Rainbow Farming: Oasis Raids and Staggered Waves

Rainbow farming is a method of farming unoccupied oases with very small armies made from multiple troop types.

The basic idea is simple: instead of sending several copies of one troop, split the attack across different troop types so Travian rounds casualties separately for each type.

A rainbow does not have to be 1 + 1 + 1. It can be 1 + 1, 1 + 1 + 1, 2 + 2, 2 + 2 + 2, 3 + 3 + 3, or an uneven combination such as 6 + 6 + 3.

The important part is not the shape of the rainbow. It is the combination of attack strength, separate casualty rounding, infantry/cavalry defense, Smithy upgrades and troop speed.

At its most advanced, rainbow farming can also use several different-speed armies released at the same time so that they arrive at the oasis in separate waves.

Why Rainbow Farming Works

Travian calculates the attacker's total offensive strength and compares it against the defender's effective defensive strength.

For a raid, first calculate:

x = (D / A)^1.5

where A is attack strength and D is effective defense.

The winning side's casualty fraction is then:

L = x / (1 + x)

Casualties are then calculated using that percentage. The critical mechanic is that casualties are rounded separately for every troop type. This is the foundation of rainbow farming.

The 50% Breakpoint

Suppose an army contains one troop of each of three different types: 1A + 1B + 1C, and the calculated casualty rate is 40%.

1 × 0.40 = 0.40 → 0 casualties

The same calculation happens independently to B and C.

Mathematically, the army has 1.2 expected casualties, but no individual troop type reaches the rounding threshold. All three survive.

Why One Troop of Each Type Is So Powerful

1 × L < 0.5 while L < 50%

A one-unit stack can survive a winning raid while its calculated casualties remain below the rounding boundary. Adding another different troop creates another independent rounding bucket. This is why a tiny mixed army can sometimes return without losses from a fight in which a larger single-type army would lose troops.

Why 2 + 2 Is Not Simply Better Than 1 + 1

Imagine a battle produces a 30% casualty rate.

1 + 1

1 × 0.30 = 0.30 → 0

Each troop type survives. Total casualties: 0.

2 + 2

2 × 0.30 = 0.60 → 1

Each troop type can lose one unit. Total casualties: 2.

The larger army contains twice as many troops but crosses a completely different rounding threshold. Adding troops also increases attack strength and therefore reduces the casualty percentage, so the real comparison is a competition between additional attack and a harsher rounding threshold.

This is why 1 + 1 can survive a specific fight where 2 + 2 loses troops. After adding still more attack, a larger composition may eventually become safe again.

Rounding Thresholds

Small one-unit rainbows maximize rounding protection. Larger rainbows trade some of that protection for greater raw attack.

Real Example: Hun 1 + 1 + 1

A clean Hun rainbow is 1 Steppe Rider + 1 Marksman + 1 Marauder with no Smithy upgrades. This combination reaches roughly 410 nature-defense strength in the standard worked example.

Two Bears have 200 cavalry defense each:

2 × 200 = 400

With 10 additional oasis defense:

400 + 10 = 410

The three-unit Hun rainbow can cross that defensive strength while keeping every attacking troop in its own casualty bucket.

1 Steppe + 1 Marksman + 1 Marauder → 2 Bears → 0 losses

Two-Color Rainbow Farming

Two-color farming can be extremely efficient because only two troops are committed to each raid. It can also reduce production pressure compared with a three-color rainbow.

Gaul Example: TT + Haeduan

A particularly useful Gaul combination is 1 Theutates Thunder + 1 Haeduan.

Smithy upgrades become part of the strategy. One practical configuration is TT Smithy level 7 and Haeduan Smithy level 6.

The important single-animal cavalry-defense breakpoints include Crocodile at 240 and Tiger at 250, before the complete oasis-defense calculation is applied.

Instead of automatically adding a third troop, strengthen the existing two colors until the pair crosses the required defensive breakpoint. Both troops then retain the powerful one-unit rounding behavior.

