Recruitment dev topic

Well I've been a bit quiet lately, but at least I crunched through a bit of stuff about how to organize unit recruitment in II.
The current idea is this:

Step 1: Urban/Rural split

I don't know how far you're aware of it, but I'm planning to let the player decide on whether he wants to concentrate more on the rural or more on the urban aspect of a province fairly early in city development (this will be simulated by using the castle/city feature).
The theory behind this is that some factions, especially the city states, will concentrate heavily on controlling cities rather than controlling the surrounding land. They will build up defensive structures in these cities, and they will recruit most of their units (and probably their better quality troops) here. Cities will have militia armies which you can imagine in the game as standing armies which are very expensive to maintain in the field but fill free upkeep slots while in cities. Thus, cities will be able to mobilize forces very quickly, but they will face problems after their forces are depleted because the replenishment rate is extremely low.
Conversely, rural provinces won't have many standing troops (maybe one or two elite units). What they get instead is the levy script which allows them to call armies of mostly low-experience units when needed. They, too, will take some time to replenish but it will be shorter than that of the missile troops. Most rural provinces will be found in more feudal areas like Burgundy, with the lowest density in northern Italy where most provinces will be urban-focused.

Step 2: City focus

Out of the box, each settlement will be able to recruit small quantities of every troop type without building stuff like archery ranges or whatever. I assume that pretty much every province would be able to cater for some equipment, horses, etc. The special buildings CA uses are in my opinion a remnant of the Age of Empires styled games where you need these for every unit type; on the other end of the spectrum, we can find civilization where every unit is recruitable in every city.
However, I don't want to completely remove this aspect of building up your provinces so what we thought of was to introduce focus decrees that allow you to concentrate on a certain aspect of warfare in this settlement. This decision will shift recruitment pools, experience and unlock more advanced units under certain circumstances. The first specialization level will be expensive and take a long time to build, so you should be wary of changing focus during the game as they will be mutually exclusive.
What I currently have in mind for these buildings is this:

Missile Focus

The missile focus tree allows you to gear a city towards recruiting high-quality missile troops. It will increase the replenishment rate and experience for missile units, but decrease the growth of cavalry and infantry units. For the levy script, it will increase the spawn chance for missile units relative to the other types resulting in a larger percentage of the army carrying ranged weapons (and they will most likely be more experienced)
It also allows you to further specialize the city by unlocking Crossbows and Archery focus.
This building represents a number of measures the ruler could pass to generate more missile troops in a province, such as settling bowmakers, increasing food allotment for missile troops during campaigns, allow people to hunt game, etc.

Archery Focus

An upgrade for the missile focus building, this focus allows you to further concentrate on bows. This will allow you to recruit the best bowmen available, and shifts the recruitment chances from crossbowmen more towards archers. A lot of settlements in Sicily will have this focus from the beginning to represent their strong archery tradition.
Historically, this would either represent a tradition of using the bow, as in Sicily, or deliberate steps taken by the government to reward soldiers who are good with a bow. These include things like tournaments and might go all the way to forcing the people to train with the bow regularly.

Crossbow Focus

Similarly to the Archery focus, this is an upgrade for the missile focus. It has pretty much the counter-effect by increasing the number of crossbowmen in your army while decreasing that of other types.
This focus will be mainly found in the technology-friendly northern Italian cities where crossbows were historically very important.


Infantry Focus

This building presents the player with an opportunity to recruit more and better infantry troops. It can be thought of as sponsoring smithes who specialise in melee weapons, better training facilities, or simply a strong infantry tradition. A lot of the better infantry units (better and more expensive equipment) can be recruited in larger capacities, but this comes at the cost of neglecting the other unit types a bit. This can be further upgraded by concentrating on the manufacture of polearm-type weapons like pikes.

Polearm/Spear focus

I think this will only be available later in the game when the technology and expertise to recruit well-drilled units bearing pikes, halberds, etc. exists. Ihis building will allow you to churn them out much more quickly and in better quality. Unfortunately it will also drastically reduce your ability to recruit other unit types, so it's a bit of a gamble.
Note that I have to investigate if there's a way to make pikes work properly for this building to make sense.


Cavalry focus

I'm not quite sure if I want to make this available in urban settlements at all, but in general it allows you to field high-quality cavalry troops, such as knights, in larger quantities.
Historically, this can represent fostering horse breeding capabilities in certain areas, giving fiefs to have more knights able to pay for a horse and equipment, and other similar measures.

