UUs only being made at Castles. Is that good game design?

Game design opinion question.

Huns and Goths have UTs that allow their UUs to be built at normal production buildings, but otherwise every castle UU (so UUs not like longboat, legionary, etc) can only be produced at a castle.

There’s no real historical logic that say a CA can be trained at an archery range, but a Mangudai or Kipchak can’t. Are they getting super duper special secret training at the castle or something? The castle that’d be filled with normal joe soldiers. If it was like an AoE1 Academy or something yeah I suppose that’d make sense, but I don’t see any historical justification for the castle only producing UU.

I don’t necessarily have a problem with UUs only being made at the castle, just pointing out I don’t think there’s any justification beyond game design.

So, is it better for the game to have UUs only trainable at Castles or would it be better if they could also be produced at Barracks/Stables/Archery Ranges/Siege Workshops (Hussite Wagon I assume would be made from SW)?

IMO IDK if it makes the game BETTER but thematically I like that the castle is where some of you most powerful options are. UTs, UU, treb. It helps make the castle “cool”.

I guess tangentially related a similar question could be asked about trebs and petards, and if it’d be better they were made at the SW.

Organ gun too.

Anyway, I think the main reason why it was done this way was to make UU’s look cooler and give more reasons to build a castle, but I know it causes many people to almost never use UU’s because they are harder to mass. From a gameplay perspective I don’t like it so much and I would rather have UU’s recruited from other buildings, but at the same time I absolutely don’t want it to be changed now. Nostalgia aside, I think it would change the meta way too abruptly.

Also, if you need historical justifications to have units recruited at the castle, the Samurai is probably the only UU for which it would make sense (I may be wrong, I haven’t really bothered checking all UU’s in my head (EDIT: There’s at least the boyar)) but it would be more logical for knights than it is for trebs and petards.

Historically, probably. Gameplay wise though? Never. This would ruin the balance.


Forget about the Organ Gun. Yeah probably that too. Good catch.

That’s really the only “problem” I see with the current design of UUs being made at castles, and honestly if it is that big of a problem that can easily be fixed by adjusting training times of UUs at the castle.

Still though interested to hear what people think. Are there other pros and cons?

It’s not that easy, though. You’d have 45 different units (including treb and petard) that would need to be changed at once while keeping balance, and I don’t think training time would be the only stat to take into account. That’s something I would maybe like to see changed if I could go back in time and convince the devs while they would be working on AoK, but now it’s far too late.

I think it’s good game design.

It helps keep the different civs more uniform. If you had to look up which building your UU is made in, that’s yet another thing that changes between civs.
Trebs and UUs coming from the castle, making the building relevant, is also a factor.

I think they could have made a game where UUs replace regular units, and then it would have made sense for UUs to be made from regular production buildings. But the current setup works well.


I think it makes sense on a thematic and flavor level. Historically, castles would’ve housed elite soldiers typically belonging to the nobility. Actual history aside, unique units can be thought of as elite noble soldiers. Thus, they are trained at the Castle.


I’m sorry I wasn’t clear. As it is right now, with UUs made at the castle, it’s very easy. Instead of changing everything, if one specific UU is too hard to mass you can adjust their training time.

just wanted to add that kreposts can train konniks.

1 Like

I always feel the design of the entire AOE series cannot well distinguish “unique units” and “elite units”. These two concepts have a lot of overlaps both from a historical and a gameplay perspective.

The “elite units” should be produced at castles to limit their availability. They should have better stats, higher cost and distinct roles.
But actually some “unique units” in castles function just like normal units with some gimmicks, or even trash units.

I’m talking about both the gameplay and the historical background.

1 Like

The short answer is game balance purposes. Castles require stone which is a very limited resource, unless you have access to trade to produce gold to buy more stone from the market.

If you move out UU’s to other buildings the the other units can be moved out too.treb to SW and petard to tc or barracks.Knights and maybe even champions should be moved to castles.Knights are nobility so they will have castles.

Ideally all horse mounted units should be moved to stable and elephant units should have its own building to be trained.

Applying how AoE3 works:

  • Except Petards, all foot units and their upgrades are in Barracks.
  • All mounted units and their upgrades are in Stables
  • All siege weapons and their upgrades, of course, are in Siege Workshops.
  • All foot units, mounted units, and a few types of siege weapons (such as Rams), also trainable in Castles.

Then, try to make the archery range have other functions.

  • Ranged units may train pretty longer than current.
  • Give the Archery Range an aura, like the aura of Folwark.
  • Barracks and Stables inside the aura of Archery Ranges can train ranged units faster.
  • Thumb Ring and Parthian Tactics are still in Archery Ranges.

