So long as I can conquer a town in 5 turns, the game will be a pushover

Another solution to this problem could be eternal guardian mechanic.

Exactly and I am for this guardian unit thing 100%. Militias simply won’t last with one fireball and lore wise, it would be perfect. Finally, I don’t want to steamroll a whole faction in couple of turn: no fun; I want to reduce them to their capitol and then, if I am powerful and worthy enough, defeat them by killing their sovereign and the guardian (a semi-god please!) giving a prize (like the wildland bosses). Think this way. Only mid to late games are heroes able to take out a boss in a wildland zone. Taking a faction’s capitol should be hard, almost impossible and mostly, fun. Fighting a huge spider, an angel, a demon, or a titan to kill a faction would be a great fealing. Whipping militias is simply boring and gives the idea that the capitol is a worthless seizing. This guardian thing would be a great thing for faction’s differentiation, for balance and well, simply for fun!!

I would say the OP's observations are spot on.  The Eternal Guardian is another subject altogether.

When the AI gets better at fielding full armies, most of this will be solved

*BUMP* Again, i don't think we want "eternal guardians" in this game. It feels like a copout mechanic for this game. Other games use different solutions and i don't think this one fits the lore or the gameplay. I agree that getting the ai to better manage its armies is the major goal. Tarth is the only faction that should be fielding small armies, and even then it should just be as a precursor to a major attack.

I would rather see stronger militia as the game goes on, over an eternal guardian.

Tougher cities would mean an emphasis on bigger armies, which would help deal with the annoying large number of small stacks running around problem.

My support for the OP's proposals as well. Very well presented! Taking cities should be a big endeavor, more so in the case of the capital. 

The wall and wall-breaker concept sounds efficient, simple, intuitive and would add another layer to gameplay of city conquest. Perhaps the capital (possibly pre-built in a 4/4 location at the start of each game) would start with a maxed-out wall already in place, and later settlements could build and expand walls as desired. 

Oh, the wall-breaker idea, it was tried in AOW1, it didn't work well there.

... and it would not solve the problem of insufficient army balance and the suboptimal way the AI deploys them

What do you mean? As far as I can remember it worked just fine. Only slightly weird thing was that a single battering ram could conquer an undefended town..

This game really needs proper siege mechanics, with actual walls on the battle map. But since that's not an option apparently, here's what I think should be done:

1) Add siege mechanics somewhat similar to the Total War games. If we can't have actual siege battles, at least this should somewhat compensate. So an army can attack a town that has no walls straight away. If it has walls, it needs to spend X turns building siege equipment. X can depend on the attackers technology levels (certain techs reduce X), the defenders wall level (better walls increase X), and hero special abilities on each side (traits like Siege Engineer, Castle Commander, etc). Also think of a way for magic to interact with this, like targeted earthquakes to bring down walls, pestilence to weaken either side, or floods to slow down the attacker.

2) Give a city way more militia. Conquering a city should be a major undertaking from turn one. In a world like in this game, a cities population wouldn't just sit around waiting to be dragged off to a Quendar slave pit.

3) Rebalance all units and their leveling, especially their health. Should be more along the lines of 10 HP + 1 per level. A level 2 unit should only have a slight advantage over an otherwise identical level 1 unit, so it might win most of the times, but could still be seriously injured in the process. This way even simple unleveled militia will continue to pose at least a slight threat to trained troops.

4) Conquered towns should require a garrison to keep it in line, at least for a number of turns. This way you can't just blitz conquer everything with one army.

5) Razing cities should have serious consequences. The player razing the city should take serious diplomatic hit not only with the original owner, but also with the original owner's friends. So committing atrocities will eventually bring in more opponents against you.

6) Combining 2) and 5), have diplomatic relations influence the strenght of the defending militia. So a Kingdom attacking another Kingdom's city over a border dispute would face maybe 10% of the population as a militia. An Empire attacking a Kingdom's city would face 20%. And after that Empire conquers and then razes the city, at the next city it will face 50% of the population. The one after that, 90%. Knowing there is a very big chance they'll be killed when conquered, the people will take up arms to defend themselves.

No offense, but if you think the AI has a horrible time taking back the city it just lost NOW, how the heck is it going to ever manage to siege the city with your new mechanic? Or heck, how is it ever going to INVADE anyone that has a proper defense (meaning you). The AI is having trouble just dealing with unrest from conquered towns. The biggest problem isn't walls or siege, but that the AI isn't putting together proper competitive armies. They are using very small groups (3/5) of units when they should be rushing for squads/logistics (for 7/9). If they can just do that, the game would be a much bigger challenge.

