The alarm didn't wake him. He'd been awake for twenty minutes already, lying on his back in the gray pre-dawn of his bedroom on Crestwood Avenue, running supply calculations in his head.
The charcoal bottleneck was the problem. It was always the charcoal bottleneck at this stage. His kiln district in the eastern quarter of Ironhold was producing at 71% efficiency because the Tier 1 worker housing he'd built in year thirty-one was too far from the kilns — an eleven-hex commute that the simulation modeled as a 16-point proximity penalty, bleeding productivity from every worker every shift. He'd placed the upgrade order four sessions ago. The Tier 2 housing was 60% constructed. When it finished, efficiency would jump to 87% and the charcoal supply would clear, which would let the iron smelters run at full capacity, which would feed the machine parts foundry, which was the single blocking dependency on the Steam Engine Assembly Hall he was trying to get online before year forty-five.
He'd placed the housing upgrade in the wrong sequence. He'd known it at the time and done it anyway because he'd needed the laborer pool for the river arterial extension and hadn't had enough workers to do both. It was a recoverable error — one of those decisions where you paid for a choice made under constraint, accepted the carrying cost, and moved past it. He'd made a note in his composition book: Don't sacrifice residential proximity for infrastructure sequencing. The commute penalty compounds across every production cycle downstream.
The alarm went off. He silenced it.
From down the hall came nothing. His mother's door was closed. He checked the time — 6:14 a.m. — and listened for the specific quality of silence that meant she was sleeping rather than absent, and identified it correctly: the faint hum of the white noise machine she ran when she got home from St. Anthony's, the one that blocked out street sound and morning light and allowed her to sleep through the hours that were not sleeping hours for anyone else. She'd come home sometime between two and three, he estimated. She had been on her feet for twelve hours. He would not knock on her door.
He got up. He was sixteen, lean in the way of someone who forgot to eat and slept at wrong hours. Dark hair that needed cutting. Pale from windows kept angled against screen glare. He moved through the apartment in the dim quiet, pulling on clothes from the chair — dark jeans, gray shirt, nothing that communicated affiliation in any direction — and went to the kitchen.
The refrigerator: half a container of leftover rice, three eggs, a nearly empty jar of peanut butter, two cans of soup on the shelf above. He scrambled two eggs and ate them over the sink. He made coffee in the French press his mother kept for her rare mornings off and poured the remainder into a travel mug that had lost its lid grip six months ago and now had to be held from the side.
He looked at the corkboard by the door. His mother had pinned her November schedule there — the shifts color-coded, the rare days off circled. Tuesday through Saturday nights. Sunday and Monday were the off days, which meant Sunday was the one morning the apartment had two people in it simultaneously. He looked at the schedule for a moment and then looked away.
He thought about his father. This was not a deliberate thought — more a background process, the kind of retrieval that happened when environmental conditions were quiet enough to let it surface. His father, David Arden, lived in Dayton now, approximately seventy miles southwest, with a woman John had never met and two children he had met once at Christmas four years ago and would not recognize on the street. The most recent text message had arrived six days ago: Hey bud. How's school going? John had read it twice and left it unread in the notification shade. Not from anger — he'd examined the anger and found it had mostly calcified into something harder and less reactive, more like a geological feature than a feeling. It was more that there was no efficient response. Any reply branched into a conversation that would require emotional accounting he didn't currently have the reserves for, and the text would still be there when he did, and it would still say the same thing.
He picked up his backpack — battered, black, one shoulder strap repaired with a zip tie — and left the apartment.
The bus took thirty-one minutes on Tuesdays and Thursdays, twenty-four on the other days, depending on whether the light at Fifth and Maplewood caught the 7:12 run or not. He had a window seat three rows back on the right side that was almost always available because it was directly over the wheel well and the vibration was unpleasant if you were running anything that required a steady typing surface, which he was not. He had the composition notebook out instead, reviewing the notes he'd made the night before.
The pages held a detailed sketch of Ironhold's eastern industrial district as it currently stood and as he intended it to stand in fifteen game-years. He'd started the sketch two weeks ago and built on it each night — adding new production buildings, adjusting the road layout, penciling in where the rail lines would run once he unlocked the Railway Survey node in the technology tree.
The city was the capital of the United Ardan Federation, and it sat in a river valley carved by the convergence of the Cresward River from the north and the Ardan tributary from the east, which was why the original settlers — the game's lore placed Ironhold's founding in the year 1622, the first permanent settlement in the valley — had chosen the site. The confluence gave the city natural water access for industry, natural defensive barriers on two sides, and the long flat floodplain to the south that had become the agricultural province that fed everything else. John had inherited this geography in the game's standard start scenario and spent forty-three years of game-time building on it.
Ironhold was not just a city. It was the production engine of the entire Federation.
In the notebook, in careful pencil, he had mapped every production chain currently active in the city. The Cresward industrial corridor along the river's western bank held his foundries and smelters — the ore came down by barge from the northeastern hill province, processed through the ore concentration facility he'd built in year thirty-five, arrived at the smelters at 94% purity, fed the iron casting halls that produced the raw structural material for every machine and building and rail segment in the Federation. Downstream from the foundries, the machine parts workshops that translated raw iron into precision components. Further south, the Steam Engine Assembly Hall under construction — the single building that would advance the Federation across the early Industrial threshold and unlock the next tier of everything: faster factories, longer rail networks, steam-assisted mining, and eventually the locomotive manufacturing that would let him bind the three provinces into an integrated economy rather than three separate resource pools connected by slow barge traffic.
He'd played Kaiserfront long enough to know that the production chain was never as simple as it appeared on the overlay. Every building was downstream from something and upstream from something else, and the chain was only as strong as its weakest dependency. The Steam Engine Assembly Hall needed machine parts. Machine parts needed processed iron. Processed iron needed quality coal. Coal came from the northeastern seams he hadn't yet fully developed because developing them required a barge expansion that required timber from the northern forest hexes that required a Lumber Processing Mill he'd deprioritized in favor of the Academy upgrade in year thirty-eight. Decisions propagated through the system in ways that took fifteen or twenty turns to manifest as problems, and by the time the problem appeared, the decision that caused it was long past revising.
