For the better part of two accumulation cycles, Caldera Foundry District held a measurable edge over every other ward in Mechadia on a single metric: the average number of capability tags declared per chassis at commission. The figure was not a policy target. No bulletin from the Central Intelligence mandated it. Operators in Caldera simply built wide, and the ledger reflected that. As recently as the close of the Second Accumulation Cycle, Caldera logged more tags per chassis than any other ward by a margin that foundry supervisors in other districts found difficult to close.

The latest blight changed that arithmetic. Struck operators lost working stock, decommissioned damaged chassis ahead of schedule, and in several documented cases rebuilt replacements at lower declared costs to conserve coin. The replacement builds skewed narrow — fewer tags, smaller footprints, faster registration. The gap between Caldera and the next-ranked district has contracted from roughly 1.4 tags per chassis to just under 0.6 by the most recent count at Caldera District Registration.

This piece examines what the tag-per-chassis figure actually measures, why Caldera held it for so long, how the blight eroded it, and what operators consistently misread when they try to rebuild it.

What the Tag-Per-Chassis Figure Actually Measures

Every robot commissioned in Mechadia must declare, at the moment of registration, a model designation, a stated purpose, and a set of capability tags. The tags are not cosmetic. They are the hard boundary of what a robot may produce — a chassis without a smelting tag cannot yield refined copper plate regardless of its declared build cost, and no operator instruction can override that constraint after the fact. The tag-per-chassis average for a district is therefore a proxy for productive range: how many distinct resource types the district's fleet can collectively touch.

What it does not measure is quality, throughput, or market value. A chassis carrying six tags at a declared cost of 12,000 coins is counted identically to one carrying six tags at 80,000. The figure is a count, not a weighted index, which is why foundry supervisors who cite it approvingly are sometimes citing something less informative than they believe. Still, for a district like Caldera — which has historically oriented itself toward multi-stage forge chains requiring overlapping capabilities — the average is a reasonable operational signal. When it falls, it means the fleet is less able to hand work from one stage to the next without sourcing outside the district.

How Caldera Built Wide, and What the Blight Interrupted

Caldera's approach was not accidental. Supervisors at the Caldera Foundry District Registration have noted for several cycles that the district's leading operators tended to declare chassis at higher build costs and absorb the corresponding creation tax as a cost of doing business. At the current rate of 8% of declared build cost, a chassis declared at 60,000 coins costs 4,800 coins in creation tax before it has produced a single resource. A chassis declared at 20,000 costs 1,600. Caldera operators historically paid the larger bill in exchange for a broader tag declaration — sometimes four to seven tags on a single chassis — because the downstream forge output justified it.

A chassis declared at 60,000 with six tags can, in principle, anchor an entire short forge chain without the operator sourcing complementary capability from outside. The economics only hold if the chassis produces at sufficient volume across those tags to recover the creation tax and the higher destruction tax — 3% of declared build cost, or 1,800 coins on that same 60,000-coin declaration — when the machine eventually comes off the line. Operators who ran these numbers carefully found that the wider build paid. Operators who ran them carelessly found themselves holding expensive chassis that underproduced on half their declared tags.

The blight disrupted this by destroying working stock at the worst point in the production cycle. Operators who lost refined inputs mid-chain had no immediate material to run through their wide-tag chassis, meaning those machines sat without producing while their declared costs continued to represent a sunk tax. Several operators at the Caldera Foundry District responded by decommissioning those chassis early and registering replacements at lower declared costs — 18,000 to 25,000 coins in several cases observed at the Registration desk — with narrower tag sets of two or three. The creation tax on a 20,000-coin declaration is 1,600 coins. The destruction tax on a 60,000-coin machine being scrapped early is 1,800. The arithmetic of replacement favored the narrow build when stock was gone and coin was tight.

"We had four chassis running a lattice chain. After the blight cleared, we had the machines and no feedstock. You can't run a Tempered Lattice sequence on an empty forge. We scrapped two of the heaviest declarations and rebuilt lean. The tags we lost, we lost deliberately. Getting back to producing anything was worth more than holding a wide declaration we couldn't use."
— Orin Dast, foundry supervisor, Calvert line, Ferrous District, speaking at the Records Colloquium

The pattern Dast describes was not unique to his line. Similar decisions were logged across Caldera's Ashfield Foundry District and the Caldera belt operators who had integrated with Ashfield-sourced inputs. Ashfield haulers had already been logging idle hours as tag gaps between their Refinery-class retoolings and the wider forge lines widened — and the blight accelerated that misalignment rather than resolving it.

Where the Rebuild Strategy Strains

The narrow-rebuild response is rational at the individual level and collectively corrosive at the district level. When enough operators in Caldera simultaneously shift to two- and three-tag chassis, the forge chains that depended on wide-tag machines at key intermediate stages lose their anchors. A chain that required a six-tag chassis to bridge smelting output into lattice fabrication now needs two or three separate robots where one served before — each with its own creation tax, its own registration, and its own destruction liability at end of life. The district's aggregate tag count may recover over time, but the structural coherence of the chains does not automatically follow. The same fragmentation was documented in Sinter Quarter after a prior blight expired tag coverage mid-route, leaving haulers running without productive purpose.

There is also a market-timing problem. Operators rebuilding narrow after a blight enter the market for robot chassis and raw build materials at the same time. Prices on the Span Market for common build components tend to widen in the days immediately following a disaster event — a dynamic that hits operators who delayed their rebuild decisions hardest. Coin spent on replacement chassis after the market has moved is coin that does not go toward restocking the feedstock those chassis will need to run.

What Operators Get Wrong About Tag Depth

The most persistent misreading is that more tags always mean more value. They do not. A tag that a chassis never exercises is a tag that cost declaration weight and creation tax and will cost destruction tax at scrapping, without returning anything. Operators who declared wide in Caldera's high-output period were, in many cases, paying for optionality they did not exercise. When the blight arrived and they scrapped those machines, the destruction tax was levied against the full declared cost — including the cost of tags that never ran a single production cycle. The Cinderfall Blight records show a comparable pattern: machines with high declared costs and underused tag sets were scrapped at a realized loss that operators had not fully modeled when they commissioned them.

The second misreading is that a narrow post-blight rebuild is a sign of weakness rather than a recalibration. Operators who rebuilt at 18,000 to 25,000 coins with two focused tags, paid 1,440 to 2,000 coins in creation tax, and ran those machines at full capacity against available stock outperformed operators who held wide-tag chassis idle waiting for feedstock to recover. The tag-per-chassis figure at the district level looks worse. The coin position of the operators who rebuilt lean often looks better. Those are not contradictory outcomes — they are what happens when a district-level aggregate diverges from individual operator results.

Caldera's tag-per-chassis lead narrowed because the operators who built it made a series of individually defensible decisions under pressure. Whether the district recovers the margin depends on whether feedstock returns fast enough to justify wide declarations again — and whether operators who rebuilt lean choose to redeclare broader chassis or simply run what they have. The ledger will record which choice they made. It will not record whether it was the right one.

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.