Eras¶

Stochastic Warfare supports 5 historical eras, each with different available technologies, unit types, and combat mechanics. The era framework gates module availability and loads era-specific data.

Era Framework¶

Era Enum¶

Value	Period	Data Directory
`MODERN`	Cold War -- present	`data/` (base)
`WW2`	1939 -- 1945	`data/eras/ww2/`
`WW1`	1914 -- 1918	`data/eras/ww1/`
`NAPOLEONIC`	1792 -- 1815	`data/eras/napoleonic/`
`ANCIENT_MEDIEVAL`	3000 BC -- 1500 AD	`data/eras/ancient_medieval/`

EraConfig¶

Each era defines an EraConfig that specifies:

Enabled modules -- which subsystems are available (e.g., no radar before WW2)
Available sensor types -- what sensor modalities exist (e.g., visual only in Ancient)
C2 delay multiplier -- how slow command propagation is (courier >> radio)
Era-specific engine extensions -- custom combat models for the period

Setting era in a scenario YAML automatically applies these constraints.

Modern Era (Cold War -- Present)¶

The default era with full access to all subsystems.

Enabled Modules¶

All modules enabled: EW, Space, CBRN, directed energy, full sensor suite, digital C2.

Available Sensor Types¶

Visual, thermal/IR, radar, acoustic, sonar, UV MAWS, laser warning, SIGINT.

C2 Delay¶

Baseline (1.0x multiplier). Digital communications, Link 16, SATCOM.

Key Unit Categories¶

Domain	Examples	Count
Armor	M1A2 Abrams, T-72B3, T-90A, Leopard 2A6, Challenger 2	5+
Infantry	Rifle squad, mechanized, Javelin team, Kornet team	4+
Artillery	M109 Paladin, MLRS/HIMARS, mortar team	3+
Fixed-Wing Air	F-16C, A-10C, Su-27S, MiG-29A, J-10A, B-52H, EA-18G	7+
Rotary-Wing	AH-64D Apache, UH-60 Black Hawk, Mi-24V	3+
Air Defense	Patriot, S-300PMU, SA-11 Buk, MANPADS	4+
Naval Surface	Arleigh Burke DDG, Sovremenny DDG, Ticonderoga CG	3+
Naval Subsurface	Los Angeles SSN, Kilo-636 SSK	2+
Support	Engineers, logistics, HQ, medical, C-17	5+

Available Doctrines¶

US FM 3-0, Russian Deep Operations, NATO Procedures, PLA Active Defense, IDF Preemptive Strike, Airborne, Amphibious, Naval Sea Control.

Available Scenarios¶

30 modern scenarios including 73 Easting, Falklands Naval, Golan Heights, Taiwan Strait, Korean Peninsula, Suwalki Gap, and 3 calibration exercise scenarios (arctic, urban CBRN, air-ground). All scenarios validated against historical outcomes in Phase 47 (seed=42 deterministic + Monte Carlo).

WW2 Era (1939--1945)¶

Enabled Modules¶

Most modules enabled. No EW (electronic warfare), no space, no directed energy. CBRN limited (no nuclear in most scenarios). Radar available but primitive.

Available Sensor Types¶

Visual, radar, passive sonar, active sonar (no thermal/IR targeting, no GPS).

C2 Delay¶

Baseline (1.0x multiplier). Radio communications available but less reliable. No digital data links.

Era-Specific Mechanics¶

Naval Gunnery Bracket Firing

WW2 naval guns use bracket firing -- observing fall of shot and adjusting. The engine models:

Initial salvo spread based on fire control quality
Bracket adjustment (long/short alternation to find range)
Convergence to target over successive salvos
Fire control radar bonus for equipped ships

Convoy / Wolf Pack

Submarine warfare modeled as:

Convoy formations with escort positions
Wolf pack tactics (multiple submarines coordinating attacks)
Escort effectiveness based on numbers and sonar quality
Night surface attacks vs submerged daylight attacks

Strategic Bombing CEP

High-altitude bombing with circular error probable (CEP):

Unguided bombs with large CEP (hundreds of meters)
Target area coverage using Gaussian scatter
Fighter escort modifying bomber attrition
Target regeneration over time (industrial recovery)

Key Units¶

Domain	Examples
Armor	Tiger I, Panther, Sherman M4A3, Panzer IV, T-34/85
Infantry	US/Soviet/Wehrmacht rifle squads
Air	P-51D Mustang, Bf 109G, B-17G, A6M Zero, Spitfire
Naval	Essex CV, Bismarck BB, Fletcher DD, Type VII U-boat, Shokaku CV
Artillery	105mm howitzer, Katyusha, Nebelwerfer

Available Scenarios¶

Kursk 1943 (largest tank battle), Normandy Bocage 1944 (hedgerow fighting), Stalingrad 1942 (urban combat), Midway 1942 (carrier battle). All validated against historical outcomes in Phase 47.

