Too many siege battles is an AI problem, not a campaign map problem

[Knight of Heaven]

You guys think there is an AI currently in this world capable of handling all that? im sorry skynet, and jarvis arent invented yet. We will make do with what we got. If less sieges helps the AI, am all for it. It realy isnt a big deal. NTW didnt had sieges for the most part. And it was fine.

[Bethencourt]

Do you see that means that main cities with walls will have some of the worst AI performance in the game? Is this true and the opposite too? I think this new lack of walls is more related to releasing a game just in time, so no need to develop more walled cities.

[Rittsy]

I do not feel worry, but that it is a waste of time needing to wait for modders or CA to solve it in the next Rome game, in ten years from now. Because where some people saw an issue, now the issue will still be there in another form and on top of that you will have the lack of walls when defending you cities. Now you have fewer armies, the AI too, but what if your main armies are engaged in the opposite side of your empire your neigbours can easly attack you with their main armies with no hope for you.

Well they're not going to change it a month before release because of a post on a forum.

I guess the player will have to rethink their strategy than, instead of playing in the exact same way they have for the past several games. Maybe they'll have to consider defensive armies, rather than purely offensive forces. If your tactics are to send out every army to the frontlines and expand quickly then there should be negatives to consider.

[Bethencourt]

Well they're not going to change it a month before release because of a post on a forum.

I guess the player will have to rethink their strategy than, instead of playing in the exact same way they have for the past several games. Maybe they'll have to consider defensive armies, rather than purely offensive forces. If your tactics are to send out every army to the frontlines and expand quickly then there should be negatives to consider.

Of course they are not. But that wont stop anyone to post his or her opinion.

Before there were not so many armies really, so you still had to take strategic balance decisions. Now going into ofensive will be, how to say? unfairly unbalanced risky?. But not necesarilly the ofensive, the defense in different fronts too.


None answers why are Rome or Carthage damned to have sieges battles when they are AI incompetent, boring and easy to win?

[Daily]

This discussion is about how/why CA cut down on Sieges. Keep it related. Stop discussing what members do and don't, it is not relevant to the discussion.

[Tim_Ward]

It is a gameplay problem. Not just an AI problem. The whole game is built around this aspect. As you stated Siege battles make the game much easier for the human player.

No, siege battles make it easier for the defending player, whether human or not. That isn't a "problem", that's the whole point of sieges.

The problem also isn't that the AI is crap at siege battles (though it is), it's that it initiates them when the chances of victory are hopeless. I seriously see the AI trying to attack castles against odds where, honestly, I would struggle to win in their place.

They had to do this in shogun 2 or else the game would be nearly un-winnable on very hard.

So, you're saying they intentionally crippled the AI so they could give it large bonuses on the harder levels? Rather than, say, gave it large bonuses because it couldn't compete with a human?

If you fight a siege battle you will win unless the AI has 2x your troops.

Try 4xs your troops, and even then it's even odds.

But, you are missing on a very important thing. Attacking the enemy.
I dont want to have auto resolve 40+ meaningless siege battles. That adds nothing to the enjoyment of the game or adds any immersion.

It's not actually compulsory to assault a castle on the very first turn of the siege.

Seriously. If there is a garrison of significant size, starve the garrison out. This has a major advantage of preserving your man power, and is probably quicker in the long run - you won't be waiting for replenishment to get your army back in fighting shape.

That's how they did it in the olden times, and there's a very good reason for that. It makes sense, it's the most effective way to play, and that's what the AI should do too. That's my basic point here... the AI should fight the player in the most effective way, and 9 times out of 10 the most effective way to take a city if there are defenders (beyond a couple of retainers) is to starve them out. The bonus is that the AI fighting the player in the most effective way is also better gameplay and eliminates the primary irritant from every Total War game since Rome 1.

It makes the AI more challenging, and it makes the game more fun, and it's incredibly simple to implement. Why is it not done this way?

I don't know about you but I would much rather have quality over quantity. I would rather have 20-30 truly epic city sieges with detailed streets and unique buildings and door to door fighting than 100 meaningless fights against a handful of samurai retainers and ashigaru

It's not actually compulsory to assault a castle on the very first turn of the siege.

Also, if the AI only has a couple of units you're going to auto-resolve the battle whether there are walls there or not, aren't you? Preventing this is the reason CA introduced the army cap, which is a whole other discussion.

The trouble with the AI starving you out is it gives you plenty of time to build a relief force and combine it with the units you have in the settlement, which is usually decisive in your favour. This simply gives you the 'Oh another siege, I'll just build a stack in the neighboring province and solve that minor problem'. If the relief force doesn't get there in time you inflict as many casualties as possible and counter-siege the ragged force next turn. I agree it's an AI problem, just not a problem with a overly-aggressive siege AI. Hopefully bigger cities and multiple capture points will mean assaulting forces have more of an advantage.