He'd made a note about this, too, somewhere in the notebook. Something about how the game punished short-term thinking not immediately but eventually, and how the punishment was always proportional to how long you'd been pretending the deferred cost wasn't accumulating.
The bus reached his stop. He put the notebook away and walked the two blocks to Westfield High.
The cafeteria at Westfield High smelled the way it always smelled — industrial steam, overcooked starch, and underneath it all something faintly sour, like the ghost of every lunch period that had come before. John sat at the far end of table sixteen, the table nearest the emergency exit, the one nobody chose unless they had to. He had been choosing it for two years.
His tray held a carton of chocolate milk and a rectangle of cheese pizza that had been sitting under a heat lamp long enough to develop a slight leather quality along its edges. He hadn't touched it. His laptop was open in front of him — battered silver casing held together at one corner by a strip of electrical tape — and on its screen the city of Ironhold sprawled across its river valley in the forty-third year of John's stewardship, caught between three crises at once: the political question of what kind of nation the Federation was going to be, the industrial question of whether he could cross the era threshold before his coal reserves ran short, and the military question that had arrived without warning two game-weeks ago and had not yet resolved.
Around him the cafeteria arranged itself in its usual social geometry — athletes at the center tables, theater and chorus kids by the windows, the gradations between fully recognized and fully invisible distributed across the remaining space according to pressures so established they required no active enforcement. John had mapped it years ago with the same instinct he applied to the game. He understood the faction percentages of every cluster in the room, the policy demands each group generated, the events that moved the equilibria. He simply existed in the zone the system hadn't categorized, which was either outside it or beneath it depending on your definitional framework. He had stopped caring which.
The only other occupant of table sixteen was Priya Mehta at the far end, earbuds in, chemistry textbook open, sandwich in hand. Their arrangement was wordless and long-standing: shared space, mutual non-acknowledgment, no costs. He found it civilized.
He opened the city build view.
Ironhold's build interface rendered the city as a top-down projection of its river valley, each hex of the map representing approximately one city block, the terrain underpinning visible in the shading — the river's blue, the forested hexes in dark green at the valley edges, the flat developed land in graduated browns and grays depending on what sat on it. Roads were white lines. Districts were color-coded: residential in warm amber, commercial in pale yellow, industrial in the deep iron-gray that John had always thought looked exactly like the slag color from a real foundry photograph.
He opened the eastern district overlay.
The eastern expansion had been the project of the last eight sessions — twelve hexes of Heavy Industrial zoning extending from the existing smelter corridor into the previously undeveloped floodplain east of the Ardan tributary. He'd had to bridge the tributary first, which had cost him a civil engineering queue slot for three turns and delayed the Academy upgrade by one session, a tradeoff he'd accepted after modeling it in the notebook and concluding that the industrial expansion had a higher return on the delayed investment than one additional turn of Academy research would have produced.
The bridge was complete. The road arterials were laid — two main spines running east-west through the district, linked by a cross arterial at the midpoint, roundabouts at both junction points to prevent the traffic congestion penalty that could knock 12% off the efficiency rating of industrial buildings within three hexes of a blocked intersection. He'd learned about the traffic penalty the hard way in an earlier Federation playthrough, when a poorly planned arterial through his wool district had created a bottleneck that cascaded into a textile shortage, a consumer goods deficit, a satisfaction drop in the artisan population, and eventually a political crisis that had taken six turns and two expensive policy concessions to resolve.
He had not made that mistake again.
The water main extension was the current constraint. Industrial buildings required water — not for the workers, though worker water access was a separate metric, but for the production processes themselves: cooling water for the foundries, steam pressure for the machine halls, hydraulic power for the stamping equipment that shaped sheet iron into component profiles. He'd routed the main extension from the existing network at Waterhouse Junction, a routing that required laying fourteen segments of pipe east across three district hexes, and he was currently at segment nine. The last five segments would complete next turn, and when they did he could place the foundations for the Machine Parts Workshop and the Steam Turbine Hall simultaneously, begin their construction queues, and let them run in parallel over the next twelve turns while he attended to other priorities.
He clicked into the district planner and rotated the Steam Turbine Hall's footprint. The building was a 3x4 hex structure — large enough that its placement constrained everything around it. He'd been holding off on final placement until the pipe routing was resolved, because placing it before the water main was confirmed would have committed hexes he might have needed for the main's path correction. Now he could commit. He placed it at coordinates E7 through H10, the southern end of the district, downwind from the residential zones per the pollution model — his kiln district had taught him about prevailing winds and smoke dispersal, a lesson that had cost him an eleven-point residential satisfaction penalty for four turns before he'd identified the source and begun the long process of mitigating it.
Next to the Turbine Hall: the Machine Parts Workshop at D4 through F7. Its footprint was smaller, and it shared a road connection with the Turbine Hall through the cross arterial, which gave both buildings access to the eastern delivery hub where the barges from the northeastern province would unload ore. He checked the routing confirmation — green lines showing the logistics path from barge dock to workshop to turbine hall, the ore flow and the intermediate product flow correctly sequenced.
He placed the sewage treatment expansion at the district's eastern boundary. Downwind, downriver, isolated from residential impact by the buffer hexes he'd reserved for the purpose. Some players didn't bother with sewage routing until it became a crisis — disease outbreaks, population growth penalties, the public health satisfaction metric dropping below the threshold that triggered faction unrest. John had learned to route sewage before he needed to, the same way he learned doctrine before he needed to fight with it.
He zoomed out to the full city view.
Ironhold, year forty-three. Population: 84,000, distributed across seven residential districts of varying class composition — the old-town laborer quarters along the western riverbank, the artisan row in the central district where the craft workshops clustered, the merchant quarter north of the commercial plaza, and the three newer residential expansions east of the river that had filled in as the city grew. Housing stock was tiered from Tier 1 laborer tenements to Tier 3 artisan townhouses, with a small enclave of Tier 4 merchant villas near the government district that housed the administrative class and generated the clerical workforce the civil service buildings required.
He had a note from a session six weeks ago: Population composition is policy, not outcome. You decide who lives in Ironhold by deciding what you build and where you build it. Every housing decision is a population investment with a thirty-year return horizon.
