My Units Seem to Take too Much Damage

[vonriel]

I'm not entirely sure if this is a glitch, or working as intended. I'm currently playing as Spain on King, and am in a war with Japan. In the course of the war, Japan, despite being very far behind technologically, has proven to be impossible to match on the field because he's out-damaging me by so much. This includes, but is not limited to:

• I took a 16 strength city from him, using composite bowmen and catapults. The bowmen did 10-15 per hit, the catapults did ~30 per hit. Next turn, it's at 13 strength, he attacks it with a swordsman and composite bowman. His bowman does 53, his swordsman finishes it off (I didn't grab screenshots of the damage)
• My pikeman, fortified in a jungle with drill 1, took 15 damage from a regular archer. As it was an archer, it likely didn't have barrage/volley like the rest of his bowmen, I forgot to check its promotions though.
• My Crossbowman took 41 damage from a melee attack done by a regular spearman. The spearman only benefited from the damage bonus a great general provides and the base penalty for being on open terrain.
• Another pikeman, fortified on a hill/forest with no promotions, took 30 damage from a composite bowman's attack. The bowman had barrage and volley and a great general strength boost. Still, 30 damage to a 20+ strength unit when the bowman had, at best, 17 strength seems very fishy.
• A Crossbowman taking 24 damage from a composite bowman. The composite bowman (in this picture and others)had these promotions.

In general, it seems like I'm taking a lot more damage than I should be from Japan, with the possible exception of the last bullet, as I'm not terribly sure how ranged damage is calculated. In the end, I don't know if it's something I have to grin and deal with, as a conscious design decision to punish the player even further than they already are punished (and if so, I'm not playing any more of this patch, ever) or if it's some sort of glitch.

 

[jwerano]

Im glad im not the only one who noticed this. It happened to me in a game vs Indonesia. Early game archers were hitting my city for 37 damage. I had completed the tradition tree so I sould have been doing extra damage with my city but my city which was fortified by an archer in return was only able to do 20-25 damage to the archers, and my archer only did 15-20 damage. It seems something is wrong with values. Maybe the AI is getting the barbarian bonus against the human player?

 

[alluton]

AI might have extra damage to city in order to compensate AI's lack of skill to actually assault city.

 

[mystikx21]

The only one of those examples that actually seems fishy is the first one. And there the AI does get an anti-city promotion to help it with conquest generally and city defences are destroyed when conquering, which weakens cities substantially. The AI bonus could be tweaked further in this regard, or may be combining with other bonuses from promotions (volley is anti-city for archers, plus anti-fortified, checked).

 

[jwerano]

AI might have extra damage to city in order to compensate AI's lack of skill to actually assault city.

Ok that makes some sense but it feels a bit too strong. I hope it gets tweaked to be a bit weaker. I cant see how I could possibly match the AI in combat under these circumstances. I'm an army minimalist and it seems i will have to shift my entire early game strategy to building 3-4 extra units instead of trying to get an early wonder.

 

[Tomice]

It doesn't seem too unrealistic for a Swordman to do a lot of damage against a str 13 city that is at very low health. Maybe it had a siege promotion? Remember these units are supposed to be able to take cities even without help from siege units.

Don't know how much health exactly a city gets back after being conquered.

 

[mystikx21]

Would be around 150 HP post conquest. Swordsman doing a lot of damage isn't surprising. The bowman at 50 would make sense, if they had volley+the AI city attack bonus.

 

[lindanealmck]

The cities seem squishy to the point that the during a war; my allied civ states were capturing enemy cities with carracks or musketmen, some of then in the size 16 range. A hit from a Battleship or Artillery would take a city from full health to zero in 1 shot. I did have a significant tech lead which may have made the difference - not sure.

 

[Ahriman]

Part of the problem on city capture is that the AI (especially for city states) doesn't seem to understand the need to keep a garrison. Without garrisons, cities are very weak and easily captured.

 

[Tomice]

Part of the problem on city capture is that the AI (especially for city states) doesn't seem to understand the need to keep a garrison. Without garrisons, cities are very weak and easily captured.

Didn't Thal decrease the strenght of ungarrisoned cities recently? And increase the effect of melee units on city strenght? This might aggravate the problem.

I can't really find the discussion any more, so I might be wrong.

 

[Ahriman]

Didn't Thal decrease the strenght of ungarrisoned cities recently? And increase the effect of melee units on city strenght? This might aggravate the problem.

Right. This is why it's much more of a problem in CEP than in BNW. Cities without garrisons can be very weak indeed.

 

[Tomice]

Can you help me remember the exact reasoning behind it? Weren't AIs to weak on the offense? Or was it about making melee units better for city defense?

