Couple questions about castles

[Severus]

Its becoming apparent that its completely unnecessary to have but a couple castles for troop production, as I don’t seem to need castles for their defensive capabilities (Its a shame the AI is so ineffectual when performing sieges) In the game the main function of castles is to produce non militia type troops. While its true historically that barons would have knights on the payroll residing in the castle, a castles function wasn't to build armies. The only advantage I see to having a bunch of castles in the game is to produce troops closer to the action. I know it can depend on circumstances, but I’m curious, how many settlements do you guys leave as castles in the long campaign? Also when converting from one type to another do you have to manually destroy the buildings? I was thinking it would make sense if the destruction and recoup of funds was automated.
I'm disappointed as i was hoping castles would be an exciting innovation in game play. During the late middle ages the construction of castles were often an absolute necessity to establish strong centralized rule against local fragmentation, and especially in newly conquered lands. In the couple games i have played so far the only thing castles are a necessity for is producing troops.

[PBI]

Castles have much better public order than cities, so they are quite useful to have in very far-flung, relatively poor provinces such as Jedda or Tripoli to avoid having to maintain a huge garrison to keep order (a castle will almost never rebel even if it has a garrison of only 10 peasant archers say). So they are useful for securing control of newly conquered lands, as you suggest they should be.

I think most people find they end up with many more cities than castles as their empire expands. I usually aim to have roughly one castle per "region", i.e. one in Spain, one in France, one in Britain etc, with the rest cities. It's quite useful to have castles on your borders however as although it is easy enough to defend a city, defending a citadel is even easier; also, I think the AI may be more averse to besieging castles than cities, so it should help to deter them from invading.

[Ratwar]

Personally, I've found castles to be quite useful for border defense. I mean, once you can upgrade an area to a Minor City, sure they aren't that useful, but a castle with Stone Walls is much easier to defend than a Town which can barely kill anything. You can really see this in the Russian campaign. If you turn all of the easter castles into towns, the Mongols will roll right through them.

Also, I like to have some near my border for troop retraining needs. It is a lot easier to fall back a one-turn march than spend ten returning to your troop base.

[Severus]

I certainly agree that a castle is much better for defense relative to pallisades but once my cities have a good set of walls up feel pretty darn confident. I didnt realize public order was given a bonus for castles, i just assumed that it was due to the moderate tax rate. It would make sense for the computer having some apprehension attacking castles, just because its a castle, even if they arent well garrisoned.
Im actually playing a long campaign (first time, havent had the game long) with Spain where i have conquered a couple cities next to Jerusalem (I gifted the holy city to the pope, There was a heretics and Imam festival going on so he can have it until i get some cardinals and assasins overthere!) Since they are so far away from Spain do you think its best to convert these to castles? They should be excellent cities to develop trade, but I have just gotten the "mongels are coming" message so maybe that wouldn't be prudent. btw if anyone off the top of their head knows about possibly recouping florins when conversion takes place id be interested to hear!

[Rhedd]

I never convert cities or castles. Ever.

Just take them as they are, no matter what.

Not only does this put you on much more even footing with the AI, but it makes the game more interesting, as decisions on what province to take next can be greatly influenced by its settlement type.

[Monsieur Alphonse]

I always try to group some castles. For instance Caen / Angers or Bern / Staufen or Hamburg / Magdeburg. This way I can train a stack within a couple of turns. Castles at the front line will be useful not only for defense but also for retraining depleted units.

[Rhyfelwyr]

Troop production is definetely the main function of castles, but thats still a pretty important rule. I got a bit overconfident with my militia armies at Venice, and was forced to have a bit of a rethink having been gubbed by an army of Feudal Knights etc.

I never convert cities to castles, although I don't do the opposite that often either. I tend to convert Inverness a lot since its too far away for troop production. Also Oslo, Helsinki, maybe Stettin, Iasi, Cagliari, Valencia tend to get converted by me a lot.

As someone else said, if you keep castles in groups they can levy a full army pretty much in two turns. A great example of this is the combination of Metz, Staufen, and Bern. Park a General in between them and you can group together a full army right in the centre of Europe. This grouping is in a great position for all the big teams eg France, HRE, Milan, England to use.

Although castles are best on the front line, by the time you have expanded enough to have a large frontline you tend to be so rich it doesn't matter, and it costs a bit to convert them anyway.

[Ramses II CP]

You'll make a dramatic amount more money if you convert seaside castles into cities, also you can get to a citadel much more quickly if you convert cities that start with a high population into a castle. Try to position your castles so that you can reinforce a selection of important cities quickly.

Also once you've conquered certain areas, the British Isles for example, it's wasteful to leave castles laying about. Even if the AI sends a sea invasion force the siege AI is so pathetic you're not going to have much trouble defending with your militia garrison, especially since you can combine them from several cities.

[mir]

I have to add that for some factions, Cities produce better troops.

As the Turks, I find that only my highest-level cities can produce Janissary Musketeers, quite possibly one of the best units around.

So I only keep one Castle in each region to produce my Qapukulu heavy cavalry...

[Monsieur Alphonse]

I have to add that for some factions, Cities produce better troops.

As the Turks, I find that only my highest-level cities can produce Janissary Musketeers, quite possibly one of the best units around.

So I only keep one Castle in each region to produce my Qapukulu heavy cavalry...

And Naffatun

[mir]

And Naffatun

Oh yes, I agree! Nothing like the smell of fresh roasted cavalry early in the morning... Mmm-mmm!