Pending worked example: TT Smithy level 7 + Haeduan Smithy level 6 against crop-oasis animals. The source document does not include verified results for this example.

Animal Defense Matters

Animals cannot be ranked using one generic defense value. Infantry and cavalry can face very different breakpoints.

Tiger vs Crocodile

Against infantry, Crocodile defense is 380 while Tiger defense is 170:

380 - 170 = 210

Against cavalry, Crocodile defense is 240 while Tiger defense is 250:

250 - 240 = 10

Therefore Tiger is slightly tougher for a pure cavalry rainbow, while Crocodile is dramatically tougher for infantry. The attacking troop composition determines which animal is actually more dangerous.

Different Oases Require Different Rainbows

For pure cavalry, the highest listed single-animal cavalry defenses are approximately 50 for Iron, 40 for Clay, 200 for Wood, and 250 for Crop, before the oasis' own defense and the complete combat calculation. A rainbow built for Iron or Clay does not need the same strength as one intended for Crop oases.

Clearing and Maintenance Are Different Jobs

Clear: Remove the existing animals with the hero or an appropriately sized army.

Maintain: Once the oasis is empty, use small rainbow groups against new nature spawns.

Recover: If farming stops and animals accumulate again, clear the oasis before restarting the small-wave system.

This lets the maintenance army be optimized for small encounters instead of the worst force that could accumulate after hours of inactivity.

Larger Rainbows

Valid configurations include 2 + 2, 2 + 2 + 2, 3 + 3 + 3, 6 + 6 + 6 and uneven compositions such as 6 + 6 + 3.

At a 15% casualty rate:

However, the larger army also has more attack, which changes the casualty percentage. Rainbow sizes therefore need to be calculated against the exact defense rather than guessed.

Smithy Upgrades Can Replace Additional Troops

1A + 1B → 1A + 1B + 1C

Adding a troop is one way to cross a defense breakpoint.

1A(L0) + 1B(L0) → 1A(Lx) + 1B(Ly)

Upgrading the existing two troops is another. The second method retains the one-unit casualty buckets and can be superior when the third troop is expensive, slow, weak offensively, or uses a congested production queue.

Production Efficiency Matters

Combat strength is only one cost. A three-cavalry rainbow may consume three sequential Stable training operations, while a mixed Barracks + Stable rainbow can train both components in parallel.

Huns illustrate the trade-off well: their offensive cavalry can make extremely strong rainbows, but several useful components compete for the same Stable queue.

Advanced Rainbow Farming: Staggered Waves

Rainbow farming becomes considerably deeper when movement speed is added to the calculation. An army travels at the speed of its slowest troop. Several Farm Lists can therefore be released together but arrive at the same oasis at different times.

Release → Wave A → Wave B → Wave C → Wave D

The goal is not necessarily to create identical armies. It is to create a useful sequence of battles.

Calculating Wave Separation

Without long-distance movement bonuses, travel time in minutes is:

T = 60d / v

where d is distance in fields and v is troop speed in fields per hour.

For two armies released at the same time:

ΔT = 60d × |1/v₁ - 1/v₂|

This tells us how far apart the two waves will arrive.

Worked Example: 19 vs 13 Fields/Hour

The same two armies produce increasingly separated arrivals as target distance increases. Distance is therefore part of the timing strategy, not only a travel penalty.

One Release Can Become Several Attacks

Suppose four viable farming groups have effective speeds of 19, 16, 13 and 7 fields/hour and the oasis is 10 fields away.

One synchronized release has created four distinct oasis attacks.

The Speed Governor

A slower troop can be useful even when its attack contribution is modest. If A + B travels at 13 fields/hour, adding troop C with speed 7 changes the entire group to speed 7.

C has now added attack, created another casualty-rounding bucket, and moved the army into a new arrival-time slot. It is functioning as a speed governor.

This means a troop's value in advanced rainbow farming cannot be judged only by attack per resource. Arrival separation can itself be strategically valuable.

