Skip to content

Tower Types and What Each One Is Actually For

Arrow, cannon, frost, flame, lightning, siege and chaos towers do different jobs. What each is good against and where each one is wasted.

By Published 6 min read

Reign Creative LLC makes Regal Tower Defense: TD Game, which this article links to, so we have a commercial interest in it. How we write and check articles.

Key takeaways

  • Towers divide into four jobs — single-target damage, area damage, control, and specialist damage — and a defence that is missing one of the four has a wave shape it cannot answer.
  • Slowing towers do not kill anything, which makes them look weak on a damage readout and makes them one of the highest-value slots on a map with a good chokepoint.
  • Splash damage is priced against crowds; against a single heavily armoured target it is usually the worst money on the board.
  • A tower that is strong against everything is a design failure, because it removes the reason to choose, and we cut one of ours twice for exactly that reason.
  • The tower list is the second question. The first is still where on the path it goes.

Every tower in a tower defence game is a bet about what is coming down the path. Arrow towers bet on volume, cannons bet on crowds, frost towers bet that your other towers need more time, and siege towers bet that something armoured is on its way. Get the bet wrong and the tower still fires, still shows damage numbers, and still fails to stop the wave — which is why a defence can look busy and lose anyway.

Here is what each role is for, and where each one is money badly spent.

Four jobs, whatever the names

Strip the art off and almost every tower in the genre is doing one of four jobs.

Role What it is for Where it is wasted
Single-target damage Killing one thing quickly — an elite, a boss, a leaker Against dense groups, where it kills one and ignores nine
Area damage Killing groups that arrive bunched together Against a single high-health target, where you pay for unused area
Control Slowing or holding enemies so everything else gets more shots On a map with no cluster nearby to benefit from the extra time
Specialist damage Beating one defensive property, usually armour or altitude On waves that have neither, where it is an expensive ordinary tower

A defence that is missing one of the four has a wave shape it cannot answer. That is not a flaw in the game; it is the reason there is a shop at all.

The seven towers we shipped, and the role each was given

Regal Tower Defense: TD Game ships seven attacking tower types, and each was written against a specific problem rather than a damage target.

Arrow. The baseline single-target tower: fast, cheap, reliable, and unimpressive per shot. It is the tower you open with because it converts a small budget into consistent damage immediately, and it is the tower you stop buying once armour appears, because armour taxes every individual hit.

Cannon. Area damage. Built for the moment a wave arrives as a column rather than a queue. It rewards narrow ground and punishes you on the boss wave if it is all you have.

Frost. Control. It is not there to kill; it is there to make every neighbouring tower better by extending the time enemies spend in range. A slow is a status effect in the ordinary game-design sense — a temporary modification of a unit's behaviour — and its value is always measured in what happens to something else.

Flame. Sustained damage over an area, which makes it strong against tightly packed weak enemies and weak against anything that crosses its zone quickly.

Lightning. Chain damage that jumps between nearby targets. Excellent against grouped enemies, and the tower we had to be most careful with — more on that below.

Siege. Slow, heavy, long-range single hits. This is the armour answer, and the trade is a rate of fire that makes it poor against swarms.

Chaos. The late-campaign specialist, built for the point in the Chaos Realm where waves stop being one kind of thing.

There is an eighth category that does not attack at all — Aura Towers, which raise the damage, attack speed, range or elemental power of the towers around them. Because they change the value of every tower nearby rather than doing a job of their own, they get their own explanation of how support stacking works.

The tower we cut twice

Lightning was the problem child.

The pitch was simple: a tower that chains damage between nearby enemies, so it rewards you for building at a chokepoint where enemies bunch. What actually happened in internal play was that a chain tower with generous numbers quietly did Cannon's job (it hit groups), Flame's job (it hit them repeatedly), and a fair share of Arrow's job (it hit whatever was nearest anyway). Three towers became decorative.

We cut its damage once, watched it stay dominant, and cut it again along with the chain count. That is not a heroic story — it is the ordinary, iterative business of game balance, which is the design work of adjusting rewards and challenges to produce the experience you intended rather than the one that emerged. The point of the second cut was not to make Lightning bad. It was to make the sentence "should I build Lightning or Cannon here?" have an answer that depends on the map.

If a tower in any game is correct everywhere, the shop has stopped being a decision.

Reading the wave instead of the tier list

The practical skill is not memorising which tower has the highest number. It is looking at what is about to arrive and asking three questions.