Having said that, the Turks do have a slight advantage that their Racetracks allow them to train Sipahi's quickly. And horse archers are all I need sometimes...

[Askthepizzaguy]

I have regional recruitment centers as castles, here is how I determine where they need to be:

When a region as large as France has no castles, I take the largest town which has poor trade and make it a castle. Poor trade = Perfect castle province, as it is otherwise a worthless settlement.

Closer to the center of the map, the oceans, and the front line against the enemy, I will leave existing castles as resupply stations, and convert them to towns later.

converting Towns to Castles:

Landlocked = good
Poor trade = better
Largest size before can become a city = best

Converting castles to cities:

Has large population = good
Has port access = better
Great trade resources = best
There is already a main recruitment facility in the region and the area is peaceful/well defended = Best candidate, when money is available.

Example:

In the Broken Crescent Mod, the Turkish Sultanate starts with a very large town called Konya, which is landlocked and has no really useful resources. Being the largest city, it's the best trader, but it's pathetic long term. Turning it into a castle also provides you a way, with barracks, to quickly gain access to the Turks' best infantry, which no other settlement can produce for a while, as the Turks' main castle doesn't grow very fast.

If I've recruited enough merchants and spies and I don't need the extra 700 gold/turn income from the trade difference between a castle and a city in Konya, It becomes a permanent castle, paired with the one directly north of it for raising between 10-12 castle/fortress level troops within range of reinforcing one another every single turn. These two provinces alone could rebuild an entire army in two turns, and Man of the Hour provides unlimited loyal, skilled generals. So if you spit out a few pieces of deadweight troops and send them over there, they can rebel and you can smash them for free generals.

Thats how you turn two otherwise worthless regions into your military command center capable of singlehandedly taking on the entire world.

[FactionHeir]

What ATPG said.

One other thing that's important in the conversion decision is that you don't want to start converting settlements that are well developed (both castles and cities), which means that generally you don't convert post turn 50 or 60 as then you will hardly regain the total loss in time for it to matter.
The best time to convert is within the first 20 turns when things are relatively undeveloped and frontlines still mobile.

[Diehard_TH]

Agree, that castles are pretty useless and all that is required is one per 'region'. As my frontline moves outwards a convert any redundant castles to cities, especially if they have large populations or are on the coast.

I basically only leave 'Elite' castles that i have upgraded to produce certain types of units, the rest i convert at the earliest opportunity.

I was surprised to fight my first defensive seige though, as the AI in MTW1 never initiates seiges.

[Martok]

I was surprised to fight my first defensive seige though, as the AI in MTW1 never initiates seiges.

Actually, it did, although it was admittedly still rare. The only time the AI has assaulted my castles is when I have only a very few men (less than 100), and the enemy army outnumbers my garrison by a significant ratio (probably around 5:1 or better).


Out of curiosity, what was the force ratio in your recent (M2TW) siege battle?

[Lord Fluffy]

I just had a small force of turks, perhaps around 500 men, laying siege to Aleppo, with a defending garrison of a tad over 800. This is M2TW vanilla v1.2

Hungary has been sieging Constantinople, several times they had only 300 men against my 2000 men garrison.

[CountMRVHS]

Like Rhedd, I almost never convert cities or castles.

The exception to this is when I think one place really "ought" to be a city or a castle. For example, as the English, I converted Bordeaux (I think that's it) from a castle to a city, as the Gascony region (SW of France at any rate; my names might be wrong) was a financial powerhouse for the English, and having a city allows better trade, etc, obviously.

[Rhyfelwyr]

I've discovered as Portugal that Toledo makes an excellent city. As a Huge City, Toledo is bringing about 4,500 Florins each turn. Trade is at over 3,000.

Probably the single highest income I've ever had from a city. Definetely worth converting as soon as possible.

[TheLastPrivate]

For some reason valencia doesn't make as much money even tho it has ports open to the rich medterrenean sea trade...

[Monsieur Alphonse]

I've discovered as Portugal that Toledo makes an excellent city. As a Huge City, Toledo is bringing about 4,500 Florins each turn. Trade is at over 3,000.

Probably the single highest income I've ever had from a city. Definetely worth converting as soon as possible.

I am playing the retrofit mod. Toledo as a huge city is making over 6000 florins (without a governor).

[Old Geezer]

I always convert Bordeaux and Metz to cities. I leave one good castle in a region and convert the others to cities when I have the cash. Jedda and Dongala I convert and grow sometimes to get trade up for Cairo. Arguin I have been leaving as a castle to produce the troops needed to keep the area under control. IIRC Timbuktu starts out as castle and I always convert it. All the extra cash from the cities makes it much easier to crush the Mongols and Timurids.

[G^2]

Like many on this thread, I also rarely convert castles and cities. Mostly because its allot more simple in terms of troop production and unrest to have about 1/3 or your provinces castles. That being said, often I will convert a castle on the coast into a city, providing there is another castle in an adjacent providence. Hamburg, Inverness, Bordeaux, Corith and Sicily (Palmareo?) come to mind. When playing as Vince, I often buy Bologna and convert it into a castle. 1) It is within one march distance from four cash cows in northern Italy. 2) It prevents Sicilian aggression from the South. 3) I've always had a large amount of unrest when I've left it as a city. This is the only city I convert to a castle. Toward late game I take settlements as they come for the most part.

[Philbert]

I mostly agree: I convert most castles into city except for settlements with a strategic position off the coast, like Bern.
However, I do find that in the middle east I generally leave all 3 castles (Gaza, Accre and the other one...) as castles to produce the troops I need to counter the Mongols.