Gesendet von meinem GT-I9300 mit Tapatalk 4

 

[mystikx21]

Both. The other issue was city strength was extremely high at most tech levels using default strength. That put a higher premium on city buildings like walls as well by lowering the strength. The city is supposed to be harder via buildings adding both strength and a larger volume of hit points.

 

[mystikx21]

But there does seem to be a very high AI city attack bonus being used as well (100% on both city attacks and defending against city ranged strikes). That's way too high and should be reduced.

 

[Ahriman]

Can you help me remember the exact reasoning behind it? Weren't AIs to weak on the offense? Or was it about making melee units better for city defense?

Gesendet von meinem GT-I9300 mit Tapatalk 4

The other reason was that Thal was trying to make it more meaningful to have a non-ranged unit as a garrison. Melee unit garrisons have always been pretty pointless, since they couldn't attack without taking damage, and gave barely any strength bonus to the city.

But I think it may be wiser to accept that melee units will never make good garrisons, and accept that garrisons should be ranged units and try to make sure that the AI can always provide a ranged unit to do so.

I'm not sure though. Thal's change here is an interesting experiment, and I've mostly seen the problem with city states, which are terrible with their units in general, major civs still seem mostly able to get a garrison into their cities.

The attack bonus is particularly weird with city states though. For some reason, two of the city states that I became allies with were at war with each other (I guess they were at war when they were both allies of other players who were at war, then I took over both alliances and they seemed to keep fighting), and one of the city states was able to conquer the other with a caravel and a musketeer.

 

[stackpointer]

But there does seem to be a very high AI city attack bonus being used as well (100% on both city attacks and defending against city ranged strikes). That's way too high and should be reduced.

Does the 25% on the <CityDefense> tag actually do anything? When I was testing the bonuses the AI get for the other thread, I didn't actually see the percentage in the damage calculations.

The attack bonus is particularly weird with city states though. For some reason, two of the city states that I became allies with were at war with each other (I guess they were at war when they were both allies of other players who were at war, then I took over both alliances and they seemed to keep fighting), and one of the city states was able to conquer the other with a caravel and a musketeer.

I've seen the never ending city state wars bug in one of my own games (and in another thread). This seems to be a bug introduced with the release of BNW (or the summer patch).

 

[Tomice]

I'm not sure though. Thal's change here is an interesting experiment, ...

Agreed. I'd try changing stuff that has no other benefit first. Always garrisoning with ranged units is a no-brainer, some choice would be nice.

 

[Ahriman]

I've seen the never ending city state wars bug in one of my own games (and in another thread)

Right, it's a weird bug, but it's even weirder that one of them was able to successfully conquer the other with just a couple of units, because of how weak an unoccupied city is and the large city attack bonuses.

 

[mystikx21]

I have no major issue with the garrison unit change. It would be better if the AI would stick to using ranged units and attacking, but the AI isn't terrible at that (just not perfect). And this does at least provide some margin of choice, as well as a means of healing wounded melee units or using them for quick counterattacks to weaken nearby threats.

The bigger question for me surrounds the later ground unit strengths in general, which was a much larger change and should probably be re-balanced closer to the old GEM figures, and the very large AI city attack bonus.

@stack. I'm not sure if the citydefense tag does anything but it is an available default unit modifier.

It's possible it is acting like the old civ4 tag for garrisons rather than as a counter to city ranged attacks. That would be noticeable in increasing garrison strength on say, swordsman or catapults for testing. It would also be noticeable in increasing garrison strength generally for AI units over and above what the human gets, in which case, it would be pointless to give the AI both a city attack and city defence modifier. I have not noticed if it is actually working as Thal intends it or if this is its function by testing it. If it's just doing the latter, we would need to invent a new function to do what is desired by reducing city attack strengths (only) against siege/sword units rather than a general cover promotion which would do the same thing but potentially severely nerf ranged attacks.

 

[mystikx21]

From looking through the dll and the civil war scenario, it appears the city defence modifier is intended to be something impacting garrison strength, if anything.

That modifier will need a complete modification to look for defence against city ranged attacks to be viable as a special effect weakening city attacks. Some alternatives might be to

a) weaken city ranged attacks. Relying on garrisoned units and armies to do most of the fighting. This would weaken the point of having the city ranged attack at all and wouldn't provide any special advantage to units with city attack effects like swords and siege weapons.

b) Use cover or anti-ranged promotions to provide the same effect and increase the strength of ranged units while leaving cities at the same value. This would make cities and non-siege/city attack units more vulnerable to ranged attacks though. It would also weaken elements like the battering rams powerful anti-city effects on attack/defence.