Fortifications and Sieges

[LDiCesare]

Morale plays a part in all fights. There is an option for surrendering during a siege, but I don't know how to handle it so there are fights anyway in the current code.
Unfortunately, the scenario won't have food/health stuff plugged in yet., because I want the square to be able to say how much food it has etc., and as long as it is a whole square opposed to a city, there is plenty of food. So the code will wait for further developments on the map (square in square) and econ (food for city) side for a moment.
But the scenario should still be fun (even though it is very simple, about 50 turns).

[Martin the Dane]

Well I dont know if this should go in the D7.2 thread as well, but it's a siege/wall isue.

While playing Delenda as the Romans I noticed that it's not allways good for a defending army to be behind walls.

The situation arose twise, in two successive games. I had an army roughly with ~150% the force of the Cartahanian TF marching upon Rome. When they besieged Rome both sides only lost a few men.
The first time, the Cartehanians moved on, so I got to fight them in the open, the second time, they stayed and gathered troops eventually wearing down my economy by taking the surrounding countryside

To make this work the besieged should be able to opt for a fight. This would be equal to meeting the oncomming forces in front of the walls.

Another optin could be to make sorties.

As it is now walls are only good for one thing, holding an undermanned possition until help arrives, and then abandon the walled in square, to let the attacker become the besieged, and then destroy the fortification.

[LDiCesare]

The defenders do make sorties, but the ratio of strengths to decide to attack may be 2:1. That is because it would be quite easy otherwise to attack a walled city with 0.8:1 forces once, twice, three times, etc. to damage them, and then finish them with a big army. If that seems unbalanced, I can change the ratio from 2:1 to 1.1:1 or something in between.

[Martin the Dane]

Would it be possible to make a besiege order, let the attacker enmass troopes "around" the walled in "city" and pund away at the walls with siege weaponds, and only use nonsiege weaponds to defend the sieg weaponds. It would then be up to the defender to make sorties into the enemy camp, in order to destroy the siege weaponds.

[LDiCesare]

Martin, I agree. We would like to limit the engagement of attackers to when the siege weapons are ready to deal most damage. I didn't go that far in the code because the main weapon in sieges is hunger and we don't have the data needed yet to model that part.

That way, attackers wouldn't attack before the walls are breached, and defenders would make a sortie when they start starving or, indeed, consider that there are too many siege weapons which must be destroyed. I think starving is the most important factor, though, and would rather have the starve/resources code first.

(edit)
What I mean is the besiege order should actually be the default behaviour and the code (or call it AI) should manage that automatically based on odds and other available data in order to reduce micromanagement.
(/edit)

[Martin the Dane]

Originally posted by LDiCesare
The defenders do make sorties, but the ratio of strengths to decide to attack may be 2:1. That is because it would be quite easy otherwise to attack a walled city with 0.8:1 forces once, twice, three times, etc. to damage them, and then finish them with a big army. If that seems unbalanced, I can change the ratio from 2:1 to 1.1:1 or something in between.

I have thought a little about this, and there are couple of things.

First. I think the ratio is to high, is it possible to edit that in an .xml file? Playtesting will have to show the best ratio.

Secondly. When the besieged party have a clear view of the approach to the besieged city, (like Rome in Delenda when Hanibal walks down the cost) the player (wether computer or human) should be able to chose battle over being besieged. Well they should be able to do that at any point, but that's the most obviouse case.

Finally I found out that the best strategy for defending a fortified possition is to split your army and leave a minor force inside the walls, while you main army waites outside, and fall upon the attacker as they settle in for a long siege. Im not sure how this relates to real life situations, but it might not be such a bad idea.

[LDiCesare]

The xml file doesn't allow to tweak that ratio. I'll have to allow it.
I want to allow players to give orders to the units, like in civ where you say sentry/fortify. These orders would be attitude-setting, so you could have fortify vs. berserk, and setting berserk would mean you will fight outside the walls. Would that suit you? I think it could work (when I code it ...)

About Carthago and sieges. Augmenting the defense value of the wall actually augments the number of damage dealt by the defenders because they will suffer much less damage while dealing as much.

About strategy, the AI currently doesn't appraise enemy forces very well, but indeed your "bait" sounds like a good idea.

