Eras and technology

Eras give the game a time dimension, so the economy changes as the century does. That is more than a calendar that shows a year in the corner: the rules shift for a reason a player can understand, because in 1905 you cut with a crosscut saw and skid with oxen, and in 1955 you do not. This page explains why that is worth building, how the era table turns it into numbers, and where it could go wrong. The numbers themselves, once settled, belong on the reference page, not here.

Every historical statement on this page that comes from memory is marked (unverified). The game copy should not repeat any of them until a local source has been checked, and the town’s historical society is a better source than anything written here.

What static rules cannot do

A game with one fixed set of rules has a ceiling, and the simulator has been showing us where it is. In the latest five-day runs, a forester who keeps replanting ends at 128,118 and a forester who never replants ends at 28,464, level with a casual player at 27,182. Replanting accounts for about 100,000 of that lead. That is a solved economy: there is one behaviour that beats everything else by a wide margin, and once a player discovers it there is nothing left to discover. The farmer with twelve fields, at 37,544, is in the same position from the other side. Farming earns roughly 280 to 400 board feet per crew-day against about 1,000 for a mature timber parcel, so it is a way to earn while the forest grows and never a competing plan. Balance findings has the full table and the caveats.

Static rules also give the town nothing to be. The shared town fills, its markets settle into Balanced, and a visitor on day 60 sees the same board as a visitor on day 6. The pillar that the first player to fill a gap does best depends on gaps opening, and a fixed world opens each gap once. Finally, a static economy gives no reason to come back that is not a timer. Pick a lane already worries about this at the level of milestones, and the open question there (a clock or player-caused milestones for opening markets) is one this design answers in part.

A time dimension attacks all three. If the rules change on a published schedule, then a lane that leads today may not lead in the next era, a town has a history it moves through, and “what is different now” is a reason to open the game that has nothing to do with a countdown.

What technology does to an economy

It would be easy to model an era as a price table: lumber is worth less, wheat is worth more. That is not what technology does, and the design avoids it on purpose. Technology changes costs, and prices follow, unevenly and late.

A steam donkey engine (unverified as to dates and details) did not make logs sell for more. It made it cheaper to get logs from the steep ground to a landing, which changed which parcels were worth cutting at all. A railroad or a truck road did not raise the price of lumber at the mill; it cut the share of that price that the hauling ate. A chain saw lowered the labour and the wear of every cut. In each case the change is on the cost side of a transaction that already exists, and the effect on the player is that some activity that used to be marginal becomes profitable and some that used to be the backbone becomes ordinary.

Technology also deletes and creates trades. The teamster who hauled by wagon is a lane that shrinks when trucks arrive, and a garage and trucking lane appears. The saw filer’s trade shrinks when saws hold their edge longer. Because the game already has a wanted board that shows underserved and crowded markets, this is easy to express: an era moves the demand and the cost of lanes, the board reflects it, and the first player to notice a new gap is rewarded exactly as the pillar says.

Third, technology shifts demand between species. Mills do not want the same wood in every decade. Building styles change, a disease can take a tree out of the market, and a new use can raise another. This is the one place in the table where the game bends a number up or down for a reason that is not technology, and it is where the design gets its sharpest tool, because the forester’s whole lead is built on replanting one species.

The levers, and what each one stands for

Every lever in the era table is a multiplier on a constant that already exists in the game, with 1.0 meaning the 1900s hand-tool baseline. That choice is the main thing keeping the system testable: an era is a short list of multipliers and nothing else, so the rules engine does not grow a new mechanism per era. The table in the next section gives the values at a glance. The mapping is this.

crewRest multiplies the crew rest. Lower is faster work, and it stands for power saws and more hands. skidLoss multiplies the fraction lost between the stump and the landing, which animals, then donkey engines, then crawler tractors reduced (unverified order). haulShare and tripTime multiply each mill’s haul share and wagon trip time, and stand for steamboats, then railroads, then trucks and paved roads. sawWear multiplies the sharpness lost per cut, so it eats into the filing trade and the filer’s fee. ownMillRate multiplies the logs per hour of the player’s own sawmill, for larger boilers, then electric drive. cropYield and cropPrice move together in opposite directions after the 1920s: draft horses and tractors raise yield, and mechanised farms in general depress prices (unverified as a claim about this valley). speciesDemand multiplies what mills bid for a species, on top of each mill’s own specialty multiplier, and is where blister rust and shifting markets live.