Light Cavalry focus

An upgrade for the general cavalry focus, concentrating your settlement on light cavalry means that you will support the breeding of faster horses, allow multiple citizens to band together and pay for one mounted soldier, etc.
The benefits are obvious: This will enable horse archers and quick cavalry units where it's appropriate and we all know how strong they are when used right.

Heavy Cavalry focus

An alternative upgrade for the cavalry focus allows you to further go down the route of having heavily armed and armored mounted units. In a lot of cases this will mean that you get to recruit knights in shiny armor, but it might also decrease your income (which would represent the large number of fiefs that you won't get any benefits from anymore).


Training focus

This building has a few different levels. Rather than specializing your province on a certain unit type, it has been decided that the populace should be trained better in the weapon types they prefer to use instead.
It will increase experience through the bank, and possibly unlock certain well-trained units (Tuscanian militia units, elite bodyguard units and so on).
Historically, this represents an effort to drill the soldiers better and make sure they will stand their line rather than encouraging a certain type of weapon to be prevalent in this area.


Keep in mind that the different trees and buildings inside the trees will be mutually exclusive, so you can't have more than one focus per province. I think that this system will make for important and interesting gameplay decisions and at the same time allows us to represent army traditions which exist at the start of the mod's timeframe.

I'll get back to you when I have further additions to this. Right now I'm implementing the system for the generic early unit roster I put in a while ago to test stuff (and which only contains the most basic units), which should give me a decent idea of how well it works and if the AI can actually cope with it.

Alright, here's another update for those of you who are still following our progress:

I decided to drop the levy script idea (at least for now). The script as I intended to do it would be too huge to sensibly be implemented for even a low number of units.

I found a really good replacement system using the normal way of recruiting units, however this of course means that one of the main ideas of the levy script (not being able to influence your army composition too much) is not valid anymore, but it's not as bad as I wasn't entirely sure that it's a good idea anyways since it may take away too much freedom from the player.
So, the new system works like this:

Urban cities generate (relatively) high experience militia units. These units have free upkeep, therefore you'll usually keep them in your cities and use them for defensive purposes or maybe for limited campaigns as they are very expensive to maintain in the field (a measly peasant unit will be about as expensive as an elite unit is in vanilla). Since they're a kind of a standing army, they'll of course get to keep their acquired experience, too, which means you'll arrive at pretty good armies using them. They won't cost anything to recruit because it's every citizens duty to answer a call to arms.
There are a few disadvantages to the urban recruitment system to balance for that:
- Units grow slowly
- Pool caps are really low
- You can only recruit or retrain one unit per turn

Rural cities on the other hand recruit a lot of cheap units very quickly. They are not quite as expensive to maintain (still much more costly than in vanilla though) but aren't usually very experienced. Castles will get to recruit up to 10 units per turn (which are not free upkeep though), which means that instead of keeping a standing army in them, you'll be able to pump out a decent-sized army within a few turns in case of war. They won't cost much to recruit though, the main cost is upkeep not recruitment because soldiers would have to pay for their own equipment.

The system in general works out very nicely. You'll use your cheaply drafted castle units first, then possibly send in some more experienced city veterans to carry on the fighting. Especially valuable are missile militia troops because they won't suffer so many casualties.
It's also designed in a way to make sure that wars won't last too long because after you expended your initial invasion force, recruitment will be slow indeed (on a magnitude of waiting something like 10 turns for a new unit in an urban city) and you will be very vulnerable to attack from a third party if you overextend yourself.


Before I forget it: I also implemented a global limit on diplomats for each faction (3, but you can recruit some more before the script kicks in properly so you'll usually have something like 4 or 5)

Particularly like the idea of a cavalry/archer/infantry etc tradition.

Also i think the initially large but very slow replenishment type recruitment idea could make for a much more realistic feeling game where you get decisive battles followed by the losing side wanting peace and a long lull while they waited to rebuild their army.

Sounds cool.

Quote:

---

Originally Posted by nikolai1962

Particularly like the idea of a cavalry/archer/infantry etc tradition.

Also i think the initially large but very slow replenishment type recruitment idea could make for a much more realistic feeling game where you get decisive battles followed by the losing side wanting peace and a long lull while they waited to rebuild their army.

Sounds cool.

---

Yeah I was actually playing a bit with it already (recruit pools are still a bit high in my current build but I'll lower them later) and it at least feels like an army is worth something.

