Are siege battles the toughest? (Lusteds battle AI)

[Dave Scarface]

Hello guys,

As the title asks, are siege battles the toughest battles to win if you are up against a stronger AI stack? My mod uses Lusteds AI as most of you already know. And it sometimes gives me a serious run for my money if I don't have a good enough garrison. I started a new campaign today as the Turks, and managed to capture Edessa at the start. The trouble is you can't recruit any units in Edessa when you first take it over. So you rely on mercs to help defend it until reinforcements can arrive. Anyway I had another army on the way when Antioch suddenly launched an attack on me with quite a good early stack.



The odds looked to be against me and I was starting to get a little scared of losing I must admit.



In the end I put up a heroic defense but my general was very severly injured in the battle afterwards. And this battle could have been easily lost. Now I am quite confident about winning 95% of field battles even if I am outnumbered. But siege battles? I must admit they do test me.

Look how many men I had left? most of the defenders died and there is virtually no garrison left to defend it again, it was horrific I tell you! I even had to throw all my horse archers into the melee fight to try and scrape a win.



Dave

[edders]

I think it depends very much on morale and unit placement, to be honest. I find holding even basic settlements relatively easy if I can funnel them through the gate, especially with the burning oil feature.

Oddly, it's cavalry who tend to be the biggest problem during sieges! Even with their charges being screwed up by going through the gate heavy cav are still able to absolutely slaughter the spearmen defending it, even in schiltrom mode, thus causing them to flee far quicker than I'd hope. Part of the problem is the nasty deployment bug, which often means I have to start units quite a way back from the gate, then start the battle and rush them into position - this can screw things up a bit on higher diffculties due to how easy it is to get fatigued.

In these scenarios I find the town center to be the most defensible point - I keep a cavalry unit to "man" the wall defenses, then run away as they break through. Both AI and unit cohesion tend to get messed up as the enemy moves from out of the streets and into the wider central square, with units charging in piecemeal and cavalry in particular flying past the frontal defensives to mill around uselessly at the flanks. I'll usually set up a defense like so:

S S S S
I I I
C C C C
A A A

EDIT: Crap, the diagram is screwed up when I actually try to post it... anyway there are massive gaps between each unit.

S=Spearmen in schiltrom
I=Other infantry
C-Cavalry
A=Archers

I deliberately leave gaps as the cavrly will often charge through at full effectiveness, miss, slow down and trot into the waiting units behind/at the flanks.
It's a bit harder to use this tactic in the mod compared to vanilla due to the lack of flaming arrows (over a long period of time these would utterly destroy morale), yet it can still be even more effective than a standard wall defense. Back in vanilla I was able to hold of successive waves of mongols in a completely destoryed, plague-ridden city with this tactic, and the formation tricks are just as effective in your mod.

[Harry Lime]

:hmmm: Interesting tactic. I've never even considered making the town square the main focus of defense. I always try to defend the walls with infantry, have archers on the walls but set back away from the fighting, horse archers back in the streets with skirmish off & spearmen defending the gate in a V or double V formation depending on how many I've got.

I think ultimately the success of defending a siege lies with the capability (command rating) of the General leading the army but like most things I could be wrong. I'll have to check out Lusted's AI & see if it makes a difference.

[roy34543]

I always thought that assaulting a enemy town is the hardest part of the game.
And damn boring too.

[Dave Scarface]

For me it depends on the garrison that I have to defend the settlement. If its lightly defended and a large stack attacks it I dont even play the battle sometimes (I will autoresolve). Because when the AI breaks through your gates with much better men and greater numbers you can do nothink, and they will surge to victory.

Edders, your tactic looks good on paper assuming you have that many units in your settlement ready to defend it. But I personally don't tend to leave large garrisons defending my walls. Its very expensive and can be a waste of manpower. However it is risky as well playing against Lusteds AI that attacks with full stacks. I never play with fow off unless testing, so the AI does catch me out on occassion. So this is why for me at least siege battles are very challenging.

Dave

[Dave Scarface]

I always thought that assaulting a enemy town is the hardest part of the game.
And damn boring too.