The specific species story is worth spelling out, because it is the part most likely to be wrong. White pine (wp) falls from 1.05 in the second era to 0.7 and then 0.5. The historical reading is that white pine blister rust, a fungus introduced to North America in the early twentieth century, spread through Idaho’s white pine forests and ended white pine as the dominant commercial species there (unverified, including the dates), while Douglas-fir (df) and others took up the demand as the century went on. The game does not need the history to be exactly right, since it uses it only as a reason that a species a forester loves stops paying. It does need to be true enough that a local reader does not wince, and that is a verification task, not a design one.

The era table

Era	Years	Technology	Gets cheaper	Gets less valuable
Homestead	1900 to 1909	Crosscut saws and axes, ox and horse skidding, wagon haul, small steam sawmills	The baseline	Nothing yet
Steam and rails	1910 to 1919	Steam donkey engines, logging railroads, lake steamboats, bigger steam mills	Skidding, hauling, trips, the player’s mill	The teamster stable (fades to 80 percent of its benefit)
Gas engines	1920 to 1939	Logging trucks, crawler tractors, electric mills, graded roads	Hauling and trips again, saw wear, crew time, farm yield	White pine bids, teamster stable (55 percent), filing shop (85 percent)
Power saw and highway	1940 to 1969	Chain saws, big trucks, paved highways, consolidated mills	Crew time and saw wear most of all, hauling, farm yield	White pine, lodgepole and larch bids, crop prices, teamster stable (40 percent), filing shop (60 percent)

All four rows are a draft and all four are first guesses. The dates and the technology column are unverified. A fifth era, for recreation and lake homes, is a placeholder and is not modelled, because it is a different game (waterfront and view value rather than timber). The Silver Valley expansion does not fit neatly into any of these rows.

How the clock works, and why two triggers

The design decision is that the era clock is town milestones plus a calendar floor. The shared town advances to the next era when it meets that era’s needs (a settler count, a number of roads, or a total of lumber produced, as each row of the table says), and no later than a stated number of days after the previous era began. Neither trigger alone is good enough, and it is worth saying why, because a later designer will be tempted to drop one.

A pure milestone clock lets the town stall. A valley with six keen players and forty empty slots will not reach thirty settlers, and an era that never arrives means the new trades never open and the old lead never resets. That is the static economy again, with a better excuse. A pure calendar clock fails the other way. It moves the town whether or not the people in it are ready, it makes the era change something that happens to the players rather than something they did, and it gives impatient players a reason to feel the game is being run on a timer that has nothing to do with them. The pillar that the studio does not steer outcomes leans toward the milestone.

Combined, the milestone is the story and the floor is the guarantee. A healthy town reaches the era through its own growth, and the page that tells players so can say “your town earned this.” A stalled town reaches it anyway, on the floor, and the announcement says that too, because the thin-town case is the one where hiding the difference would be dishonest. Three more reasons favour the floor. An era change is a marketing moment, a dated event (“the first steamboat arrives in Harrison”), and marketing wants a date it can point at. It is a fairness matter between valleys, since two towns that start the same week should not differ by years of technology because one of them found more friends. And it bounds the testing problem: with floors, the longest the game can stay in one era is known, and the simulator can cover every transition in a fixed span.

One trap in the milestone side is already visible in the shared world design. Settler counts can be inflated by townsfolk, and the shared world caps townsfolk households at half the total. The same cap should apply when settlers count toward an era’s needs, or a thin town could be carried by labelled stand-ins into an era its real players have not earned.

How fading works, and why nothing vanishes

The second decision is that old trades fade but never vanish. A perk or a lane declines toward 30 percent of its benefit (fadeFloor) as eras pass, and stays playable. In the table above, the teamster stable’s halving of hauling shares goes from full benefit to 80, 55 and then 40 percent across the eras, so by the last row it is a modest discount and still a discount.

The reasons are mostly about respect for what a player did. Someone who built a barn and then a stable in the first era paid for it in lumber and in days. If the third era turned it into a decoration, that is the studio taking something from a person who followed the rules, and it is the kind of thing that makes players distrust the next announcement. A fading perk says: you were right to build this, and the world has moved on.

There are two further benefits. Old trades keep a floor under variety: a player who prefers the quiet pace of hand work can stay on it and still be in the game, which is closer to the pillar of picking a lane than a hard cutoff would be. And an old trade becomes a premium niche in the way real ones do. Hand-split cedar shakes, a wagon service for the farms the highway skipped, a filing shop that still does the old saws (all of these are illustrations, not decisions) are smaller markets in a later era, and a small market with few sellers is exactly what the wanted board marks Underserved. Nostalgia crafts are not only flavour; they are lanes the board can show.