Another important point to note is that when you disband a unit in a settlement, the pools will increase in vanilla. I'm currently wondering whether I should keep that or disable this behaviour (there's a setting for it).

Alpaca, I find your Focus idea highly creative, big props on that.

What was wrong with the levy script size? I mean once you take into the original large size, what about these new ideas (rural/urban) made it balloon much more? (I'm trying to remember the details of how it worked...)

Also like the low replenishment idea. Any concrete examples with campaign tests about that?

Quote:

---

Originally Posted by SigniferOne

Alpaca, I find your Focus idea highly creative, big props on that.

What was wrong with the levy script size? I mean once you take into the original large size, what about these new ideas (rural/urban) made it balloon much more? (I'm trying to remember the details of how it worked...)

Also like the low replenishment idea. Any concrete examples with campaign tests about that?

---

Well the levy script how I had planned it would need quite a few executions for each settlement/unit/faction combination. Add to that additional changes to the unit weights for different buildings and the like and you quickly end up with hundreds of thousands of lines of code, not all of which can be easily created with automation.

The main problem is not the size per se but the time I can spend on other features which are more interesting (because as I said I'm not even sure that I want to limit the player so harshly). In addition it could be that the script will stop the player from doing anything for a few seconds whenever he triggers it, and that's not very good, either.
I realized that the levy script was mainly interesting to develop, but it's mediocre from a gameplay point of view. Since I have to be somewhat economical with my scripting time I dropped it.

The low replenishment right now is mainly a concept, I haven't played more than maybe 20 turns into a campaign yet because I have to do some work on the AI and features before really going into testing things, but so far it felt quite interesting to actually be forced to be careful about not losing half of your army on one front because this would mean you'll be open for backstabbing on the other side. Incidentally, it also makes stable alliances more interesting and since I'll try to make the AI heed marriage agreements most of the time, they'll become very important for keeping your back free from danger.

I forget, the only way increase or change pool replenishment %, is by script right? Or is there some building attribute that I'm not remembering at the moment.

Also, would be nice to brainstorm how farms might affect replenisment % along with the population rate. That would be really awesome, all of this adding up to a much more realistic portrayal of the realities of the world.

Quote:

---

Originally Posted by SigniferOne

I forget, the only way increase or change pool replenishment %, is by script right? Or is there some building attribute that I'm not remembering at the moment.

Also, would be nice to brainstorm how farms might affect replenisment % along with the population rate. That would be really awesome, all of this adding up to a much more realistic portrayal of the realities of the world.

---

Didn't quite get what you mean there. If you're basically asking how this is coded: It's only using complex conditionals for the recruit pools in the EDB.
However, this will only work if you use a building with no other bonuses, which is another reason for the focus building tree.

Interesting idea about farms influencing recruitment. I didn't get around to properly work on tech trees yet but I have a growing feeling that I should :laugh4:

Quote:

---

Originally Posted by alpaca

Didn't quite get what you mean there. If you're basically asking how this is coded: It's only using complex conditionals for the recruit pools in the EDB.
However, this will only work if you use a building with no other bonuses, which is another reason for the focus building tree.

---

I see, so you mean something like:

recruit spear_militia AT CERTAIN RATE if event_counter CERTAIN_RATE =1
recruit spear_militia (same unit, in same building) AT HIGHER RATE if event_counter HIGHER_RATE=1

?

That's probably a better idea than using scripts. As you know, you can lower or raise pools using scripts (at least in Kingdoms), but that's probably less efficient, and more time consuming, than just defining it in the building using event_counters.

Another idea I thought of just now, that would utilize the ability of changing unit pools: if something bad happens, a script would set those pools to zero. Let's say you link it to barracks, and if barracks is destroyed (sabotaged by a spy) then all of the recruitment pools in a city become null, since units cannot train or muster there anymore.

Quote:

---

Interesting idea about farms influencing recruitment. I didn't get around to properly work on tech trees yet but I have a growing feeling that I should :laugh4:

---

Wouldn't that be nice if very high farms development would guarantee that your troops would replenish faster? I think that's a cool addition to an already great idea.

The thing isn't about farms giving better bonuses but about how a "farming" building should be implemented in the first place. Building farms wouldn't be quite what you do to increase agricultural output, it'd rather be stuff like passing decrees to settle people in unsettled or sparsely inhabited areas, building granaries, developing technologies and so on.