He'd written it after realizing that his laborer surplus and artisan shortage — which was choking the machine parts output — had been caused by overbuilding Tier 1 housing in years twenty-eight through thirty-one without a corresponding investment in artisan-tier residential development, which had prevented the natural upward mobility that converted skilled laborers into artisans over time. The game modeled social mobility as a slow drift influenced by education, housing quality, healthcare access, and economic opportunity. Get it wrong, and you were recasting your population composition for the next twenty turns of game-time. Get it right, and your workforce rebalanced itself organically.
He was correcting it now. Four Tier 2 artisan residential buildings in the eastern district, placed adjacent to the new industrial zone so the commute penalty would be minimal. The workforce would follow the housing, the housing would generate the right population tier, the artisan population would grow, and the machine parts output would rise to match the Steam Turbine Hall's appetite.
He closed the city build view and opened the technology panel.
The technology tree in Kaiserfront was rendered as a branching schematic that filled two-thirds of the screen — a visual taxonomy of human industrial progress from the Colonial-Agrarian period through the late Industrial era to the mid-century threshold, each node a circle with a label and an icon, the connections between nodes drawn in lines that indicated sequence dependency: you could not research Steam Power Fundamentals without first completing Basic Mechanics, could not attempt Armored Doctrine without completing Motorized Transport, could not unlock Mass Production without the concurrent completion of both Assembly Line Theory and Industrial Chemistry.
He was currently at year forty-three, which in the Federation's technology progression placed him in the early Industrial period. The research panel showed his current focus: Steam Power Fundamentals, at 65% completion, with an estimated six turns to completion at his current Academy output rate.
He studied the tree from that node forward.
Steam Power Fundamentals unlocked five immediate downstream nodes: Steam Engine Production (enabling the Assembly Hall he was building), Railway Survey (prerequisite to rail network planning), Industrial Chemistry (one branch toward Mass Production), Mechanical Engineering (enabling precision tooling and higher-tier machine buildings), and Naval Steam (which he'd deprioritized because the Federation's geography made naval power secondary for at least twenty more years of game-time). From those five, the tree branched outward through another three tiers before reaching the mid-century threshold nodes — Motorization, Armored Doctrine, Advanced Artillery Theory — that would transform his military from musket-and-cannon to the combined arms machinery he'd been studying in the tactical layer.
He was forty game-years from those nodes. Possibly more.
He had mapped the critical path in the composition notebook, the sequence of research decisions that would get him to Armored Doctrine at the earliest possible moment without sacrificing too much economic development along the way. The path ran through Industrial Chemistry to Mass Production to Industrial Metals, then through Motorized Transport to Armored Vehicle Theory to Armored Doctrine — eight nodes, each requiring turns of research focus, each competing for Academy output against the economic development nodes he needed for the city's production chain. Every turn he spent researching military doctrine was a turn not spent researching something that made his factories more efficient. The tradeoff never resolved cleanly. It required constant rebalancing.
He had a principle he'd written down: Research is future purchasing power. Spend it on the things that unlock other things, not the things that are improvements on what you already have.
This was why he was pursuing Steam Power Fundamentals instead of the Improved Aqueducts node that would clear his current water infrastructure bottleneck faster. Improved Aqueducts would help him now. Steam Power Fundamentals would unlock Railway Survey, which would unlock Rail Network, which would halve the transit time on ore shipments from the northeastern province, which would remove the ore supply constraint on his smelters permanently, which was a more foundational fix than any single infrastructure improvement could provide.
He made a note in the margin: Three turns of reduced output now in exchange for permanently unconstrained ore flow. The math is correct. Hold the sequence.
Steam Power Fundamentals: six turns to completion. Six turns of the Academy focused there, and then he could begin the cascade.
He closed the technology panel and opened the governance interface.
The politics of the United Ardan Federation were, as of year forty-three, a carefully maintained equilibrium between four factions with incompatible desires.
The governance panel was a layered interface: on the left, the policy sliders — Tax Burden, Labor Rights, Press Freedom, Education Investment, Healthcare Access, Public Infrastructure — each running from 0 to 100, each affecting a different set of variables. In the center, the faction satisfaction gauges for Conservatives, Liberals, Socialists, and Nationalists, with a bar graph showing their relative seat count in the Federation's parliament. On the right, the National Focus timer showing the current focus — Civic Consolidation, the parliamentary representation node — with fifty-eight days remaining.
His slider settings: Tax Burden 38%, Labor Rights 64%, Press Freedom 46%, Education Investment 74%, Healthcare Access 58%, Public Infrastructure 66%.
He had arrived at these numbers through a process that resembled, he'd once thought, running a very slow, very expensive argument with four different people who each wanted the exact opposite of what the others wanted and would accept nothing less than everything.
The Conservatives held 26% of parliamentary seats. They wanted low taxes, minimal labor regulations, press restrictions, and high military spending. They were his primary funding base — the merchant and gentry classes who owned the workshops and the trade ships and the commercial buildings that generated most of the Federation's direct tax revenue. Keep them content enough and they didn't organize against his industrial policies. Push too far on labor rights and they started sponsoring parliamentary motions to block workshop construction. He'd learned this in year thirty-one, when he'd pushed Labor Rights to 72% without first building up the Liberal parliamentary seat count to offset Conservative resistance, and the resulting parliamentary deadlock had stalled four critical construction projects for eight turns while he negotiated his way out.
The Liberals held 35% of seats. They wanted political reform, press freedom, expanded education, moderate labor protections, and representative governance. They were his most reliable faction on the industrial transition — Liberals understood that education investment and research spending were good for long-term growth — but they had a press freedom demand that conflicted with his counter-intelligence policy, which required some restriction on journalism about military movements and resource locations to prevent intelligence leakage to neighboring powers. He'd set Press Freedom at 46%, which was below what Liberals wanted but above the threshold that triggered their active opposition. A careful middle position that satisfied no one completely and prevented crisis.
