Please register for Total War Access to use the forums. If you're an existing user, your forum details will be merged with Total War Access if you register with the same email or username. For more information please read our FAQ’s here.


Sieges-problems and possible solutions

HorseWithNoNameHorseWithNoName Registered Users Posts: 1,001
The current siege battle system has imo mainly two problems:
1. The biggest problem is that battles play out the same very often, same garrison, same procedure
2. Another problem is that, although this has become more difficult, it is still incentivised to game the system in immersion-breaking ways: When you are attacking, you are incentivised to pull your units where the towers cannot reach you, so for the player, only one tower effectively exists which he can either destroy or rush to the walls under its fire. When you are defending, the fact that the AI is trying to use the battering ram is highly abusable. (There are more exploits, but these are imo the ones that you cannot avoid noticing.)

I propose the following changes:
-While sieging a settlement on the campaign map, you can "zoom in" on it and choose a part of the settlement's walls to attack. This will determine what form the wall will have in the actual assault. During the siege time on the campaign map, you can also choose wall segments and towers as targets for your arty (if you have any in your army). The chosen wall segments/towers will sustain damage per turn (maybe 50%) and will be damaged/destroyed in the eventual assault.
-What kinds of units are in a garrison is affected by what kinds of military buildings you have in a settlement. These will not be additional units but instead weaker units of preferably the same type will be substituted for them.
-Towers will have a very slow turnrate for their cone-shaped targeting area so that up to half of the towers can, given enough time, target an area which on one edge of the map and has a distance to the walls equal to the range of long range arty.



  • daelin4daelin4 Senior Member Registered Users Posts: 16,419
    Very interesting thoughts, here are some of my comments:

    1) it does seem interesting because it allows you to effectively skip the initial part of siege battles, but unfortunately also still makes for a very straightforward procedure. If you bring enough stacks you can basically have one consist entirely of artillery, then quickly smash the settlement's defenses in a few turns, and then just auto-resolve in the next (well not really; you can just bring like four stacks and immediately auto-resolve a siege battle for easy win).

    2) I do think garrisons configurations need to be considered, either now or in future titles. You can do this fine enough up to Shogun2 thanks to no generals required for armies, but since you require generals now, garrisons are considerably limited in scope due to their reliance on a few buildings' ability to grant bonus units. Dedicated defense building chains for both minor and capital settlements is a good step forward, but a better one would have been to utilize existing military buildings to effectively train garrisons. Even Rome2's Hannibal at the Gates DLC campaign basically spawned immobile garrisons for the AI when you entered Rome or Carthage region, top simulate a reinforce garrison (yes, we're talking Realm Divide spawned out of thin air armies for free BS reinforcements).
    From a technical standpoint this seems easy enough- every settlement contains a general, in the form of a governor, that is used to produce an army of whatever size and composition the player desires; unlike regular armies however, these garrison armies are entirely limited to what units the settlement can train, what mercenaries (in case of Warhammer, RoR if any) are available for hire, and what units a regular armies decides to merge with it. This allows you to shuffle and reconfigure garrisons and army compositions around, without the trouble of little captain armies running all over the map like they did from RTW through Fall for the Samurai (though frankly even then it wasn't that bad).
    This then leads to interesting diverse strategies like deciding whether to leave a skeleton garrison and maximize the power of your army to go out on campaign, keep a larger garrison in case the enemy comes back, or anywhere in between. You can then dispense with the whole buildings grant bonus units idea as it's far less flexible. Units can be disbanded or recruited at will, given the usual financial and recruitment slot constraints. Frankly, the regular configurations of garrisons in Warhammer are pretty bland- it is not like being able to configure a garrison to have 20 Thunderers, with its regular cost to recruit and maintain- is an automatic win for every siege battle, but it at least allows for interesting scenarios. Which then adds to the replayability factor.
    The only reason siege battles would be memorable in this game, is not because they are so diverse and cool, but simply because they're all the same!

    3) Towers is a little tricky. The design is based on the fact that some factions can get artillery quite early, and some factions simply have better artillery, so all towers need to be universally designed to be able to handle all threats at some level, at the start of a campaign. But I have never really found towers to be such a huge problem- you bring enough units to get through the tower shooting gallery phase, you've probably a good chance to winning the fight. If anything, units within minimal range of towers should just reduce their range to very short, so that they still function by attacking units very close, ie units climbing the walls, but not at the cavalry or artillery many more yards away that pose no immediate threat. i think that is the best alternative to just locking out unit types or having a max range that the opponent can exploit.

    Corrected action is the most sincere form of apology.
  • RifugioRifugio Member Registered Users Posts: 1,125
    edited February 2017
    A while back there was another big topic about that was talking about the siege mechanics, I think it makes sense to have the siege weaponry (and ideally mining) represented as part of the strategic siege map mechanics - the longer the settlement is under siege being bombarded by artillery and undermined by sappers the more damage will appear on the battle map. I'm not sure I see a need to "zoom in" on it and choose a part of the settlement's walls to attack, though. This could be implemented automatically so when the battle starts you would be presented with the cumulative damage your siege has done (and possibly secret tunnel routes that your sappers have dug - if not concentrating on undermining walls). I would not feel the need to micromanage the targeting.
    Rifugio said:

    I agree with @Tajl Historically actual siege artillery, and mining took days, if not weeks to show results. So would more realistically be represented as part of the long term siege mechanics of the game... Perhaps it would be good to have some evidence of siege weapons actually having caused damage to the settlements defenses depending on the length of siege prior to the assault and amount of artillery/existence of miners/sappers etc. Taking out random towers, damaging gates and causing attrition in defenders/morale.

  • HorseWithNoNameHorseWithNoName Registered Users Posts: 1,001
    2. Abolishing auto-constructed garrisons would indeed open up even more options for the player, but @Ephraim_Dalton found in tests he made that the ability for the AI to conquer opposing AI settlemets is very sensitive to the quality of the garrison, so maybe retaining auto-constructed garrisons is required for now. I am also not sure how easily abusable that would be.
    3. I am not that fond of my solution myself, but the ability of the player to invalidate all but one tower for the whole battle seems too strong to me (and even that one tower can be destroyed be arty while drawing its fire with a durable character). I am not sure getting rid of their min range would be that good as siege towers or clumped up units (ladder) may be too easily destroyed.

    @Refugio I suggested the "zoom in" (could also be realized with a schematic tactical view) because I think it can give a good context to your siege (and changing the wall's layout should not be that difficult to put into the game). But you bring up a good point and I think you are right when you say that it would probably be better to have siege damage (with some restrictions) randomly affect targets (walls/towers). Thinking about it, this would players from always choosing the best solution and would lay more emphasis on playing the actual siege battle well tactically. Good call.

Leave a Comment

BoldItalicStrikethroughOrdered listUnordered list
Align leftAlign centerAlign rightToggle HTML viewToggle full pageToggle lights
Drop image/file