What I mean with the recruit pools is stuff like this:

Code:

---

recruit_pool "crossbow_militia" 0 0.002 0.5 3 requires factions { all, } and building_present core_building and building_present_min_level control_level control4 or building_present ai_control

---

But yes, these can be linked to event_counters, too. I'd like to create some random events with that stuff.

I don't intend to keep spies in the game as spies by the way. They're just too powerful. I'll reveal the whole map at the start of the game, then you'll have to make do with diplomats (who will have a larger line of sight than in vanilla) but are capped to a certain limit. I'd like a system best where I can restrict spies to, you know, spying, but I didn't see a way to do that.
I'll probably also limit assassins somewhat because their sabotage ability is ridiculous, too. One assassin can disable a building, so it should be pretty expensive (let's say at least as much as it costs to repair the building).

Quote:

---

Originally Posted by alpaca

I'll probably also limit assassins somewhat because their sabotage ability is ridiculous, too. One assassin can disable a building, so it should be pretty expensive (let's say at least as much as it costs to repair the building).

---

That's not realistic. It won't take as much resources to train a saboteur as it takes to make a whole building. Probably you could even just recruit some low-life that had nothing to lose. Saboteurs are used because they are effective, despite their rather low chances of success (or at least of getting away after the sabotage). If you want to make it harder for them to succeed, I guess a few law bonuses in settlements will fix that easily and would IMO be better than removing assassins in general.

Quote:

---

Originally Posted by bovi

That's not realistic. It won't take as much resources to train a saboteur as it takes to make a whole building. Probably you could even just recruit some low-life that had nothing to lose. Saboteurs are used because they are effective, despite their rather low chances of success (or at least of getting away after the sabotage). If you want to make it harder for them to succeed, I guess a few law bonuses in settlements will fix that easily and would IMO be better than removing assassins in general.

---

Well if it was so easy to let important buildings crumble to dust using saboteurs, why are any of them still standing today? I don't think historically they were a very powerful tool. They might have caused some mischief but I don't believe they were able to cause very serious problems usually, despite the high fire risk in cities at the time.

What factors, I wonder, affect the likelihood of sabotage? Is it the law_bonus?

Quote:

---

Originally Posted by SigniferOne

What factors, I wonder, affect the likelihood of sabotage? Is it the law_bonus?

---

A shot in the dark: Public security would be a candidate.

Setting fire to buildings would (probably) usually not destroy the whole building if the victims were vigilant, but could destroy its function for a time. In the game this would be the one turn it takes to repair. However, I fail to find any resources on sabotage in medieval times, so I guess you're right that it didn't play a very large role until the invention of explosives. Making them cost as much as a building doesn't make any sense though, even if this function should be less effective.

Assassination was rampant however. And later, in the renaissance, even many regents were killed.

Can you add public security bonuses to buildings, or only characters? This sounds like a good way to discourage sabotage. It would also hinder assassination in the settlement, but allow assassination outdoors.

Even before the Renaissance. I can tell at least two major starting characters in the mod were assassinated : Boniface of Canossa and the young burgundian heir, Rodolphe.
What I think doesnt work with assassins is that they have to train by butchering whole family trees before being even usefull which is completely irrealistic. I think, they should be a one shot attempt with larger consequences to diplomacy and/or influence and a training level only based on a focus or a ruler trait (or anything else but not by being a serial killer, cause its not hitman). Sabotage I dont see any reason to keep in that state and since I guess it will be another one of these hard coded thingies we cant mod, its probably better to make it unusable IMO.

Quote:

---

Originally Posted by Solo

Even before the Renaissance. I can tell at least two major starting characters in the mod were assassinated : Boniface of Canossa and the young burgundian heir, Rodolphe.
What I think doesnt work with assassins is that they have to train by butchering whole family trees before being even usefull which is completely irrealistic. I think, they should be a one shot attempt with larger consequences to diplomacy and/or influence and a training level only based on a focus or a ruler trait (or anything else but not by being a serial killer, cause its not hitman). Sabotage I dont see any reason to keep in that state and since I guess it will be another one of these hard coded thingies we cant mod, its probably better to make it unusable IMO.

---

Agreed, assassination should stay in by all means (and yes, their skill should just be randomly, maybe influenced by some buildings or governor traits) - however sabotage is too powerful for balance reasons. It wouldn't be so bad if the AI would use it, too, but the way it currently works just isn't good.

I wonder if you could transfer the traits' features into buildings. Thus you could assign public_security 100 to even the lowest level of walls, and thus make sabotage practically impossible.