Staggered Roman Farming

Romans can construct several distinct speed and strength tiers against the same oasis. A system can include:

• Wave 1 — EI: fast initial hit.

• Wave 2 — EC: different movement tier and greater individual strength.

• Wave 3 — Imperian + Legionnaire: infantry-limited later arrival.

• Wave 4 — heavier Imperian group: slower, stronger follow-up.

All four Farm Lists can be released together. They do not need to be equal and they do not all need to use identical rainbow logic. The objective is to construct a useful sequence of battles.

Stronger Later Waves

A staggered system can deliberately increase strength as arrival time gets later:

Optimization = combat strength + arrival time

Distance Becomes a Strategic Variable

A closer oasis is not automatically superior for staggered rainbow farming. Short distances can cause different-speed waves to arrive almost together. Greater distance can create useful spacing, but it also keeps troops away longer and reduces cycle frequency.

Distance should therefore be treated as a strategic variable rather than simply a penalty.

Tournament Square and Long-Distance Rainbows

Beyond the applicable threshold, Tournament Square changes the movement calculation. The first section of the journey is travelled normally and the bonus applies to the later portion.

T = T(first section) + T(bonus section)

Long-distance staggered-wave calculations therefore need to model the journey in sections rather than multiplying the entire travel time by one bonus. Changing Tournament Square level can also change the spacing between different waves.

The Elephant Exception

Elephant defense is a large jump from the normal Bear/Crocodile/Tiger maintenance range. A tiny rainbow should not automatically be enlarged until it can handle an Elephant, because doing so may destroy the efficiency that made the rainbow useful.

Where an Elephant is present, calculate that encounter separately.

What Makes a Good Rainbow?

There is no single best rainbow without first defining what is being optimized.

The Right Question

Instead of asking, “What is the best rainbow?”, ask:

What is the cheapest composition that produces zero casualties against the expected defense while giving the speed, carry and production characteristics I need?

For advanced farming:

What combination of zero-loss groups gives the greatest number of useful staggered arrivals from one release?

The Three Layers of Rainbow Farming

Level 1 — Rounding

Use different troop types to create independent casualty buckets. Example: 1 + 1 instead of two identical troops.

Level 2 — Breakpoint Engineering

Combine troop choice, quantity and Smithy levels until the smallest possible army crosses the exact defense threshold required. Example: TT L7 + Haeduan L6 rather than automatically adding another troop.

Level 3 — Arrival Engineering

Build several viable farming groups with different limiting speeds and release them together so they arrive at different times.

W₁, W₂, W₃, W₄ released together → T₁ < T₂ < T₃ < T₄

The Complete Rainbow Model

Rainbow = f(troop types, quantities, Smithy, animals, infantry/cavalry, distance, speed, production cost)

The best result is the composition that satisfies the player's objective with the least waste. For simple maintenance that may be 1 + 1. For another oasis it may be 1 + 1 + 1. For heavier farming it may be 3 + 3 + 3 or an uneven composition.

For an advanced farming network, it may not be one rainbow at all:

Several different rainbows → one synchronized release → multiple staggered arrivals

That is the real depth of rainbow farming.

Proposed Calculator Integration

The source guide proposes the following calculator integrations. These are planned examples and workflows, not implemented links or verified presets on this page.

• Open in Combat Calculator — pre-load the exact troops, Smithy levels and animals used in an example.

• Find Zero-Loss Rainbow — select tribe and oasis type and search viable combinations.

• Compare Rainbow Sizes — test 1+1, 2+2, 3+3 and larger groups against increasing animal counts.

• Build Staggered Waves — enter village/oasis coordinates and available troops to calculate viable compositions and their exact arrival sequence.

Related guides and tools

For inhabited targets, see player village farming. Read more about Oases: bonuses, animals and annexing, Smithy upgrades: what they do to your units and Tournament Square and the 20-field rule, or open the battle simulator to enter your own scenario.