Few questions about sieges

[Vercingetorix_Defeated]

Sorry to badger you guys with more questions, but I thought I'd make a thread to ask a few remaining questions about sieges.

As you might have seen in the other thread, I used a method that Wlesmana discovered to remove all wall defenses, including boiling oil, from sieges. This makes them a lot more fun to play, in my opinion.

However, there is still a problem. Namely that the enemy campaign AI seems reluctant in the extreme to actually attack during a siege....they prefer to just sit there for the 15 years or whatever it is that is required to starve my city into submission.

Does anyone know if there is some kind of way to force or really encourage the AI to attack an enemy city in siege as soon as they have siege stuff built?

Historically, few factions were patient enough to just camp around outside the enemy walls for over a decade when they could just assault it instead .

My other question is if there is a way to modify how many units die of starvation when inside a siege. (ie, after one turn, every unit inside the city loses 8 men, is there a way to change that?)

[Darth Alec]

They always assault in my games. And Sieges never lasted longer then a few months in reality, and a year was really pushing it.

[DimeBagHo]

In my campaigns the AI always attacks as soon as they have built some siege gear. Unless the beseiging force is all cavalry, as often happens with the horse factions. Then they wait you out.

[Scutarii]

My AI always attacks two turns after besieging. Don't know why.

[Vercingetorix_Defeated]

Hmm, it is odd.

The faction that is currently besieging and not attacking is the Seleucid Empire, somewhat surprising given that they have a full stack of around 4500 foot soldiers, decent ones too, thorakitai, phalanx pikemen, few archers, etc. The city has a full stack of units defending it, but they're all low grade: militia hoplites, skirmishers and a few archers. Nothing good that should scare the enemy from attacking.

It wouldn't be a problem if the units inside the city didn't lose a few men per turn, that's a bit annoying. Does anyone know how I change how many men a unit inside a sieged city loses per turn?

[Scutarii]

I don't think you can. At least, I've never heard of it being done... If you have a full stack, they will be reluctant to attack, but they should still attack eventually.

[CaesarVincens]

The time it takes for the AI attack during a siege is based on how quickly they build "enough" siege equipment. So if they have a full stack of infantry, they'll want to build at least half a dozen pieces, but that takes them three or four turns.

[Vercingetorix_Defeated]

Ok yeah..after waiting a few turns, I've confirmed what I thought earlier. The AI refuses to attack settlements of mine. Even forts. A full stack of Seleucid infantry with 4 rams refuses to attack my fort, instead starving it out. Their armies wait outside my cities for the 8 or 9 turns necessary to starve them out.

It's almost as if the game knows I've fixed sieges and now refuses to fight them.

Does anyone know which text files are related to how many men are lost in a sieged city per turn, and how long it takes to starve them out, etc? I would appreciate it if anyone could let me know where I might change those values.

[Scutarii]

The number of turns to starve a city out is connected to the wall levels, I think, so its either more than simple text-file editing or its next to impossible. The other one is not possible, AFAIK.

[RedFox]

Try moving out half your troops and hide them in nearby (if possible) woods. The AI won't attack if you have the advantage.

[Vercingetorix_Defeated]

Try moving out half your troops and hide them in nearby (if possible) woods. The AI won't attack if you have the advantage.

Can't move them out of the settlement when it's under siege.

So far, after testing a bit by reloading saves a few times, the AI either starves my settlement out or decides to move away at the last minute and break the siege. It's almost as if they're programmed to deliberately not attack the walls themselves .

Do you know if the siege thing where you lose a few men per turn is hard coded?

[Roman Warlord]

Got to say though while I have experienced what you are saying a couple of times, this was due to me having what appeared to be a stronger force.

However this only ever happened to me on 1 out of 100 times, the rest of the time they attacked.

[CaesarVincens]

Do you know if the siege thing where you lose a few men per turn is hard coded?

Yes, for goodness sake, everyone has been saying yes. This is hardcoded.

[zowrath]

After playing MTW2 and ETW for some time now I wanted to go back to RTW a bit, but wondered if there was any fix for the AI not attacking in sieges. This thread is from March, has there been any improvement? I remember last time I played the AI only attacked if I had considerably less forces than them in my cities, otherwise they would starve it. Which is completely rational of course, but a bit boring as a player.

All the times when I am attacking myself during a siege I wonder how I would have fared if it was me that defended with a full stack against a full stack. But so far I haven't been able to test that out

[Publius Quirinus]

No one else has mentioned it, but could campaign difficulty be related to the AI's reluctance to assault settlements? When I used to play on E/E, the AI hardly ever assaults my cities, but they do it a lot more often now that I play on M/M. And I remember reading a blitz AAR that seems to imply that playing on VH makes the AI assault your cities immediately.

[Scutarii]

There's no fix for the AI. But this problem does seem to be quite imaginary on difficult levels above Medium. My AI always assaults 2 turns after sieging (one more than necessary, but they still assault), and if they have elephants, right away.

[zowrath]

There's no fix for the AI. But this problem does seem to be quite imaginary on difficult levels above Medium. My AI always assaults 2 turns after sieging (one more than necessary, but they still assault), and if they have elephants, right away.

So what you're saying is that those that don't want to play on harder difficulties will not experience much sieges? Which then I guess it is hard coded somewhere....

[Scutarii]

Not necessarily. My experiences on medium with sieges are the same as on H and VH, the AI assaults on turn 3 of the siege, sometimes longer if the garrison is big.

[CaesarVincens]

I've always played on hard or very hard, but in my experience, the AI will take two to four turns building siege equipment depending on how large their army is. Once they have finished all of their equipment, they will assault.

[GoldShield Guard]

I've always played on hard or very hard, but in my experience, the AI will take two to four turns building siege equipment depending on how large their army is. Once they have finished all of their equipment, they will assault.

There's no fix for the AI. But this problem does seem to be quite imaginary on difficult levels above Medium. My AI always assaults 2 turns after sieging (one more than necessary, but they still assault), and if they have elephants, right away.

I confirm the above two posts. On H and VH that I am always playing, they assault as soon as they have enough for them siege equipments. If they have elephants or onagers, they attack immediately.