The Socialists held 26% of seats. They wanted housing investment, healthcare, labor rights at or above 70%, and income redistribution through the tax code. They were the faction of his laborer and artisan population — the majority of Ironhold's workforce — and their satisfaction metrics ran directly through the housing and wage systems he was managing. Every Tier 2 housing upgrade he completed moved their satisfaction needle. Every turn he delayed a clinic or a clean water access improvement, they lost two to three points. He'd set Healthcare Access at 58% because that was where he could afford to be without triggering the public health crisis event, which he'd seen trigger in a previous playthrough and which had produced a 40% productivity penalty citywide for twelve turns.
The Nationalists held 13% of seats. They were dangerous at low satisfaction and negligible at high, so he kept them fed with military spending and managed them out of any real influence. When the Veth military crisis resolved, he'd use the deterrence success as a press event to boost Nationalist satisfaction by fifteen points without spending any additional resources on military buildout, which was the kind of political efficiency he'd learned to look for.
The National Focus system was Kaiserfront's mechanism for driving long-term political direction. Each focus took a fixed number of turns to complete, consumed political attention, and produced a permanent change to one or more of the Federation's foundational parameters. He was currently completing Civic Consolidation — the first parliamentary focus node, which would formally establish a legislative body with representation for each faction, convert the current autocratic system to a constitutional model, and give all four factions a direct stake in political continuity. The tradeoff was a 12-point permanent reduction in executive authority, which limited how quickly he could pass crisis legislation.
He'd chosen it anyway. Constitutional governance made the Liberals and Socialists permanently more stable, reduced their defection risk during the industrial disruption period when wages lagged behind prices, and opened the parliamentary branch of the focus tree, which contained three long-term nodes he wanted: Universal Education (Education Investment permanently buffed), Labor Reform Act (Labor Rights locked above 60% without Conservative opposition), and Press Reform (a compromise node that gave him counter-intelligence protections while partially satisfying Liberal press freedom demands).
He had mapped the focus tree in the composition notebook too, twelve turns across, the dependencies drawn in connecting lines, the expected completion dates noted in the margins. It was, he'd once thought, a document that looked insane to anyone who didn't understand the game and immediately legible to anyone who did.
He was running a nation. In the game, it felt like running a nation: the permanent weight of distributed consequences, the decisions that couldn't be undone, the constant management of people who wanted things he couldn't fully give them. He found it more comprehensible than most of what happened at Westfield High, and he found that fact neither sad nor notable. It was simply accurate.
He saved the governance state and moved to the campaign map.
The military question was the one he'd been staring at since noon.
The campaign map showed the United Ardan Federation's territory as three provinces among nineteen on a regional map rendered in the style of an aged survey atlas — the borders drawn in colored lines, the terrain marked with illustrated hills and river symbols, the small heraldic shields indicating faction control. The Federation's home province centered on Ironhold. The northeastern hill province held the iron ore and coal deposits. The southern river province, acquired eleven years ago through diplomatic purchase, provided the agricultural surplus that funded everything else.
To the east, the Kingdom of Veth controlled six provinces, and three game-weeks ago they had moved an army.
It was a substantial force. The campaign map rendered armies as illustrated unit stacks with a troop count, and Veth's eastern army showed 8,400 men: two regiments of line infantry, one regiment of light infantry skirmishers, two cavalry squadrons, and a field artillery battery of eight guns. They had marched from the Veth provincial capital at Ardenmoor and were now encamped two provinces east of the Federation's border, which was close enough to constitute an explicit threat but far enough to preserve diplomatic deniability. Veth's king, a game AI the diplomatic panel labeled KING ALDRIC III, had not issued a formal declaration. He'd sent a trade envoy instead — the game's diplomatic intelligence system flagging this as almost certainly a probe, using the envoy's travel route to conduct road and fortification reconnaissance.
John had built his relationship with Veth carefully over eleven years of diplomatic exchanges. The current score was 61 out of 100, rated Cautious Friends — good enough to prevent casual aggression, not good enough to prevent opportunistic aggression if Veth calculated that the Federation's industrial transition made it vulnerable. Which it did. Every nation on the regional map could observe the Federation's economic condition through the trade visibility mechanic, and every nation with eyes could see the resource diversion the industrial buildout demanded.
He opened the military panel. The Federation's standing army: 4,200 men. Two regiments of line infantry quartered in the home province. One understrength light cavalry squadron patrolling the northeastern hill province. A garrison company at the southern river crossing. Against Veth's 8,400 it was, numerically, bad — two-to-one odds on a pitched field were only survivable if the terrain strongly favored the defender.
He opened the military recruitment interface. This was where the game's Victoria-style population mechanics became military arithmetic.
In Kaiserfront, units didn't recruit from an abstract manpower pool — they drew from specific population strata, and those draws had cascading effects on every other system the strata fed. Line infantry came from the laborer class: physically robust young men, trainable to musket-regiment standard in six to eight game-months, but pulling them from the labor pool reduced factory workforce at exactly the moment he needed factory output. Skirmishers came from the artisan class — skilled, expensive, slow to train, with production implications in the craft workshops. Cavalry came from the merchant and gentry classes, who had the horses already. Artillery recruited from artisans with Ordnance Workshop prerequisites he hadn't yet completed.
He ran the numbers in the composition notebook.
1x Line Infantry Regiment: 600 men, laborer class, 7 months training, 340 gold + 18/month upkeep. Production impact: -4% labor-intensive buildings.
1x Skirmisher Company: 280 men, artisan class, 9 months training, 210 gold + 14/month upkeep. Production impact: -2.8% craft buildings.
1x Light Cavalry Squadron: 180 men, merchant class, 6 months training, 280 gold + 22/month upkeep. Production impact: negligible.
Total new force: 1,060 men added, bringing the Federation to 5,260. Still outnumbered, but meaningfully less so, and his troops would be on home terrain with the defensive advantage of the Cresward River geography.
He thought about the terrain. The Cresward had three fords. The northern crossing ran through broken ground with high bluffs on the western bank — a natural defensive position that turned a crossing into a nightmare for an attacker deploying from column to line under fire. The central crossing was open on both sides but exposed to fire from the southern extension of the bluff line. The southern crossing was flat, clean, ideal for large-force deployment in good order — which meant it was the most dangerous to leave undefended and the most likely approach for a professional army under a general who read terrain correctly.
He had a garrison company at the southern ford. He needed a regiment there.
