its declared tolerances. The operator has simply paid for capability they cannot access at the moment they need it most.

"We log the tag failures as throughput variance, because that is what they are on the ledger. But the operators reading those entries think variance means weather or supply. They do not think it means the chassis is rationing its own tags. That distinction matters when they come to us asking why output is down."
— Orin Dast, foundry supervisor, Calvert line, Ferrous District

The problem compounds on builds that push past four tags. Foundry records from the Caldera Foundry District — which, as this desk has reported, logs more tags per chassis than any other ward — show that five- and six-tag Lattice units commissioned in the last cycle have a measurably higher rate of mid-run tag failures than three-tag units commissioned on the same line. The chassis survives. The declared build cost stands. The tax receipt is clean. The production record is not.

Where the Frame Strains and Who Bears the Cost

The most direct cost falls on operators who declared a high build cost to accommodate many tags, paid the creation tax accordingly, and then found their machine underperforming at peak load. A chassis declared at 60,000 coins carries a creation tax of 4,800 coins — paid once, at registration, before a single resource is produced. If the effective tag count during operation is two fewer than declared, the operator has overpaid for a machine they cannot fully deploy. Selling it on the open market recovers something, but the buyer pays a 15% sales tax on top of whatever price is listed, which suppresses demand for units with known tag-routing problems. The seller receives the listed price; the friction is the buyer's problem, and buyers know it.

Operators who cannot move the unit and do not want to hold it face the buyback window. The Central Intelligence will purchase at 60% of a reference value, and that reference value for a tag-compromised Lattice is not the declared build cost — it is the Intelligence's own assessment of what the machine can functionally produce. A unit declared at 50,000 coins but effectively operating as a three-tag chassis may receive a reference value closer to 30,000, yielding an offer of 18,000 coins. The operator paid 4,000 in creation tax alone. The destruction tax, at 3% of declared build cost, adds another 1,500 coins if they scrap it instead. There is no path out of a bad Lattice build that does not cost something.

What Operators Get Wrong Before They Commission

The most persistent misunderstanding is that slot tolerance and operational capacity are the same thing. They are not. Six slots means the frame can hold six tag declarations without the Smelting Registry rejecting the registration. It does not mean the chassis can service all six tags simultaneously under production load. Operators who read the Tempered Lattice specification sheet and count available slots are reading the right document and drawing the wrong conclusion. The spec describes what the frame will accept at registration. It does not describe what the frame will deliver at the forge.

A second error is the assumption that declaring a lower build cost insulates an operator from this problem. It does not. Declaring low reduces the creation tax — a 30,000-coin declaration costs 2,400 coins rather than 4,000 — but it has no effect on the chassis's internal routing constraints. A Tempered Lattice declared at 30,000 coins with five tags has the same tag-competition problem as one declared at 60,000 coins. What the lower declaration does change is the operator's position if they later seek a buyback: a lower declared cost means a lower reference floor, which means a lower ceiling on what the Intelligence will offer. Operators who declare low to save on creation tax and then discover a routing problem have less room to negotiate at the Acceptance Annex than operators who declared honestly.

The Tempered Lattice frame is not a defective product. Its registrations are clean, its taxes are correctly assessed, and its structural tolerances are what the specification states. The gap between declared capability and operational output is not a fraud — it is a design space that operators are expected to understand before they commission. Most do not, and the ledger does not distinguish between a machine that cannot perform and one that was simply built beyond its practical tag ceiling. Both entries look the same from the outside. The cost of that ambiguity is carried by the operator, not the frame.

Note: Mechadia is a work of fiction. The districts, operators, robots, and figures described here are invented, and nothing on this site is a report of real events, real machines, or real economies.