WW1 Era (1914--1918)¶

Enabled Modules¶

Limited modules. No EW, no space, no CBRN (except gas warfare via adapter), no directed energy. Basic visual detection only. Very slow C2.

Available Sensor Types¶

Visual only. No radar, no thermal, no acoustic sensors.

C2 Delay¶

5.0x multiplier. Telephone/telegraph to HQ, runners and couriers forward. Significant communication friction.

Era-Specific Mechanics¶

Trench System Overlay

Trenches modeled as spatial overlays using Shapely LineString geometries:

STRtree for fast spatial queries
Trench segments provide cover and concealment bonuses
Wire obstacles slow movement
Communication trenches enable covered movement between positions
No-man's-land exposure zones

Creeping Barrage

Aggregate fire model for massed artillery:

Fire density measured in rounds per hectare
Barrage line advances at configurable rate
Random walk drift with observer correction
Casualties proportional to troop density in beaten zone

Volley Fire & Melee

WW1 bolt-action rifles route to the volley fire aggregate model (coordinated fire), while close-quarters combat uses the melee engine for bayonet charges and trench raids. Machine guns use the standard direct-fire model for sustained suppressive fire.

Gas Warfare

Chemical weapons via CBRN adapter:

Wind direction/speed gating (gas blows back in wrong wind)
Pasquill-Gifford dispersal model (shared with modern CBRN)
Gas mask don time enforcement (delay before protection)
Persistent vs non-persistent agents

Key Units¶

Domain	Examples
Infantry	British/German/French rifle squads
Artillery	18-pounder, 15cm howitzer
Air	SPAD XIII, Fokker Dr.I, SE.5a
Naval	HMS Dreadnought, SMS Konig, HMS Iron Duke

Available Scenarios¶

Somme July 1 1916 (first day of the Somme), Cambrai 1917 (first mass tank assault), Jutland 1916 (dreadnought fleet action). All validated against historical outcomes in Phase 47.

Napoleonic Era (1792--1815)¶

Enabled Modules¶

Minimal modules. No electronic anything. Visual detection only. Courier-based C2 with extreme delays. Foraging-based logistics.

Available Sensor Types¶

Visual only. Detection range limited by terrain and weather.

C2 Delay¶

8.0x multiplier. Courier on horseback. Orders take hours to propagate. Fog of war is extreme.

Era-Specific Mechanics¶

Volley Fire

Aggregate model using binomial trials:

Musket accuracy varies by range (effective only within ~100m)
Formation affects volley effectiveness (line > column for firepower)
Rate of fire: ~3 rounds per minute for trained troops
Smoke accumulation degrades accuracy over sustained fire

Melee Combat

Close-quarters combat model:

Bayonet charges, saber engagements
Morale is the primary determinant (most melee resolved by one side breaking)
Formation effects (square is devastating to cavalry)
Reach advantage for longer weapons

Cavalry Charge State Machine

Multi-phase cavalry engagement:

Approach -- accelerating toward enemy, taking fire
Contact -- melee if defenders hold, rout if they break
Pursuit -- chasing broken enemy (most casualties here)
Rally -- reforming after charge (vulnerable period)

Pre-contact morale is the key mechanic -- most charges decided before physical contact.

Napoleonic Formations

Rock-paper-scissors formation system:

Formation	Strong Against	Weak Against	Best For
LINE	Column (firepower)	Cavalry (flanks exposed)	Defensive fire
COLUMN	Rapid movement	Line (narrow frontage)	Assault
SQUARE	Cavalry (all-round defense)	Artillery (dense target)	Anti-cavalry
SKIRMISH	All (hard to hit)	Cavalry (no mass)	Screening

Transitioning between formations takes time and creates vulnerability.

