Sieges: when to starve out; when to assault?

[econ21]

Just noticed a couple of possibilities during sieges that I had not exploited up to now:

1) [Rather obvious really] when starving out a garrison, you can withdraw the occupying army and replace it with a smaller force of more second rate troops. That way your front line army is freed up to attack somewhere else and does not suffer any casualties from attrition [siege engines in particular seem to suffer horribly from attriton].

2) [This may be cheesy] When sieging provinces with ports, it is still possible to use the port while the enemy is holed up in their castle. This can be important as ports are usually destroyed at the end of a successful siege, which can strand your armies (esp. if the province is an island).

More substantively - anyone got any advice on whether to starve out a garrison or auto-resolve an assault? Usually the castle is full to bursting and so is predicted to fall within a year, in which case I wait. But I've noticed I still suffer quite large losses besieging. Anyone got any rules of thumb about when to starve and when to assault?

[GoldenKnightX2]

For me, the chose of sttarving the garrison or assualting it depends on the number of enemies in neighboring provinces. http://www.totalwar.org/forum/non-cgi/emoticons/biggrin.gif

[HopAlongBunny]

I will usually wait it out if the castle will fall within 2yrs. Any longer than that, I doubt your casualties will be lower beseiging than assaulting.

I really dislike the fact that assaults no longer "preserve" the structure.

btw, I use your 2nd observation quite frequently. Once the area is secure, why leave a perfectly good army there to die by seige attrition or be cut-off when the port disappears? http://www.totalwar.org/forum/non-cgi/emoticons/smile.gif

[Maelstrom]

I mass produce Highlanders to handle all my sieges/castle assults - Cheap, effective and disposable

[The_Emperor]

Quote[/b] (Maelstrom @ Aug. 22 2003,10:29)]I mass produce Highlanders to handle all my sieges/castle assults - Cheap, effective and disposable
I simply summon an army of Mercs to do the dirty work for me... Very disposable and you can send them on their merry way anytime

[Al Shama'ar]

One thing I do when I get the "... will not fall without a direct assault" message, is to withdraw from that province and come back again the following year and try to kill/capture the remaining troops (the ones that were inside the castle).
In the 2nd year I rearrange my troops carefully so that the enemy chooses to fight and doesn't retreat immediatly (again). You know what I'm talking about, no large armies and such.

Thus I can usually take control of the province in only two years, and I have never seen a besiege army being reinforced after I withdraw that first year.

Al

[Kristaps]

Quote[/b] (Al Shama'ar @ Aug. 22 2003,10:47)]One thing I do when I get the "... will not fall without a direct assault" message, is to withdraw from that province and come back again the following year and try to kill/capture the remaining troops (the ones that were inside the castle).
This works as long as another faction does not jump in on the opportunity to chew up the depleted garnison... It could well be your ally... It has happened to me a couple times.

[Kristaps]

Quote[/b] (Simon Appleton @ Aug. 21 2003,20:38)]More substantively - anyone got any advice on whether to starve out a garrison or auto-resolve an assault? Usually the castle is full to bursting and so is predicted to fall within a year, in which case I wait. But I've noticed I still suffer quite large losses besieging. Anyone got any rules of thumb about when to starve and when to assault?
One thing I've noticed: with the VI expansion, the AI does not get 'will fall immediately' castles anymore. Even when the castle is full to its maximum capacity. Nevertheless, whenever a player has a substantial army in the province, it always says "the castle will fall immediately" in the case the player retreated to the fortification...

[Oaty]

One thing about putting second rate armies on guard to besiege a castle. The A.I. knows very well what valour means and is very quick to recognize a zero star general trying to keep them subdued. The Byzantines taught me very well what this means. Belive me 300 valour 5,6 and 7 troops can smash the oblivion out of 1200 valour 0 troops. My optimistic view after this battle was, At least there are 50 less of these elite warriors to deal with. And in reality I do'nt think I even caused that many casualties but that was a while ago. Just make sure they do'nt have good troops in there when you do this

As far as sieging the castle though I almost always siege a wooden fort unless there are elite forces in there that would actually cost me more than starving them out.

Another thing to consider is how rich is that province you could end up getting an extra 1000 florins by ending the siege earlier.

I do all sieges on case by case scenario and sometimes if it is the king that is besieged and I have about 2 territories left to finish that faction off I let him back out.

Now I bet a lot of you are wondering why I do this. Heres the keypoint provinces that are owned by rebels are a lot harder to keep there loyalty after conquering. So I destroy whats left of there faction then I besiege him again, and this time for good.

[Red Harvest]

Money is an issue.

1. I would never use mercs for sieges unless I intended to fight out the siege and needed them to fight it. Mercs are more expensive for upkeep, and the expense is multiplied for the besieger. Mercs can't be retrained or combined, so partial unit losses in a siege really hurt.
2. I often do an assault to raise command stars with a low rank general.
3. I often do an assault to reduce the expense of maintaining a large besieging army in the field (upkeep is like 1.5x or 2x for besieger.)
4. I will do an assault to get rid of older low valour units I would otherwise have little use for--this lets them go out in a useful capacity rather than just disbanding to reduce upkeep. They might even gain valour and be worth recombining and saving.
5. I will do an assault to get income the next turn, and to start upgrades sooner or start building specific units from that province sooner.
6. I will do it to prevent a rebellion since loyalty shoots up when you attack.

[Australianus]

The loyalty of the province can be crucial, particularly in VI. If you assault you have a chance of the booty before you end up with 3000 loyalists ruining your invasion.