The cost of fading is that it adds one more number per perk per era to tune, and that the 30 percent floor is a first guess. A floor too high means nothing ever really goes out of date and the era feels like a coat of paint. A floor too low is vanishing under another name.

What eras do to the balance problems

The simulator’s findings give an unusually concrete test for this feature, so it is worth being plain about the hope and the risk.

The hope is that an era change is a recurring rebalancing event. Because the levers act on the constants the forester’s advantage rests on (white pine demand, skidding loss, hauling share), an era change resets which lane leads. The forester’s compounding lead comes from keeping stands in timber. When white pine bids fall to 0.7 and then 0.5 of baseline while Douglas-fir rises, the lead does not go away, but it now depends on the player choosing a different species, and a forester who kept replanting white pine out of habit gets less for it. That is the kind of “one behaviour stays on top” problem that cannot be fixed with a static rule without making the rule feel arbitrary, and a historical reason makes it feel earned.

The risk is the opposite. Some levers help the player who is already ahead more than the one who is behind. The clearest case is crewRest. The casual bot is capped by how much work it can bank when it returns, not by how fast the crew refills, so a lower crew rest speeds up the continuous player and does almost nothing for the player who checks in every eight hours. If an era shortens crew rest from 1.0 to 0.5 and nothing else compensates, the 4.7 times gap between the leader and the casual bot could grow, and the finding that the gap comes from check-in frequency (a casual player who replants but checks in every eight hours also ends near 27,400) says it would grow for a structural reason. The same worry applies to ownMillRate, which rewards the player who already built a mill. These are the levers to watch first, and a catch-up design for the casual player, still an open decision, may turn out to be a precondition for the power-saw era rather than a separate feature.

There is a second way eras can make things worse. Several levers multiply together for a player who exploits all of them, and a net worth that jumps at the boundary is a symptom. The target set for the simulator is that the net worth of a standard bot immediately after an era change is no more than 1.5 times what it was just before, given the same play.

What could go wrong

Era whiplash. If every row changes many levers by a lot, a player’s plan stops working on the day of the announcement and the game feels like it punishes success. The fading rule helps, but so does keeping each change small and telling players what is coming. The era change should be announced a few days ahead, with the new numbers visible, so that the event is a thing to prepare for. That is also the best argument against surprise species shifts: the blister rust reading is dramatic and the drama should be foreshadowed.

Pay-to-skip pressure. A world that moves on a schedule invites the question of why a player cannot buy the next era, or buy a perk that survives it. The design position is no. An era belongs to the shared town, so skipping it by purchase would be one player spending the town’s time.

New players joining a late era. A newcomer who arrives in the power-saw era has never owned a donkey engine, never built a stable that faded, and has an economy where white pine is worth half. Is that a disadvantage? It may be the opposite, since they start with the best tools and a cleaner choice of lane. It is also possible that they start with nothing against a town of mature forests and settled markets. The own-town sandbox can teach any era, but whether late joiners get catch-up tools is open.

Content cost. Each era needs art, copy, an announcement, a perk set, a wanted-board entry for the new trades and a historical check. Four eras is a manageable number; ten would not be. The design stops at four timber eras for that reason.

Tuning burden. Nine levers across four eras is thirty-six numbers, plus the fade values, and every one of them interacts with the others and with the constants underneath. The simulator is how that is kept sane, and the tuning targets below exist so that a change is judged by a rule written down in advance.

Test criteria

The tuning targets give three measurable checks, each to be run on the era-fast simulator variants. First, the leader should end no more than 3.0 times the casual bot (leaderToCasualMax), which is a real tightening on the current 4.7. Second, every lane should reach at least half of the leader’s net worth (anyLaneAtLeastShareOfLeader is 0.5), which the current results fail badly, since the farmer is at under a third of the forester. Third, an era change should not multiply a standard bot’s net worth by more than 1.5 (eraChangeNetWorthJumpMax). The third is read as a comparison of a bot’s net worth just before and just after the change with the same play.

These targets are demanding. The first two are not met today even without eras, so an honest first result may be that eras alone do not meet them and something else (the catch-up question, the farming yield) has to move too. Seeing that early is worth more than seeing it after launch.

Where this sits

Eras are one more case of the project’s habit of saying why a rule exists. The shape (multipliers on existing constants, published dates, no hidden changes) is chosen so that the rules stay public and a ruleset change stays a dated, testable event, which is what the shared game’s rules engine is built to carry. Nothing on this page describes land or timber as rising in value; an era changes what work costs and what mills bid, and a fading trade is not a loss of capital but a change in how useful an old tool has become.