I don't see how this is 'the trouble'. I see only good things coming from this, from the AI's perspective.

1) If you've actually pulled off what you say, you've turned the battle from a siege assault which the AI will definitely lose to a field battle that it will only probably lose. Win for the AI.
2) If you pull that off and manage to assemble a truly overwhelming force, the AI will retreat and keep it's army instead of trying to attack the city and losing it. Win for the AI.
3) In many situations, you won't have the money to do that or you won't be able to do that quickly enough and the AI will actually capture the city. Remember, the AI will still be handicapped on harder difficulties and will likely outnumber you. Remember, also that in Rome 2 armies are capped. You can't just build a stack on a whim anymore. Win for the AI.
4) If you do pull that off, you've just spent a bunch of money on a defensive stack and not whatever else you would otherwise have spent it on, i.e they've disrupted your plans. Instead of, for example, building a fleet and army in Italy to invade Africa, you've made an army in Spain to defend a city. Win for the AI.

With the current behavior, the AI throws a stack away in a way which costs you almost nothing. Almost any other outcome is preferable from the AI's perspective. Even if it forces you to build an army and fight a field battle which it loses, it's at least cost you money and it will at least inflict more casualties on you in the field than it will trying to get over the walls. You at least had to pay to get rid of that stack.

Also, it's only good things for gameplay.

1) Instead of a boring siege battle, you'll get a more interesting field battle.
2) It gives you, the player, a Decision to make. Do I try and relieve the siege or do I let the AI take that city for now? Do I divert Legio IV from the front lines or do I raise a new army (*can* I raise a new army?) Making Decisions is the bread and butter of strategy games, and more of them you have to make, and the more meaningful they are, the better in general the game is.
3) It will make the game more challenging, as it means just having enough units in a city to see of a full stack trying to take the walls is no longer enough to keep it safe.

So you're blaming the AI trusting auto-resolve (which it doesn't always. I've had AI armies siege me out of settlements), rather than sieges being enourmously advantageous for the player.

Siege battles are enormously advantages to the defender. This is not a gameplay problem, it's how it's should work. Defender only = player because of this erroneous AI behavior to always attack when it could and should starve the garrison out.

You know what the problem with shoguns sieges was (other than having 3-4 per turn...). It was the same castle, that you used exactly the same tactics for, for every siege. Did anyone ever utilise multiple levels of a castle? I would just sit my army on the top level, shoot everything to death with archers, and sacrifice whatever garrison I had to make sure nothing ever got to the top. Again, and again, and yet again. The most fun campaign I've had so far is playing as the Ikk's, because you can run a few loan swords out to kill off the small groups the siege AI loved (now there's an AI flaw...).

I agree the siege maps in Shogun 2 weren't very good, but no amount of awesome maps is going to make the amount of sieges in Shogun 2 any more entertaining. Last time I played I'd just had like 3 or 4 in a row. The same armies, attacking the same castle and receiving the same slaughter. If they'd decided to starve that castle out I'd have had a real ing problem - not enough monies to make a new full stack and my other armies were in different parts of the map.

Am not sure re the point the OP is trying to make. Basically this is a thread to vindicate CA's decision to have less sieges?

It is agreeing with CAs decision to have less sieges but disagreeing with its method for achieving that. You could get rid of 9 out of 10 sieges battles - and get rid of the most boring and pointless 9/10ths - simply by changing the AI behavior to starve out garrisons in most circumstances and still have walls round every city.

I think less sieges is because CA havent had enough for Rome2 in dev and found a better alternative so choose an easy way out and then argue it's more fun the current way...

I don't think that makes any sense at all. They will already have made several generic siege maps. It is not extra work to make those maps happen in every city rather than some cities.

It's a decision they made for gameplay reasons. It's the wrong decision, but it's still one they made for honest reasons.

Hopefully this doesnt mean Rome2 is going to be awfull on this score...or the capture the flag type scenarios silly.

Sieges have always been capture the flag. Admittedly, you usually had to kill pretty much everyone before you got to the flag...

When you are getting attacked, the AI has to attack, otherwise, it would be a boring game.

You are not correct. The AI 'having' to attack is what makes it a boring game.

[baptistus]

Now you have fewer armies, the AI too, but what if your main armies are engaged in the opposite side of your empire your neigbours can easly attack you with their main armies with no hope for you.

Cool ! ^^ That sounds good to me. You have to plan carefully yours invasions. IMHO, this is exactly what CA have in minds with this two features (minor city with no walls, restriction of the number of armies ). It could be a good challenge to defend a huge empire !

[Bethencourt]

You are not correct. The AI 'having' to attack is what makes it a boring game.

I like your point of view, have you mentioned the surrendering option?