Yeah, especially with the garrison script, some battles can take me over 1 hour, and I don't always win at the end of it If it starts off bad and the first men in your attack rout it gets harder and harder. Attacking great cities with ballista towers is especially hard. Sometimes 2 stacks are required to be sure of victory.

[edders]

Edders, your tactic looks good on paper assuming you have that many units in your settlement ready to defend it. But I personally don't tend to leave large garrisons defending my walls. Its very expensive and can be a waste of manpower. However it is risky as well playing against Lusteds AI that attacks with full stacks. I never play with fow off unless testing, so the AI does catch me out on occassion. So this is why for me at least siege battles are very challenging.

Dave

Well, not just on paper since it works every time I utilize it. Even with smaller forces I've found it fairly effective due to the non-routing feature of the TS and the enemy AI breaking up and tiring out its forces when it heads for the center.

However, you're correct to the extent that it only works optimally with the right mix of units, and is less successful in the early stages where you don't have schiltrom units, and cavalry still pwn spearmen stat-wise.

As to the fow issue, I also used to also play that way but was also terrified of being caught off guard, hence I'd spend ages 'taming' every single province (i.e. watchtowers EVERYWHERE, spies hovering over the borders etc.). It was extremely boring, however, and I always thought the ability of full-stack enemy forces to march all the way through open countryside for miles, and take a city by suprise just because their movement points were just over the view limit of a settlement, was just plain silly. Therefore I do always play with fow-off these days.

And I agree that siege assaults are indeed the most costly but also the most god-awful boring part of the game. I pretty much always auto-resolve with these unless it gives me a patently stupid result.

[vietanh797]

most likely that I always have 6 group spearmen, 2 group Heavy infantry(HI) the last is about 4 group archer this could help me win even i face a full stack in beginning of this mod(about 50 turn first) all spear defend the main gate
archer in the wall and the HI stay behind wall then go to defend where AI use ladder or tower
if I can may have one or two group light cavalry to catch all run away enemy

but if I siege I will use 6 HI 4 spear and 2 archer with 2 heavy cavalry this is better to att a town

[Edward the Bear]

I used to fight the siege battles, but it took too long, so now I just tend to wait it out (if I'm the attacker) or auto-resolve it (as the defender.)

[Dave Scarface]

Waiting out a siege is not always an option unless you have surrounding regions under control. Playing the Turks for instance a prolonged siege against Antioch at the start would be impossible with the garrison script their. Because reinforcements would flood from Aleppo and Krak de Chevaliers (both castles).

What I did in my old Kingdom of Jerusalem campaign was take out Egypt's other surrounding regions first before an attack on Cairo. Btw it took me over 70 turns to get into this position! and it was not easy. Here is a screenie I still have



Dave

[edders]

Heh, I did the same as you in my campaign, though I took at least one of the more southernward settlements as well before braving the capital.

I really like the garrison script - I removed it at first but found the AI would be wiped out far too easily by each other. With it small factions are able to hold out much longer, and crusades are gruelling affairs (if more than two or three factions joined in vanilla it was basically a game over for the target settlement.). I remember in one campaign I had a close ties with the Pope as Milan and wanted to take out France, yet despite the fact pretty much all of Europe was at war with them (Spain, Portugal, England, Milan, Venice, Denmark, Sicily) it took multiple crusades just to kick them out of Toulouse, Paris and Rheims. At this point they were reconciled with around three settlements left and remained alive (if weakened) for the rest of my campaign despite intermitently battling HRE, Spain and England. This was with all the money scripts disabled, as well.

[Edward the Bear]

Waiting out a siege is not always an option unless you have surrounding regions under control. Playing the Turks for instance a prolonged siege against Antioch at the start would be impossible with the garrison script their. Because reinforcements would flood from Aleppo and Krak de Chevaliers (both castles).

What I did in my old Kingdom of Jerusalem campaign was take out Egypt's other surrounding regions first before an attack on Cairo. Btw it took me over 70 turns to get into this position! and it was not easy. Here is a screenie I still have

Dave

I don't have a problem with field battles, I always keep several stacks of 5-7 Turkomans/Mamluk Archers around to ensure that if the garrison sallies I can wipe them out.