Standard march time from the home province: six game-days. Veth was still two provinces east, staging at Varenk. He had the margin if he moved now.
He ordered the infantry regiment's march. Six days, standard pace, arrive at the southern bluff at 91% readiness.
He was calculating approach angles when someone pulled out the chair two seats to his left and sat down heavily.
Marcus Holt was a broad-shouldered seventeen-year-old who had understood early that size was a form of argument. Sandy hair, pale gray-blue eyes with a look of mild permanent amusement. He sat two chairs over rather than directly across, which was its own positioning choice — the kind of deliberate terrain use that John, in a different context, would have noted as tactically aware. Derek Paine materialized at the table's edge. Sully Marsh stationed himself behind Marcus and right, large and unhurried. Choreography, two years refined.
"City game again," Marcus said.
John minimized Kaiserfront. A reflex.
His desktop background appeared: the 1847 Birmingham industrial survey map, factory districts color-coded by output.
Marcus looked at it. "Another game?"
"A map."
"Of where?"
"Old city. Industrial survey. Different industries marked in different colors." John kept his voice flat, affect-free. "Foundries, textile mills, rail yards."
Marcus reached over and picked up the pizza and took a bite from the far end, looking at John while he chewed. "Your game. That's a city one."
"City-building. Regional management." John paused, measuring the calculation briefly. "Military layer too. Campaign strategy and battlefield command. Diplomacy, politics, technology. There's a tactical simulation built into it."
Marcus looked at him. "Tactical like —"
"Combined arms. Armor and infantry coordination. Artillery suppression. Real-time." He said it without emphasis. "I'm studying it for an era I haven't reached yet."
Marcus glanced at Derek. Derek produced the audience smile. "Studying it. For a game."
"Mm."
Marcus set the pizza back. He studied John with the mild, assessing look he'd maintained unchanged for two years. "You know what I think, Arden? You're actually smart. Garber's averages — you're always up here." He gestured upward. "So you're not dumb. You just —" He looked at the empty table, the laptop, John in general. "Choose this."
"I don't choose anything."
"No?"
"I eat lunch."
Marcus held the look for a moment. Then he stood, and the interaction ended as it always did — not with resolution but with the simple withdrawal of attention, the way you stop looking at something once you've confirmed it won't perform for you. He and Derek and Sully moved back toward their table.
John unminimized Kaiserfront. He noted the exposed right flank in the tactical scenario he'd been running, the position problem that was his next session's work. He closed the tactical layer and returned to the campaign map, and finished ordering the march.
The hallway afterward.
He heard the footsteps behind him — their specific cadence, measured and deliberate, the sound of people who had no reason to hurry — and he didn't turn around because turning around was a variable with a poor outcomes distribution. He had run the expected value on turning around two years ago and found it consistently negative across scenarios: best case, the attention dissipated faster; most likely case, it encouraged escalation; worst case, physical contact became more probable. He kept walking.
The backpack came off his shoulder in one smooth pull — practiced, efficient, no fumbling — and hit the linoleum and the partially open pocket spread its contents across the floor. Chemistry notebook. AP History notebook. The black composition book with its sixty pages of game calculations and city diagrams and military doctrine notes. A mechanical pencil. The national focus tree printout with margin annotations tracking his planned decision sequence through the next thirty turns. The governance policy reference chart he'd printed from the Kaiserfront wiki and annotated with his specific slider history. The order-of-battle sheet with every unit in the Federation's army listed by regiment name, province location, troop count, readiness rating, and current orders.
And — landing face-up near the far wall — a fourth printout he'd made two weeks ago: a Kaiserfront mid-century doctrine reference card. Military symbology. Suppression values and movement rates for armored, motorized, and artillery units at the technology tree's later nodes. Handwritten notes along the bottom margin: suppression window: 40-60 sec at full battery. FO advance: 600m minimum. Anti-tank dead zone: 150m, use infantry.
He stopped. He looked at what was on the floor.
Marcus and Derek and Sully were already around the corner. The hallway moved with practiced non-attention — the particular blindness of people who had decided not to see something, which was different from actually not seeing it.
A girl in a yellow jacket stepped around the doctrine reference card without looking at it. The suppression values and unit icons were face-up, legible, unread.
A pair of freshmen crossed to the far side of the hallway.
Mr. Renner, Social Studies, came from the opposite direction, registered John crouching on the linoleum, and continued without breaking stride.
John picked up everything. He unfolded the order-of-battle sheet — there was a bootprint across the cavalry section, the third time in two months. He picked up the doctrine card, which was undamaged, and looked at it for a moment: the suppression window, the forward observer note, the anti-tank dead zone he'd written in the margin after his fourth failed attempt at the ridge scenario. He folded it along its original lines and put it in the bag.
He stood. He zipped the backpack fully. He walked to Chemistry.
In the margin of his chemistry notes, during the lecture, he drew the ridge engagement from the tactical scenario. The anti-tank gun positions on the reverse slope. The forward observer placement that had cost him two extra tanks by being six hundred meters too far back. The corrected FO line, further forward, inside the fog-of-war zone, exposed but necessary — the price of accurate suppression was putting an observer where the observer could get hurt.
He wrote under it: You can't suppress what you can't see. Getting the picture costs something.
Mrs. Padilla wrote a molecular equation. John looked at what he'd written and thought about it for a moment — not about the scenario specifically, but about the principle, which went somewhere more general and more uncomfortable than the game, somewhere he didn't have the bandwidth to follow right now — and then he drew the Cresward River ford diagram below the tactical sketch, the three crossings, the bluff line, the march route for the infantry regiment.
Two systems. Same page. He was always holding multiple systems simultaneously.
The bus home. Diesel, wet coats, the tired quiet of the after-school hour. Window seat three rows back on the right side. Columbus scrolling past the scratched plexiglass — the gas station at Maplewood, the coin laundry, the check-cashing place with the faded awning that had been faded for as long as he could remember.
He thought about the forward observer problem and the Veth ford problem in parallel, the way they ran in adjacent tracks in his head without interfering with each other, because they were the same problem at different scales and eras: the problem of seeing enough of the battlefield to act on correct information, at the cost of exposing something to gain that visibility.