Cool ! ^^ That sounds good to me. You have to plan carefully yours invasions. IMHO, this is exactly what CA have in minds with this two features (minor city with no walls, restriction of the number of armies ). It could be a good challenge to defend a huge empire !

You already had to do that when having walls. And it lacks of historicity it had in previous and old games, and before anyones says, the historicity is not opposed to gameplay. Most of the time is the same thing.

[Tim_Ward]

I actually forgot it exists.

[baptistus]

You already had to do that when having walls.

No because you will have A LOT more time to call reinforcements. For example, one province with 4 cities with walls in Rome 1 (even if there is no province in R1) = 9 or 10 turns to take all the city with one full stack, even if there is only one unit in every city, and you will lost men at every siege 'cause of the towers. In rome 2, I think you will take a undefending province in 5 turns. But Tim_Ward explain it better than me lol

Also, it's only good things for gameplay.

1) Instead of a boring siege battle, you'll get a more interesting field battle.
2) It gives you, the player, a Decision to make. Do I try and relieve the siege or do I let the AI take that city for now? Do I divert Legio IV from the front lines or do I raise a new army (*can* I raise a new army?) Making Decisions is the bread and butter of strategy games, and more of them you have to make, and the more meaningful they are, the better in general the game is.
3) It will make the game more challenging, as it means just having enough units in a city to see of a full stack trying to take the walls is no longer enough to keep it safe....

[Bethencourt]

No because you will have A LOT more time to call reinforcements. For example, one province with 4 cities with walls in Rome 1 (even if there is no province in R1) = 9 or 10 turns to take all the city with one full stack, even if there is only one unit in every city, and you will lost men at every siege 'cause of the towers. In rome 2, I think you will take a undefending province in 5 turns. But Tim_Ward explain it better than me lol

Which is part of the fun, here they come! the reinforcements! But now loosing cities undefended will be, well another city lost, no possible reinforcements. Do you think taking a province 4 cites in five turns is fun? doesnt that mean Italy with three? regions then 15 turns? well if you add one army per region maybe 25 turns? For sure the numbers are wrong.

[Ichon]

I guess the player will have to rethink their strategy than, instead of playing in the exact same way they have for the past several games. Maybe they'll have to consider defensive armies, rather than purely offensive forces. If your tactics are to send out every army to the frontlines and expand quickly then there should be negatives to consider.

Exactly- players have had a huge advantage in earlier TW games able to rely on walls in sieges where AI couldn't. For the first time players will have to consider leaving significant defensive armies.

Capturing walled regions in most TW games as often been easier than a field battle since AI often divided its army. The largest part of time to expand for me has always been occupation time where army just sits still until newly conquered region is pacified.

[baptistus]

Do you think taking a province 4 cites in five turns is fun?

It is an example To resume, in R2 you will have 2x less time to call reinforcement if a province is Under attack by a full stack, so you have to take your decisions carefully. In R1, during the sieges of the first cities , you have enough time to create a strong army in your major city to counter attack quickly without call reinforcement from the other armies. "Making Decisions is the bread and butter of strategy games"

[Bethencourt]

It is an example To resume, in R2 you will have 2x less time to call reinforcement if a province is Under attack by a full stack, so you have to take your decisions carefully. In R1, during the sieges of the first cities , you have enough time to create a strong army in your major city to counter attack quickly without call reinforcement from the other armies. "Making Decisions is the bread and butter of strategy games"

As walls were in those times and in strategy game. If you can raise an army in such time, so it is a loss of gameplay, maybe it is not walls fault but the recruitment system.

Exactly- players have had a huge advantage in earlier TW games able to rely on walls in sieges where AI couldn't. For the first time players will have to consider leaving significant defensive armies.

Capturing walled regions in most TW games as often been easier than a field battle since AI often divided its army. The largest part of time to expand for me has always been occupation time where army just sits still until newly conquered region is pacified.

And what will make them less easy now if there is going to be less armies? Well this "less armies" thing I am not too convinced really because watching campaign screeshoots there seems to be plenty of them.

[Ichon]

And what will make them less easy now if there is going to be less armies? Well this "less armies" thing I am not too convinced really because watching campaign screeshoots there seems to be plenty of them.

Not sure what you meant about less easy but I agree from the previews so far there does not appear to be a lack of armies on the campaign map... not entirely sure what SHogun 2 campaign map looks like without fog of war most campaigns but I'd guess actually a bit less armies than we see so far in Rome 2. Although Shogun 2 AI surely used too many fleets maybe that won't be such a detriment to it now.

[baselhun]

Bumping old thread. Ca will "counter" this issue by reducing "cities"

[WhiskeySykes]

I'm in the minority, at least of opinions back then.

CA will not innovate siege warfare, until they've taken action points out of the abstract, formed a rudimentary logistics mechanic, and the cai can process slightly complicated behaviors, based on these stats.