The size of units is also why militias just fail. Groups of 3 against groups of 7? It's not a funny joke. I wouldn't mind seeing militias either start at the level of the city (for more HP - so they can last longer), and maybe give them some basic armor... but the problem is still the size of the units.

Totally agree: I had a paragraph prepared saying as much and forgot to copy it in. You must have read my mind. Will put back in.

Well, all I can say is that the AI in Galactic Civilization 2 (Stardock's last game) did almost exactly the same thing, with transports instead of siege weapons. It worked pretty well then.

Whenever I play the game, the AI almost always has a higher military rating then I do, but I can still walk into their towns and take them before they can respond. The problem is not that they don't have good enough units, it's that they don't have them in the right place. And to be honest, I don't suspect the AI will ever be particularly good at this. We could easily beef up AI income and tech rates to give them better armies, but it would feel like a false, unfair advantage. Instead of giving the them a huge unfair advantage in order to compete, wouldn't it be more fun to play a game they were more suited to?

I played GalCiv2 quite extensively, and to be quite honest, they never managed to invade me... ever, even on the hardest difficulty, due primarily to starbase protection bonuses. A couple of garrison units and the AI just get totally WTF ownt the moment they get in range. But at least in GC2,  you had to build tons of constructors to get those SB online, here... barracks, monuments, altars, walls? Cities are easy to take if they are undefended, but if they are properly garrisoned... good luck. The problem right now isn't a lack of a siege mechanic, but that the AI doesn't have the right armies. Last game I played (I wrote about here: https://forums.elementalgame.com/424776 ) Gilden had SOOOOO many defenders that he literally couldn't fit them all in his cities, but I still ran them over with my units because he was using very small groups (and those were high level cities with plenty of militias). Would it have change anything it I needed to bring a unit with "wall-breaker" trait? Nope. Worse, it doesn't help early game rushing at all, because you won't even be able to build a wall until level 3 cities.

Adding a siege mechanic does nothing to help the AI, it hinders it. Instead of building combat ready units, it now has to worry about including wall-breakers in its armies, and properly protecting such armies. Afterall, if the player cast a strategic spell to destroy or immobilize that wall-breaker army... what is the AI going to do? It'll stand around doing nothing. The player? He'll have his wall breaker ready outside enemy territory, merge that wall breaker into his main army the moment war is declared, rushes the city, nothing really changed.

If there's anything that seriously needs changing (aside from better AI unit/armies) it's that roads in enemy territory needs to stop giving movement bonuses. That would at least slow down the invasion somewhat and give the AI a bit of time to respond to war declaration. THAT is the main difference between GC2 and FE. In GC2 it takes a while to get anywhere... not so much in FE. There's just no time for the AI to react.

One way to help things out would be to create a set of societal norms.

So make it so that when you declare war, you get a few turns of preparation before war actually starts. If you break this tradition, then you get an oathbreaker penalty with all other players, and maybe you even lose diplomatic capital and such. It might even generate unrest in your cities!

I also think that enemy units should take attrition damage while in enemy territory. Maybe you could build a structure in your city to increase the damage in its area of influence. It would be something like 1 or 2 damage per turn but it could add up over time. Scouts should get a perk that prevents this damage.


I think AoW 2 did city battles pretty well, though I never have much trouble with AI opponents in 4x games. The main problem that the AI always has is a lack of understanding on how to make a killer stack. They don't understand that having 5 powerful units is better than 20 weaker units.

Until the developers figure out a really good formula for calculating the threat of an army, and also teach the AI how to use fight optimally in a tactical battle, the AI will never put up any resistance to a seasoned player.

I like being able to rush enemy cities. I am okay with an extra militia for the capital city, but anything more than that is too much. I don't think that taking out one opponent in the early stages is a problem. If it makes the game too easy, raise the difficulty, add more opponents, etc. I never had a problem with this in any other 4x game.

As for taking over cities, I think that losing half of your population is rediculous and way to penalizing if you lose and retake a city to have only 25% of the starting amount. I do agree that militia should not instantly convert. I also think that they should copy the revolt system of AoW 2 where if you don't have enough troops the city will kick you out and become independent.