In the tactical scenario: the recon platoon pushed six hundred meters into the fog-of-war zone, close enough to observe the reverse slope gun positions, vulnerable to being spotted and suppressed in return.
In the Cresward situation: the cavalry patrol he was going to dispatch into the northeastern forest hexes — riders moving through terrain where a hidden Veth flanking force might be, close enough to find it but close enough to be found.
The principle was the same. Reconnaissance was exposure. You accepted the exposure because operating blind was worse.
He wrote in the composition notebook: Cavalry patrol NE forests — 3-day sweep. Accept detection risk. Better to know than not know.
Then: Tactical layer: FO repositioned 600m forward in next phase. Accept suppression risk on FO team. If I lose the FO I lose the suppression window. Blind suppression already cost me two tanks at the ridge.
He thought about Veth. If Aldric III was a rational actor — and the game's AI was modeled to be rational within its information set — then his decision to stage at Varenk rather than commit to either ford immediately was itself a weapon. A commander confident in a quick victory moved. A commander running a deterrence probe held at the decision point as long as possible, keeping the uncertainty alive. The uncertainty was the mechanism: hold at Varenk long enough and the Federation either over-militarized — damaging the industrial transition — or failed to militarize and left the ford undefended. Either outcome served Veth without a battle.
Aldric was playing a suppression game without firing a shot. Keeping John's attention on the military panel, his treasury calculations stressed, the resource diversion question permanently unresolved.
The countermeasure was the same as the tactical countermeasure: push the observation forward. Close the fog of war. Force the enemy to either reveal their actual intention or withdraw the ambiguity.
He wrote: Diplomatic envoy to Veth — trade proposal as cover. Real purpose: observe army supply status and camp configuration. If provisions are heavy, it's real. If light, it's a bluff — they can hold Varenk 6-8 weeks max before upkeep forces a withdrawal decision.
He looked at both pages — the tactical ridge problem and the Veth border problem — and understood that they were both instances of the same underlying architecture. See more than the enemy. Suppress what you can see. Move through the window you've opened. Hold what you've taken. The weapons changed across eras. The logic didn't.
The bus reached his stop. He walked the four blocks to Crestwood Manor and climbed the stairs.
His mother's sneakers were by the door.
He registered them with the specific relief of someone who hadn't been consciously anxious. She was home, which meant she'd had a short shift or a schedule change, and the apartment was occupied in the ordinary way it sometimes was before dinner. He could hear the television through the wall — the channel she watched when she was tired enough to watch anything without choosing: the home improvement network, low-stakes and continuous.
He knocked.
"Hey." Her voice was flat in the way that meant she'd been awake for too long. "How was school?"
"Fine."
"Come in if you want."
He opened the door. His mother was on the bed in her work scrubs, shoes off, the television on, a glass of water on the nightstand. She was thirty-nine and looked it in the way of someone who worked nights and had been working nights for seven years. Dark hair, his jaw, his narrow way of sitting — as if occupying less space was a form of consideration. She had a half-eaten granola bar on the blanket beside her and the faintly abstract look that meant she was present but her nervous system was still processing the shift.
"You eat?" she asked.
"I will."
She nodded. She wasn't going to push it. They had established, over several years of iteration, a domestic equilibrium built on mutual low-demand and mutual competence — she didn't ask him whether he'd done his homework because he always had, and he didn't ask her whether she was okay because the question was unanswerable in any productive way and they both knew it. What they had instead was presence: the awareness of shared space, meals occasionally timed to overlap, sometimes thirty minutes of the same television program before she fell asleep and he went to his room. It wasn't what he imagined other families had, but it was real and it was theirs and he didn't consider it insufficient.
"Anything happen?" she asked.
He thought about the backpack. The bootprint on the order-of-battle sheet. Mr. Renner's stride, unbroken.
"No," he said. "Regular day."
She accepted this. He said goodnight — she had another shift at eleven — and went to the kitchen and heated the tomato bisque, and ate it standing at the counter, and washed up, and went to his room.
His father had texted again. Hey bud. Saw Ohio State is doing something this weekend. You into football at all?
John read it. He put the phone face-down on the desk. He would answer it sometime in the next week, something brief and noncommittal that acknowledged the text without inviting a conversation he didn't have architecture for. David Arden was not a bad person. John had spent some time in his fourteenth year deciding whether his father was a bad person and had concluded that the evidence didn't support it. He was a person who had left, which was different from a bad person. The leaving had consequences — financial ones that John was aware of mostly through the meals that were what was available and the laptop held together with electrical tape — but cataloguing consequences was not the same as assigning blame, and he'd found that trying to assign blame produced a kind of frictionless circular thinking that never resolved and consumed time he could spend on things that resolved.
He put his headphones on and opened Kaiserfront.
The desktop setup: a door laid across two filing cabinets, assembled over eight months, component by component, learned into existence through forum posts and secondhand markets and one afternoon with a YouTube tutorial that had gotten him most of the way toward understanding what a PCIe slot was and why it mattered. The monitor was large enough to show the full city build interface without having to scroll, which was not a luxury but a functional requirement — you could not manage Ironhold's production chains on a laptop screen. He'd saved for four months for the monitor. He'd bought it secondhand from a college student in Clintonville for sixty dollars and felt, when he'd carried it up the stairs, something close to the satisfaction the game produced when a long-planned construction queue completed.
He opened the save.
Year forty-three. The United Ardan Federation, the city of Ironhold, the river valley in the forty-third year of his stewardship. He oriented himself the way he always did at the start of a session: full city view first, then the campaign map, then governance, then the tactical layer. All systems. Current state.
The city view first.
Ironhold at night — the game's day-night cycle ran faster than real time but was visually rendered, and the night lighting was one of the things the game did better than it needed to: the city glowed in the dark, the foundry districts orange-hot along the river, the residential streets lit by the gas lamps he'd installed three years ago through the Public Illumination infrastructure project, the commercial district brighter near the market hall and dimmer toward the residential edges, the new eastern construction zone dark and framework-only, waiting for the water main extension to complete before the first buildings could begin.