[Richard Bruns]

I just played a game of delenda, and something appeared to go seriously wrong when I attacked Carthage. I had a colossal army, with dozens of legions and about 14 siege weapons. They had two phalanxes. My power circle was enormous, theirs was the minimum size. I fought for several turns, each turn saying that I had won the siege. But suddenly, after another message saying that I won the siege, five siege weapons retreated to another square. The Carthaginians seemed to be entirely unhurt, and they sent a legion to wipe out all of these siege weapons the next turn. My entire army had vanished, to no effect. I cheched the battle log, and it showed that I lost 40 elements before they suffered a single casualty. The few casualties they had were all suffered after the wall was breached, but this happened after my force was almost comeletely wiped out.

What is going on? Why does the GUI tell me that I am winning if I am getting slaughtered this badly? And what does it take to destroy Carthage? The SiegeWars scenario worked perfectly, so what is going on here?

The battle started like this:

Fortifications built. Bonus of 1.4952434
Attackers assault enemy fortifications.
Fight at range Long Range with 41 vs 3 elements.

It is almost as if the siege weapons were not even being used, for some reason.

[LDiCesare]

I will check the sieges in Delenda.
The model is currently awkward because tha besiegers don't wait for the walls to be breached to attack. That means they suffer huge casualties. In order for them to wait, I have to make it worthwhile, which means I need food shortage for the besieged. Thus the model is quite unrealistic now.
The retreat is due to a "defeat", but defeats happen once in a while during long battles. I can probably prevent retreat from a siege.
Another thing I can do is totally destroy the walls when they are breached. That way, the turn walls are breached, attackers no longer have to suffer the huge defense bonus from defenders, and the following turn they could flank the defenders. Currently, they don't flank, which means they still take a lot of casualties even after wall breaching succeeded.

[Martin the Dane]

I have played Delenda a couple of times and never had any problems at all wiping out the defenders in Carthage, but I never tried bringing any siege veapons.

To me it seems like there is some problem in the code as to figuring out when to fight, and when to just let the oponent waste himself on the walls. Hmm could this be related to the trouble I had defending Rome with a huge army?

[LDiCesare]

I think the problems may come from our various releases of D7, D7.1, D7.2, D7.2.1... which don't sport the same behaviour one another. I quadrupled the walls strength in the last version for instance, and there are ai changes too and other things which may change the behaviour. See the new thread I just posted.

[Richard Bruns]

Destroying breached walls completely would be very good. Overwhelming force should have some advantages, and the defender should have to rebuilid the wall again after suffering that much damage.

What I want to know is why my attackers didn´t suffer this kind of attrition in the SiegeWars scenario, which theoretically has better walls. I had much less of an advantage in the siege of Mordor than the siege of Carthage, but suffered fewer casualties. Should I play again and post both logs?

Small sieges already seem to last forever without fights in the game. Seville was besieged for dozens of turns without much happening. The log even said "Siege of Seville proceeds with little or no fighting." I certainly recommend that attacks not take place until the walls are breached. Also I would like to prevent the raising of troops in a besieged area. Simply move the province capitol somewhere else if it is besieged, the same way it is moved if it is taken oer.. IMO that is the whole purpose of the siege model. Sieges are supposed to last a long time without fighting, until the defender sends an army to attack the besiegers.

By the way, what happens in the military model if a TF is sent to attack besiegers? Do they join the troops inside the city instantly, or do they fight outside the city?

[LDiCesare]

By the way, what happens in the military model if a TF is sent to attack besiegers? Do they join the troops inside the city instantly, or do they fight outside the city?

I am afraid they dig a tunnel, enter the city, and reinforce troops there without noone noticing. When we code "squares in squares", I will have different squares for armies in and out and lots of things will go better.

What I want to know is why my attackers didn´t suffer this kind of attrition in the SiegeWars scenario, which theoretically has better walls. I had much less of an advantage in the siege of Mordor than the siege of Carthage, but suffered fewer casualties. Should I play again and post both logs?

The damage mostly comes from the fact that in Mordor there aren't any replacements coming in the city, or hardly any sizeable ones. Maybe also due to differences in units attack strengths (in Delenda, they deal way more damage).

[Richard Bruns]

The difference in attack strength is probably the cause, since I was getting chewed up even in the first battle, before reinforcements arrived. I think it would be a good idea overall to lower Delenda attack values while keeping the defense and health high. That way both the normal fights and the big sieges would be less deadly and last longer. I´ll test this with the testbed when it comes out and tell you which settings I like best.