Finally, I think roads should be nerfed by about 50%. In modern warfare roads do function for both sides equally, probably the same for a fantasy world. I do think there should be a penalty though, maybe 50% again.

Game design vs. AI design are often at odds.

Players are very good at creating a very sharp tip to their spear. AIs, by contrast, are good at defense in depth.  

The general problem in FE, is that the current design gives a big advantage to the player with the sharpest spear and that's inevitably, the human player.

Even in online play, that's how it usually sets up.  In strategy games like Starcraft, Company of Heroes, or Age of Empires, the strategy works against humans too -- the player will create their super army and blitzkrieg their way into the enemy's vitals.

In Galactic Civilizations, since I wrote the AI and design the game, I was motivated to make sure the design was more weighted to the defense in depth.  

First, planets were typically far apart. 

Second, transports were required to take planets.

Third, transports moved slowly.

Fourth, players either had to plod around with transports in their fleets (which gave the AI time to counter) OR they had to control the board between the invasion origin and invasion destination which gives a huge advantage to defense in depth.

I don't think Kael is likely to make a drastic change like having an occupation unit.  But there are plenty of other things that came come into play here that would help.

I know it's been said before, but... one way would be to make capturing cities a lengthier, more difficult affair.  If an attacker had to seige and/or wear down the defenses of a city over a number of turns before taking it, wouldn't that help with what you're describing?

I've always preferred games where cities/planets are hard points.  I like the city defense mechanic in Civ5 and Warlock for that reason.  Distant Worlds also struck a nice balance.  You can smash a planet's defenses as fast as you want, but unless you're willing to conduct orbital bombardment (a diplomatic no-no), you then need to put a lot of boots on the ground and begin the long slog of actually capturing the planet.

Champions unable to capture towns is a great idea. Same with solo champions in foreign lands.

I think just upping the number and quality of Militia that a city gets, and PERHAPS making a meter showing how much of your militia has been depleted by recent combats would go miles.

I definitely want cities to be a stronger point than they are now. With how strong Champions are and how time consuming it is to raise an army early game, as well as how prohibitively expensive it is to garrison a significant force of troops at any point in the game, cities should be much more difficult to take.

I'm just spitballing at the moment, but I don't think it's unreasonable to have 2-3 Militia per city level equipped with Spears and or Clubs/Shields and the second best armor you can build. There are plenty of truly deadly wandering monsters out there which would still be able to roll up a city without an additional stationed army, but it would help really limit the number of first 10 turns wandering monster deaths as well as help draw in the early game rush kills.

This is a game of conquering your opponents.  Let's not make it impossible to conquer cities.  I think the game needs minor tweaking at most.

I like the idea of requiring x number of turns to siege a city (based on city level?). This should allow the AI time to react. If they counter by bringing in reinforcements, the attackers would now have to face the reinforcements AND the city garrison in the same single battle. Would make attacking a city interesting, to say the least. I'm sure this is how it happened throughout history - attacking army has to tackle both the city garrison and any reinforcements from the flanks.

Personally I think that the underlying issue is that that the "strategic map game" is, well, messy. Adding some ZoC rules would allow players and AI to intercept enemy armies before they reach the cities, this is currently rather hard (and messy) to do. This would greatly enhance the strategic part of the game as well. A simple example of a ZoC rule: moving next to an army that you are at war with stops movement, only allowing you to attack it this turn, or wait and move away next turn.

Something like this + one-tile cities (which might be coming next beta?) could solve a lot of problems.

I tryed to invent a mechanism, which relatively easy to script, doesn't turn upside the recent mechanism and helps the small. Oh, and think this isn't prevent the AI to conquer towns, because this idea doesn't give more deffense to towns.

the post: https://forums.elementalgame.com/424857

I strongly agree with this.  Implement ZoC in strategic mode.  When you move next to an enemy, you lose all ability to change strategic map squares.  You can use your remaining action points to attack enemies (1 for each battle), but not for moving past them.  And if you want to move past through a group of enemies if there is an open path, you can only move one square per turn whenever you move next to an enemy stack or unit.

The ZoC would slow down a bit the conquer progress, but only a bit. The player could use it well, the AI not so well. At the and we have a new opprtunity, where we are better than the AI.

On the other hand: The AI must split its armies. There must be an army outside to block your way, and there must be troops in the city.

What is the better option? To split its army, or to defend the city with all of its units?