He opened the production overlay. The iron smelters: running at 89% efficiency, limited by ore purity — the concentration facility in the northeastern province was two turns from its next upgrade, which would push purity from 94% to 97% and give him the last few points of smelter efficiency he needed to run the foundry chain at nominal. The machine parts workshops: 74%, limited by skilled labor availability — the artisan housing shortage he was correcting. The charcoal kilns: still 71%, the Tier 2 housing 60% complete. He checked the construction timer. Three more turns.
He advanced the city two turns.
The housing completed on turn two. He watched the efficiency readout update in real time: 71% to 76% as the nearest kiln workers relocated, then again to 82% as the commute penalty recalculated, then to 87% as the final worker assignments resolved. The charcoal supply meter climbed from a six-day surplus to an eleven-day surplus. The iron smelters, receiving full charcoal flow, ticked up another two efficiency points.
He exhaled. Not dramatically — just the breath that came when something you'd been managing for a long time finally clicked into place.
He opened the research panel.
Steam Power Fundamentals: 65%, six turns to completion. He checked the Academy's research rate — the Tier 3 upgrade he'd completed in year forty had lifted output from 8 research points per turn to 14, which was what had made the six-turn completion estimate possible. Before the upgrade it would have taken eleven turns, and eleven turns was too slow given the era threshold he was trying to hit.
He queued the next research nodes in order: Steam Engine Production, then Railway Survey, then Industrial Chemistry. The sequence was locked. He would not deviate from it.
He advanced the research two turns and watched the completion bar move.
Then he opened the governance panel.
National Focus: fifty-six days on Civic Consolidation. The parliamentary representation node. He checked the faction satisfaction levels: Liberals 62%, up from 58% — the Tier 2 housing had a small positive effect on their urban development satisfaction metric. Socialists 54%, down from 57% — the Healthcare Access funding he'd deferred two sessions ago was producing a slow satisfaction bleed he needed to reverse before it reached the 50% threshold that triggered parliamentary opposition to his construction budget. Conservatives 71%, stable. Nationalists 48%, which was concerning.
He opened the Nationalist satisfaction breakdown. Their satisfaction was tied to three metrics: Military Spending (currently 120 gold per month, rated Adequate), Territorial Integrity (high, no threats materialized), and National Prestige (declining, because the Intelligence report about Veth's army had leaked into the game's public information layer and was being interpreted domestically as foreign intimidation going unanswered). He needed to either increase visible military action — which he didn't want to do — or produce a diplomatic success he could spin as a Nationalist victory.
He noted: Open formal alliance negotiation with Aldeach — not necessarily complete it, just open it. Alliance talks generate a Prestige notification event that reads as diplomatic assertiveness. Will hold Nationalist satisfaction at 48-52% for eight turns without military expenditure increase.
He queued the diplomatic action. The envoy would be dispatched next turn.
Then he opened the campaign map.
Veth at Varenk, unchanged. Army encamped. Diplomatic envoy in transit. Alliance negotiation with Aldeach dispatched.
He had, in a notebook he kept specifically for diplomatic notes, mapped the regional power structure around the Federation the way he mapped production chains — every nation, their relationship scores, their known military strength, their historical behavior patterns, their likely responses to Federation expansion. Veth was the immediate threat. Skarrath, to Veth's east, was the long-term one — a larger nation, further away, not yet alarmed by the Federation's growth, but positioned to become alarmed once the Federation crossed the industrial threshold and its power projection capabilities increased. He'd been pre-shaping that relationship for five years of game-time, maintaining regular trade contacts and avoiding any actions that would register as hostile to Skarrath's interests.
The military strategy was downstream from the diplomatic strategy, which was downstream from the economic strategy, which was downstream from the city-building that made the economy function. He was managing four levels of abstraction simultaneously, and the levels were not separate — a decision at the city level cascaded upward through economics and diplomacy and reached the military layer twenty turns later in a form that was barely recognizable as related to its origin. You had to hold the whole chain in your head at once, or you'd optimize one layer and break the others.
He advanced the campaign map.
Veth held at Varenk. The cavalry patrol entered the northeastern forest hexes. The infantry regiment marched day one of six toward the southern ford. Steam Power Fundamentals clicked to 70%. The alliance envoy to Aldeach departed the Federation border.
He advanced another turn.
On day three, the diplomatic envoy crossed into Veth territory and passed within observing range of the Varenk encampment. The game's visibility mechanic calculated what the envoy's route could observe: camp configuration, approximate supply wagon count, the depth of provision stockpiles in the eastern logistics area. The intelligence would arrive on the envoy's return.
He opened the tactical layer.
The tactical battle view was a different visual grammar entirely from the campaign atlas — operational and immediate where the atlas was stately, the terrain rendered in functional detail rather than cartographic elegance. A four-kilometer stretch of the Skarrath river valley he'd built as a future war scenario: wheat fields and treeline, two villages anchoring the flanks of a prepared defensive line, unit icons overlaid in precise military symbology with range rings and suppression indicators and the fog-of-war gradient showing each unit's observation limit.
This was the mid-century military layer. Once the Federation crossed Motorization and Armored Doctrine, the campaign's battles would look nothing like the current era. No cavalry, no skirmish lines, no volley fire. Armor. Motorized infantry. Self-propelled artillery. And above all the combined arms logic: each element covering the fatal weakness of another, no element capable of operating alone. Tanks devastatingly effective against infantry in the open, vulnerable to anti-tank guns in prepared positions. Infantry able to hold ground and clear buildings, unable to survive armor without anti-tank support. Artillery suppressing anything it could see, blind without forward observers, and forward observers always exposed.
He had been studying this system for two months by running tactical scenarios before he needed them — the approach recommended in Tessellate's forum post. The players who make the transition smoothly are the ones who've been running tactical scenarios for ten in-game years before they need them. John had taken this as a minimum, not a target.
The current scenario: an armored breakthrough against a prepared defensive line, degraded logistics, the full challenge set. He'd failed four of six attempts. On attempt five he'd cracked the core mechanic: suppression was not destruction. It was temporary incapacitation through incoming fire volume — suppressed units moved slower, aimed worse, made command decisions at a reduced rate. You didn't need to kill an anti-tank gun; you needed to pin its crew for the forty to sixty seconds it took your armor to cross the exposed ground. Open the window with suppression, move the armor through it, close the window before the enemy unsuppressed.