[Jacque Schtrapp]

Always assault. I can roll over any army the AI can field. However, when it comes to time assault the castle I am on much more even footing with the AI. Damn catapult towers http://www.totalwar.org/forum/non-cgi/emoticons/wink.gif

[ToranagaSama]

Quote[/b] (Simon Appleton @ Aug. 21 2003,21:38)]

Quote[/b] ]1) [Rather obvious really] when starving out a garrison, you can withdraw the occupying army and replace it with a smaller force of more second rate troops. That way your front line army is freed up to attack somewhere else and does not suffer any casualties from attrition [siege engines in particular seem to suffer horribly from attriton].


There's an additional "strategic" decision to be made in the above scenario. Whether to move your attacking/beseiging General as well. I've been meaning to observe this more closely, but I believe if you move out the "attacking/beseiging" General, he will not get the Valour benefit from the Victory.

Of course you can keep the general and move in cheaper supporting troops.


Quote[/b] ]2) [This may be cheesy] When sieging provinces with ports, it is still possible to use the port while the enemy is holed up in their castle. This can be important as ports are usually destroyed at the end of a successful siege, which can strand your armies (esp. if the province is an island).

Don't see why this might be Cheesy. The Port hasn't been destroyed as yet and exists outside the Castle so it would sensically be in control of the beseiger, No?


Quote[/b] ]More substantively - anyone got any advice on whether to starve out a garrison or auto-resolve an assault? Usually the castle is full to bursting and so is predicted to fall within a year, in which case I wait. But I've noticed I still suffer quite large losses besieging. Anyone got any rules of thumb about when to starve and when to assault?

I'd think the basic rule of thumb, would be wether you "need" any of the buildings that might be destroyed when the Castle falls.

If you intend for that province to be a "producing" province, then in regarding to "Time" even more than "Money, but both are important, it may be wise to perserve the Castle by taking it with an Assault.

If its not going to be a "producing" province, then there's less, if any, benefit in assaulting the Castle.

Of course the decision(s) is extremely subjective to the circumstance, e.g., how many quality troops you have, Time, Money and Threats from other factions, potential and real.

In Shogun, I virtually always let the Castle fall. In vanilla MTW, more often than not I'd let the Castle fall, but utilizing the MedMod and my Hardcore Rules (very little money&#33http://www.totalwar.org/forum/non-cgi/emoticons/wink.gif the need to preserve and take a Castle is much greater.

Of course, having an overabundence of cheap (doing nothing much) peasant fodder (that you're probably going to retire anyway) is a big incentive to assault. http://www.totalwar.org/forum/non-cgi/emoticons/biggrin.gif

[Hamburglar]

I assault about 95% of the time.

Why?

1. I can often do it without taking heavy losses. (I build LOTS of siege engines)
2. My general gains experience.
3. Men still die when you starve it out, but when you assault at least your men gain valor.
4. Besieging armies pay double support cost. Often it's quite a bit of money. Usually cheaper to just assault it and replace the dead men than have the castle seep away your money.
5. Quicker you own the castle the quicker you can start using it.
6. When facing attrition losses, you don't get to choose who dies. Oftentimes you lose a lot of "good" men starving out a castle like Knights and Siege Engineers. If I storm it I get to pick who takes losses - notably the rank and file rather than the elites.

[Sainika]

I prefer to siege castles instead of assaulting because after the assaults the level of castle reduces.
Of course if the number of besieged men is very low and the time to wait is too long I assault. In that case I use a lot of siege weapons and big army. Usually I set one unit of low cost to accumulate arrows. Main attacking force is heavy-coated cavalry which can enter the castle without heavy losses.

[motorhead]

In my latest campaign as the Turks (late, expert), I had 4 Egyptian provinces under siege at the same time. I was attacking with light desert troops and no siege equipment so castle assaults didn't appeal to me as the sieges involved one keep+, 2 castles+, and one citadel+. I kept moving my fighting armies ahead and moved in siege armies (lesser troops and lesser generals, enough to keep the province from rebelling and keep garrisons from sallying out).

All the garrisons I was besieging were small, 100 troops or less so long siege times (avg 7 years) and they suffered little to no casualties each year. After grabbing the last two provinces, the Horde struck my northern border so any thought of trying to castle assault went out the window as I needed my troops and generals up north pronto. So, while waiting out the sieges I kept dropping my spies on them to see what their chances were of opening the gates. Usually first turn your chance is zero. Smaller castles began to show small chance (10-20%) after another turn or two. By the 3rd turn, keeps were in the 60-70% range and castles were creeping over 20-30%, and by the 4th/5th turn castles and even citadels hit 50-70%.

I had a surplus of spies to keep loyalty up following my invasions so I dropped 2 or 3 on a castle once it broke the 50% range and lost only 1 spy for 4 castles. Seems that as a siege drags on the chances of spies doing their work increases, even if the garrison has only lost 3 or 4 men over 5 years. Also, I was really strapped for cash during this time so even the measly 2k florins to bribe the garrisons was out of my reach. And I can't recall any infrastructure being damaged/destroyed when a spy opened the gates, so overall it wasn't too bad given the contraints I had.

[Eastside Character]

I always assault if I have enough forces to do that ,and for practical reasons it is automaticly resolved (I found that automatic resolve of castle sieges gives very good results, for me it seemes AI generates result without talking into consideration that besidged army is protected by towers and all that other "castle defences".

[Red Harvest]

Tip for many of you who are apparently confused on this matter:

The level of deterioration of castle and buildings is RANDOM whether you assault or let it fall to siege. It is random. This came from a post by one of the CA team some months ago. This applies to VI version, not the original MTW.