Garrison Mod for 0.9.2

**john_peter** · August 10, 2017, 03:53 PM

What if I want garrison script but not that much? Reading the description it seems every time I attack a settlement it will raise a large force, I don't want that, I just want the AI to have a extra small-medium force when I attack them.

**tmodelsk** · August 12, 2017, 02:48 AM

Originally Posted by john_peter

What if I want garrison script but not that much? Reading the description it seems every time I attack a settlement it will raise a large force, I don't want that, I just want the AI to have a extra small-medium force when I attack them.

john_peter - currently it's not customizable but I can do that if you help me with specification how it should work.

We have three 'scenarios' : Player vs AI (your request), AI vs AI , AI vs Player .
Lets assume we have four logical garrison sizes which will be created : no at all, small, medium, strong .

How do you think it should work in scenarios AI vs AI , AI vs Player ?

**john_peter** · August 13, 2017, 05:14 PM

Is it customizable for each city or just equal for everyone? Like, an important fortress or large city, the garrison should be strong (only if the army there is not strong already) but a minor and castle should be medium (AI vs AI and Player vs AI)
AI vs Player, small garrison in castle, fortress (only one of them could be some decent infantry? Not heavy infantry) and minor city, medium garrison in large city.

**Jurand of Cracow** · August 14, 2017, 12:54 AM

Originally Posted by tmodelsk

We have three 'scenarios' : Player vs AI (your request), AI vs AI , AI vs Player .
Lets assume we have four logical garrison sizes which will be created : no at all, small, medium, strong .
How do you think it should work in scenarios AI vs AI , AI vs Player ?

I think you need to take a few things into account:
- the player knows about the mechanics and can take countermeasures. Additionally, the player can learn from his own mistakes: if he fails once, the next time he will bring along more troops. The AI cannot neither: it stays dumb, it carries on doing the same mistakes.
- for the AI it's already now difficult to expand (recall all the players' reports on the passivity of its big armies for long times - Lithuania often just takes scripted Riga and then does nothing...). We shouldn't make it worse.
- the main aim of introduction of the garrison mechanics should be to ensure the player cannot take easily the AI settlements due to the stupidity / poor programming of the AI. It's just about fixing the engine - ideally you'd like the AI and the player to keep adequate garrisons all the time (or take a deliberate risk of losing a city after depriving it of a good garrison). The defence should be just by design. Actually, the SSHIP does for me pretty good job: I keep forces for the reason of keeping the unrest in check, but the side effect is: the settlements are adequately garrisoned.
- we should also not break our historical immersion by adding additional units coming from the Mars or something.
- the second, lesser aim of this mechanics would be to bolster up player's historical experience by meeting always garrisoned settlements (ie: in all settlements) with a delicate scaling of the size of the garrison according to the size of the settlement (more units in the bigger settlements, especially castles - they were always difficult to take down). It should also add to the fun of fighting a siege battles, what means: 1. stronger units, as only such provide challenge for the player; 2. no cavalry units as in reality these troops would take part in the defence only dismounted.

The bottom line:
- garrisons spawn only when the player attacks the AI;
- minimal sized garrisons for all settlements preventing from taking them just by chance;
- slightly more units for the larger settlements, the most for fortresses/citadels.
- strong units (eg. feudal infantry or specialized ones like Metsikhowne), but no professional units, and absolutely no cavalry
JoC

In the past I've made a few arguments in favour of the garrisons:

Spoiler Alert, click show to read:

The question is how do we imagine "reality" and "historicity". For instance, in the SSHIP scale any province should have dozens of towns or castles, not only one. One settlement per province is an utmost simplification. If we would have to make it historical, we would have to split it among number of sieges and the question about "who is left in the city” would not be pertinent.
Another issue is simultaneity of the war process: it's not like an army is coming an sieging and the defender is sitting back doing nothing. There's usually a process of mobilization among elites which fear losing their status. The men were called from the towns and villages. Or there's a problem of time - our turns are half-year, and we know how the movement issues are simplified. So in reality there was time before any siege to gather forces.

All in all: we have to suspend our disbelief. We are therefore free to make simplifications and craft it in such a style we feel historical and which is playable (satisfying gameplay, AI can handle it, it’s possible to code).

My feeling are:
- there were always duke’s forces constantly residing in the cities/castles just to prevent rebellion or maybe taking power by competitors – so at least one unit of feudal troops be in any garrison (including foot knights),
- there were citizens’ organizations (eg guilds in Western Europe) preparing for defense, carrying for walls and towers, organizing competitions in archery and fencing – so the most useful missile units for defense should be present: non-professional archers (but not peasant), later in the time also crossbows (I still have difficulty to agree it’s historical to use crossbow extensively in 12^th century…). I don’t think it should be the case with javelins even in Spain, similarly to Lifth opinion.
- spear was the cheapest weapon so militia spearmen should be the base, but swords as well if it’s in the regions’ tradition.
- units with a word in their name “Urban” should appear. In this case I don’t see a problem with “Merchant cavalry” in the cities with the guilds, but in general I agree: no cavalry, no artillery.
- peasants should be out from the game as a unit.
- from the gameplay perspective: I don’t see the need to disband the units. The players should be careful with attacking enemy settlements for risk of mobilization enemy forces. Taking each settlement should be an enterprise. Expansion should be slow.

It's also useful to read this thread: http://www.twcenter.net/forums/showt...script-in-EBII

**tmodelsk** · August 18, 2017, 06:40 AM

Jurand, thanks for consistent considerations.
I'll think and possibly prepare 'configurable' version where you can turn off garrisons for AI vs AI and AI vs Player scenarios.
So only Player vs AI garrisons will be raised.

**tmodelsk** · October 28, 2017, 08:40 PM

Originally Posted by john_peter

What if I want garrison script but not that much? Reading the description it seems every time I attack a settlement it will raise a large force, I don't want that, I just want the AI to have a extra small-medium force when I attack them.

john_peter - I've released 1.03 version of SSHIP-Tweaks with new Garrison Script v. 1.3 among others changes (-> download instructions & link).

"Garrison Script: raise units on siege" has now new parameter: Garrison Size :
Enter value:
* MEDIUM - to get medium garrisons (like in previous version, to large for you)
* SMALL - to have smaller garrisons. (default value).

Small garrisons will be raised in all settlements EXCEPT:
* CAPITALS - here will be always MEDIUM garrison
* very small settlements, smaller garrison for them means less units than one unit = zero units, so no additional garrison

Garrison Script with SMALL option will generate much more smaller campaign_script.txt file, so end turn times should be shorter then in MEDIUM version.

Small garrisons means smaller then Medium with one or two units, something like that. Check yourself :-)