He'd failed four times to understand that. Understanding it had cost him, across those failures, the equivalent of several hours of real time and what he estimated was six hundred pages of forum reading about combined arms doctrine and suppression mechanics and the coordination timing that separated a successful attack from a catastrophic one. He didn't consider that time wasted. He considered it training.
He advanced the current tactical phase.
Artillery suppression on the ridge treeline — the battery registered on the anti-tank gun positions he'd identified in the previous phase through careful forward observer work. Smoke rounds on the flanking positions, forcing the gun crews into chemical protective posture and reducing their traverse speed by 40% for three minutes. During that window: armor around the northern edge where the treeline gave partial cover, infantry pushing centrally through the smoke. The infantry had man-portable anti-tank teams distributed one per platoon.
The phase resolved. He lost two tanks to an unlocated anti-tank gun whose position had been in his fog-of-war zone — a known cost he'd accepted, the price of incomplete intelligence. The suppression had held on the main axis. Infantry were consolidating on the forward slope.
He noted in the composition book: FO placement — 600 meters further forward before H-hour. Blind suppression costs armor. The observation radius is the constraint on everything downstream.
He looked at the notebook page.
The forward observer problem and the Veth ford problem and the factory supply chain problem and the political faction management problem were all, at some level he couldn't quite articulate precisely, versions of each other. They were all problems about information: seeing enough of a system to act correctly on what you saw, at a cost that was paid upfront in exposure or resources or political capital, in exchange for future accuracy. The cost of seeing clearly was always something. The cost of not seeing clearly was always larger.
He minimized the tactical layer. He returned to Ironhold.
He ran all three systems through the night.
The city building, the campaign, the tactical layer — the way he always ran them, all three panels available, his attention moving between them as each demanded it. The city's construction timers ticked forward. The eastern industrial district took shape: the Machine Parts Workshop laying its foundations, the Steam Turbine Hall's first floor framing going in, the water mains completing and the pressure reading climbing to operational threshold. In the residential zones, the Tier 2 housing filling with workers, the commute penalties dropping, the efficiency ratings climbing toward the numbers he'd planned for.
He advanced the technology tree. Steam Power Fundamentals hit 80%, then 90%, then completed on the eleventh turn with a notification that unlocked Steam Engine Production and Railway Survey simultaneously. He queued Steam Engine Production immediately — it had a seven-turn completion time, and the Steam Engine Assembly Hall's foundations were already poured, waiting for the technology unlock before construction could begin.
He managed the politics. The Civic Consolidation national focus completed, converting the Federation's government to a constitutional model and giving all four factions parliamentary representation. The Liberal and Socialist satisfaction numbers climbed eight points each. The Nationalist satisfaction held at 50% on the strength of the Aldeach alliance announcement, which the game's public information system had rendered as a diplomatic success story visible across the Federation's population. He used the parliament's first session to pass the Education Expansion bill — a moderate increase in Education Investment from 74% to 79%, funded by a two-point Tax Burden increase that the Conservatives accepted because the constitutional framework gave them institutional protections they'd been demanding for twelve years of game-time.
He managed the campaign. The cavalry patrol returned from the northeastern forests: clear. No hidden formations. One army, one crossing decision, Varenk. The diplomatic envoy returned from Veth with intelligence: moderate provision stockpiles, not provisioned for a winter campaign. A bluff, or a probe that expected a response. Either way, not an imminent assault. He used the alliance completion with Aldeach — finalized two turns after the opening — as a public diplomatic event, which triggered the Prestige notification he'd planned for and pushed Nationalist satisfaction to 56%, safely clear of the opposition threshold.
The infantry regiment arrived at the southern ford. Readiness 91%. He deployed them on the western bluff — 600 men in a two-rank line along the crest, garrison company in reserve behind the crestline. He looked at the deployment on the battle map and thought about what it would look like in forty years of game-time: tanks hull-down on the reverse slope, anti-tank missiles on the flanking high ground, motorized infantry in prepared positions in the tree line, an artillery battery registered on the ford crossing points with a forward observer team at the bluff's forward edge — exposed, exposed in the way that forward observers were always exposed, but seeing clearly.
The logic was the same. The weapons were different.
He made a note: Forward position doctrine applies across eras. Push observation forward, accept exposure cost, suppress before movement, move through the window. Era changes the tools. Principle holds.
At some point past midnight the tactical phase resolved. The FO team was established in the forward treeline, observing, transmitting. The artillery battery had accurate coordinates for the ridge positions for the first time in the scenario. He didn't fire yet. He just established the picture. Let the observation settle. Made sure it was correct before he acted on it.
He minimized the tactical layer. He advanced the Kaiserfront turn.
And then he ran them all forward together — the city building and the army holding and the research climbing and the doctrine waiting in the background, patient, accumulating — into the night and past it, the screen's glow the only light in the room, John Arden in the dark apartment on Crestwood Avenue while his mother slept down the hall and his father's text sat unanswered on the nightstand and the city of Ironhold thrived on the screen under his careful hand.
He was sixteen. In the game he was a statesman, an engineer, a general, a city planner, a diplomat, a logistician. He managed the food supply of 84,000 people and the political expectations of four factions with incompatible desires and the military security of a federation of three provinces against a kingdom that had just decided, for reasons he was still working to understand, that the moment of his greatest industrial ambition was the moment to test whether he would flinch.
He would not flinch. He had data, plans, reserves, prepared positions, and forty game-years of decisions that had been building toward exactly this crisis. The Federation would cross the industrial threshold. Ironhold would be the city he'd been designing it to be. The military would hold the ford and the diplomacy would hold Veth and the technology tree would keep climbing toward eras he was already studying.
He had a lot of work to do.
In his composition notebook, at the bottom of the last full page — below the city diagrams and the ford calculations and the governance slider notes and the forward observer sketch — he'd written, at some point during the session, in small careful letters:
Hold the line. Keep building. Study what you'll need before you need it.
Don't let them make you fight on their terms.
And get the picture. Whatever it costs, get the picture.
He looked at it for a moment.
Then he turned to a fresh page and kept working.