Repercusions of the AI retreating to a castle?!

Printable View

11-27-2002, 13:07
+DOC+

This is a tactic employed by the AI that i find very strange?

For example, say a province has 600 men and is attacked by 1000 of my men and the fortification is castle-sized, which means it is capable of holding 600 men. The Ai in this circumstance will more often then not retreat to the castle.

Firstly this allows me to pillage the province uncontested.

Secondly, the length of time they will be able to hold out in the castle will only be 1 year from the following turn due to the sheer number retreating To me, castle siege times are still too short when the castle is near capacity.

If there is no relieving force, then this is nothing short of AI capitulation as it'll lose 600 men in 1 year for almost no cost to me. IF there is a relieving force, whether they will be able to do any better than the initial 600 v 1000 is debatable since they are attacking me and i now have the advantage of being the defender. More often than not though there is no relieving force.

Now....

With that number of defending troops, imagine they decided had defended their province whether they knew they were going to lose of not.

Firstly, if they had lost they'd have certainly have killed many of my troops in the first place.

Secondly, they'd now have a lot less than 600 men in a 600-man castle, therefore resulting in the approximate siege time being considerably greater.

What all this essentially means in gameplay terms is that i have now lost many men, face a longer more costly siege and as a result of both of these will be subjected to a greater likelihood of loyalist revolts in the besieged province.

That's my thoughts.
11-27-2002, 14:26
Puzz3D

True DOC. It appears that the AI is only evaluating the current turn, and not looking even one move ahead. It's as though the AI doesn't know there won't be any relieving force coming. On the other hand, realistically an opponent would retreat to a castle if they were facing defeat on the battlefield because they would survive a little longer in the castle. They would then also have some hope that a relief force could get to them before they had to surrender.
11-27-2002, 15:09
The_Dude

I agree with you to, but retreat to the castle should allow a longer siege as it's planed, maybe twice the normal length. If it's after a battle, we can supose that all the food and water supplies are not in the castle, so normal length.
BTW longer siege would be a great improvement, as for AI assaulting castle. Why attack a 600 men castle if you just have to wait one turn? Most of the time I don't assault castle, you can lose a lot of men...
11-27-2002, 16:09
BatkoMahno

It would be more realistic and fun if siege time was much longer and religion conversion in provinces was prolonged to 200 years. It shouldn't be possible to convert all muslims(or catholics) to catholicism(or muslim religion) in almost less than 50 years. It should take more than 3 generations to switch religion if not more( even if you build a lot of churches and etc.).
More fun if AI takes the battle even in bad odds specially if there is no relieving forces in nearby provinces.
11-27-2002, 22:09
MizuKokami

maybe it's just the ai's way of saveing money for better forces. http://www.totalwar.org/forum/non-cg...icons/idea.gif
11-28-2002, 07:30
YunDog

the simple solution to this would be that the province doesnt fall till the castle falls and the castle shouldnt fall unless attacked - now we can realistically lay seige to castles the whole autosiege thing is a hangon from the STW days when you couldnt - you should have to take every castle - maybe this would mean you would have to make castles less prevalent - but it solve alot of problems - but then the AI is not so good at seiging and will get slaughtered but that happens to me too http://www.totalwar.org/forum/non-cg...s/rolleyes.gif
11-28-2002, 08:16
hiver77

I'm sorry if i missed a previous discussion on this (haven't played the game since early October http://www.totalwar.org/forum/non-cg...ns/biggrin.gif ), but wasn't the patch supposed to triple the length of castle siege? It didn't seem to change it in my patched game.
11-28-2002, 11:48
deutschlanduberalles

It didn't change it in my patched game either, and my units cost the same in upkeep whether they're sieging or not. (I've checked with the economy sheets and done about a million reloads to make sure this was the case)
11-28-2002, 12:40
+DOC+

Siege times are still too short, whatever changes were made weren't great enough. http://www.totalwar.org/forum/non-cg...cons/frown.gif

Regarding siege support costs, this depends on the difficulty level you're playing at. I think it goes like this:

Normal = normal
Hard = x1.5
Expert = x2

I'm sure someone will correct me if i'm wrong. I think the AI follows either the normal or hard values no matter what difficulty level the human plays at, although which one i don't know?
11-28-2002, 13:37
Fearless

If everyone agrees sieges are not long and costly enough were is the mod to correct this.......or is this not possible
11-28-2002, 16:33
TheLastEuropean

Maybe the AI intends to send a relief force but becomes caught up on other fronts. It's difficult to know what the total state of AI factions are. I know there are times I retreat into a castle from a superior force with the intention of relieving the siege a turn or 3 away but end up not doing so because I need troops elsewhere etc. In my games, more often than not the AI WILL relieve it's besieged troops (when it can&#33http://www.totalwar.org/forum/non-cg...icons/wink.gif
11-28-2002, 16:54
einar

Quote:

Originally Posted by [b

Quote[/b] (BatkoMahno @ Nov. 27 2002,09:09)]It would be more realistic and fun if siege time was much longer and religion conversion in provinces was prolonged to 200 years. It shouldn't be possible to convert all muslims(or catholics) to catholicism(or muslim religion) in almost less than 50 years. It should take more than 3 generations to switch religion if not more( even if you build a lot of churches and etc.).
More fun if AI takes the battle even in bad odds specially if there is no relieving forces in nearby provinces.

Yes, i think this is very UNREALISTIC. The conversion of a determined province should take more than 50 years. 200 years is more realistic, and the province should conserve a residual minimun percentage of the original religion that the province had (10%, etc). This minimun should be very difficult to convert.

The AI retreating castles is unrealistic, too. The garrisons should have a maximun of 200 men and several years until surrender.
11-29-2002, 15:03
Lord Romulous

couldn't agree more gentlemen.

please devs, if u make a add on fix the castles role in the game. make castle battles more involved too.

and add more diplomatic option, and more win game critera.

just conquering is getting boring now.