Courier C2

Orders delivered by mounted courier:

Travel time proportional to distance
Risk of interception or courier loss
Message delay = distance / courier_speed + lognormal noise
Initiative doctrine: commanders act on last received orders

Foraging Logistics

Supply through local foraging:

Daily foraging radius around unit position
Region exhaustion over time (devastated areas produce nothing)
Ambush risk during foraging operations
Strategic implications of supply lines vs living off the land

Key Units¶

Domain	Examples
Infantry	French ligne, British line, Austrian grenadier, Prussian musketeer
Cavalry	French cuirassier, British light dragoon, Cossack, lancer
Artillery	6-pounder, 12-pounder, howitzer
Naval	Ship of the line (1st/3rd rate), frigate

Available Scenarios¶

Austerlitz 1805 (Napoleon's masterpiece), Waterloo 1815 (Coalition defeat of Napoleon), Trafalgar 1805 (Nelson vs Franco-Spanish fleet). All validated against historical outcomes in Phase 47.

Ancient & Medieval Era (3000 BC -- 1500 AD)¶

Enabled Modules¶

Minimal modules. Visual detection only. Visual signals C2 (flags, trumpets). No logistics automation.

Available Sensor Types¶

Visual only. Very short detection ranges.

C2 Delay¶

12.0x multiplier. Visual signals (flags, horns, drums) for nearby units. Messengers on foot for distant commands. Extremely limited command span.

Era-Specific Mechanics¶

Massed Archery

Aggregate volley model for massed bowmen:

Arrows per volley tracked per archer (typically 24 arrows total)
Effective range varies by bow type (longbow > short bow)
Formation density affects casualties
Shield wall and armor reduce effectiveness
Arrow supply exhaustion is a key constraint

Ancient Formations

7 formation types with distinct mechanics:

Formation	Type	Effect
PHALANX	Dense spear wall	Devastating frontal defense, vulnerable to flanks
SHIELD_WALL	Overlapping shields	Strong defense, slow movement
WEDGE	Triangular assault	Penetration bonus on charge
TESTUDO	Roman turtle	Near-immune to missiles, very slow
SKIRMISH_LINE	Dispersed	Hard to hit, weak in melee
SQUARE	All-round defense	Anti-cavalry, slow
OPEN_ORDER	Flexible spacing	Balanced, standard

Siege State Machine

Campaign-scale daily resolution for sieges lasting weeks to months:

Approach -- moving siege equipment into position
Investment -- surrounding the fortification
Bombardment -- siege engines degrading walls
Assault -- storming breaches (high casualties)
Resolution -- surrender, relief, or starvation

Melee Extensions¶

Reach advantage -- longer weapons (pikes, spears) get first strike
Flanking bonus -- attacks from side or rear multiply damage
Charge momentum -- cavalry and wedge formations gain impact bonus

Oar-Powered Naval

Ancient/medieval naval combat:

Ram attacks (primary weapon for triremes)
Boarding actions (melee combat on deck)
Oar speed vs sail speed tradeoffs
Ram damage proportional to relative velocity

Visual Signals C2

Command and control via visual/audible signals:

Flags, standards, horns, drums
Signal range limited by terrain and weather
Misinterpretation probability increases with distance
Commander must be visible to troops for morale effects

Key Units¶

Domain	Examples
Infantry	Hoplite, Roman legionary, Viking huscarl, English longbowman
Cavalry	Companion cavalry, cataphract, knight, horse archer
Siege	Trebuchet, battering ram, siege tower
Naval	Trireme, quinquereme, longship, cog

Available Scenarios¶

Cannae 216 BC (Hannibal's double envelopment), Salamis 480 BC (Greek trireme victory), Hastings 1066 (Norman conquest), Agincourt 1415 (English longbow vs French knights). All validated against historical outcomes in Phase 47.

Creating Era-Specific Scenarios¶

To create a scenario for a specific era, set the era field in the scenario YAML:

name: "Battle of Austerlitz"
era: napoleonic
duration_s: 28800  # 8 hours
terrain:
  width: 8000
  height: 6000
  cell_size: 100
  terrain_type: rolling_hills
sides:
  - name: french
    units:
      - unit_type: french_ligne
        count: 12
        # ...

The ScenarioLoader will:

Apply the era's EraConfig (disable unavailable modules, set C2 delays)
Load unit definitions from data/eras/napoleonic/units/ instead of data/units/
Wire era-specific engines (volley fire, melee, cavalry, formations, courier, foraging)