For me both (1) having UUs made only at castles and (2) having UUs made in a cheap wood only building (with possibly rebalance) are a fine design models. It mainly depends on how much the UU should be in game.

I am very concervative with game design, so I dont want to have most UU available out of castles. Just like I wouldnt want to have warcraft 3 races share common units or heroes.

And I also do not mind the huge imbalance of UU utility between civs, where some civs like Mongols with 95% play UU no matter what the opponent does, and some civs like Teutons with never make the UU no matter the situation.

There do need to be any justification besides “game design”.

Most in game stuff leading to the current game meta is only justified by “game design”.

Castles are already the coolest and most popular unit/building in the game. I expect most of the unranked playerbase to think stuff like “at last in castle age, I can finally make castles!” and “oh no! my opponent has castles ! I cannit do anything until imperial age !”

Oops, sorry I misunderstood…

NP. I wasn’t as clear as I could have been.

I think the answer is more complicated than the design simply being “good” or “bad”. It has some good features and some not-so-good features, works well in some cases, and less well in others.

Thematically, it makes sense (has “historical logic”, if you like) for unique units that represent elite or aristocratic warriors within their civilisation, e.g. samurai, boyars, keshiks, conquistadors. It also makes sense in some other cases: for instance, longbowmen were usually commoners, but the start of the English mass use of longbows coincided with Edward I’s large castle-building projects; the war wagon is (probably) based on the Mangam hwacha, which I think was mostly used to defend fortresses. There are probably other examples like this, where a historical justification can be found.

From a gameplay point of view, I think the key benefit is clarity and consistency of the tech tree – which was important even with only the original 13 civs, and certainly with the 43 we have now, at least if you want random civ to be a reasonable option.

On the other hand, owing to the high cost of castles, the current design works best for the more powerful unique units, like conquistadors, janissaries, mangudai, war wagons – units that it’s worth building a castle for. As a result, many less strong unique units are rather niche. If I were redesigning AoE2 from scratch, I’d suggest that every civ should have a krepost-like building (with a more generic name), but it’s probably too late for that. To me, he krepost seems wasted on Bulgarians, since konniks and knights overlap too much.

Again, there’s no simple yes or no here. A small number of unique units are like this already, with at least four different implementations. There are a couple where I think the implementation doesn’t work well:

Legionary: This only adds visual flavour – from a gameplay point of view, it’s not really different from having a civ bonus for two-handed swordsman. However, the visual flavour is jarring: long swordsmen’s equipment regresses by several hundred years when you upgrade them.

Houfnice: it’s awkward that it has to inherit all the strengths of the bombard cannon. I think it would have been better as a separate anti-personnel unit.

This seems like a special case worth addressing. Hussite wagons are kind of rubbish anyway, but to do the job they’re supposed to (pair with Castle Age hand cannoneers) they probably need to be more easily available as well as receiving a buff. So making them available from the siege workshop seems sensible to me.

1 Like

The bead part of the game design is actually that you often just can’t make use of the UUs as you don’t really get the opportunity and games ending way before UUs can become a thing.

And not that the UUs have these restrictions.

Just have a look on how Mangudai are balanced and well established due to their long training time in castles.
They could never be that iconic if it wasn’t for this inconvienience. And still, Mongols can decide to NOT go for them,

I think it’s important that UUs shouldn’t always be played and the castle restriction is a good way to do it. The only sad part is, that most games don’ even reach the state where you can make this decision. But this has nothing to do with the general UU design concept.

Ofc there are several types of UU which suffer more from the castle inconvenience. Mostly Cavalry, but also Archer UUs. They need to have really nice selling points to make that transition worthwhile. But this is again not the mistake of the castle as production building. It’s the clear indicator that the military timings are atm still super meta dominating. You just can’t have a castle at the beginning of castle age to start producing. But you can make Knights or the xbow upgrade right away.

1 Like

I’m not sure I agree with the full analysis.

You name Mangudai. In most games where I play Mongols, I do actually use Mangudai. Because they’re worth it.

Rattan archers are also often worth it.

Looking at Longbows in contrast, they’re pretty much never worth it. And it’s not because they can’t be worth it, it’s just bad design that’s specific to Britains. Again, Mangudai are usually worth it, so it’s not that the game doesn’t go late enough.
I think if castle-age longbows had 1 extra range, and imp longbows were a bit cheaper to tech into, they’d be worth it sometimes. And that would result in a better game.