How many? Many weak enemies arriving together is area-damage ground. One or two large ones is single-target ground.

How tough per unit? Armoured units reduce incoming damage per hit, which means slow heavy shots keep most of their value and fast light shots lose most of theirs. This is the single most counter-intuitive interaction in the genre, and it is why adding more arrow towers to an armour problem feels like it should work and does not.

How fast, and how far? A fast enemy spends less time in range, which raises the value of control and lowers the value of anything with a wind-up. Our later worlds mix wolves and drakes into slower formations for exactly this reason.

The enemy-side version of all three questions — armour, flight, elites and how mixed waves are built — is covered separately, because it is genuinely a different way of looking at the same board.

Where the tower list stops mattering

It is worth being blunt about the limit of this article. Choosing the right tower type is the second question. The first is still where it goes, because a perfectly chosen siege tower on a straight with three seconds of dwell time will lose to an ordinary arrow tower sitting inside a switchback. If you have not read the path yet, the shop cannot help you — why placement beats damage is the prerequisite.

The other thing a tower list will not tell you is the order to buy in. Our campaign runs fifty levels across ten worlds, and the reason the same seven towers stay interesting across all of them is that each world changes the path shapes and the enemy mix rather than simply inflating health bars. A tower that carried Stone Pass can be the wrong purchase in Frozen Crossing with identical stats.

Players coming from run-based games sometimes expect towers to work like roguelite power-ups, where the aim is to stack multipliers on a single build. They do not. A tower is a fixed position on a map, and its value is set by the ground under it as much as by its own numbers — how in-run power-ups actually combine covers the other model, and the contrast is instructive.

The short version

  • Cover all four jobs before you optimise any of them.
  • Build area damage where the path is narrow and the wave arrives bunched.
  • Treat control towers as multipliers on your best cluster, not as damage.
  • Answer armour with heavy single hits, not with more light ones.
  • Distrust any tower that seems correct on every map — including ours.

The campaign, and the rest of the studio's short-session titles, are listed under Action & Arcade Games.

Good to know

Frequently asked questions

What are the main tower roles in tower defence?
Four, regardless of what a particular game calls them. Single-target damage kills one thing quickly and is your answer to a tough individual enemy. Area damage hits a group and is priced against crowds. Control does little or no damage but slows, freezes or otherwise extends the time everything else has to shoot. Specialist damage exists to beat one specific defensive property, usually armour or altitude. A defence missing one of those four has a wave shape it cannot answer.
What does a frost tower actually do?
It applies a status effect rather than killing things. Slowing an enemy increases the number of seconds it spends inside every other tower's range, which means a frost tower's real damage output is the extra damage your other towers get to deal. That is why it reads as weak in isolation and is worth a premium slot beside a strong cluster, and why two frost towers side by side are usually much less useful than one frost tower beside two damage towers.
Are splash damage towers worth building?
Against groups, yes — that is the entire point of them. Against a single high-value target they are poor, because you are paying for area you are not using. The practical rule is to build them where the path is narrow and the wave arrives bunched, and to avoid leaning on them for a boss. If your defence is made mostly of area damage, a wave of one enormous enemy will walk straight through it.
Which tower handles armoured enemies?
The one your game marks as armour-piercing or siege, usually at the cost of a slow rate of fire. Armour in most tower defence games reduces incoming damage by a flat amount per hit, which punishes fast, weak shots far more than slow, heavy ones. That means the answer is rarely 'more arrow towers', because each individual arrow loses most of its value. The enemy-side view of this is a separate topic from the tower list.
Is there a single tower that beats everything?
There should not be, and if there is one in a game you are playing, it is a balance bug rather than a strategy. The genre only works if the choice of tower is a real choice, which means every tower needs something it is bad at. We cut one of ours twice during development precisely because it was quietly answering three other towers' problems as well as its own.

Sources

  1. Tower defense (opens in a new tab)Wikipedia — accessed
  2. Status effect (opens in a new tab)Wikipedia — accessed

About the author

— The independent app studio that builds the apps covered on this site and publishes every article here. On Google Play it publishes as Reign Collective Apps.

We use AI tools to help research, draft and edit our articles. Each article lists the sources its factual claims rely on, and details about our apps come from their live Google Play listings. No article on this site has been reviewed by an independent expert unless the reviewer is named on it.

Spotted an error? Tell us — how corrections are handled is set out in our editorial policy.

Keep reading