[Balance] Ranged Combat (Units & Cities)

Lord Tirian

Erratic Poster
Joined
Nov 30, 2007
Messages
2,724
Location
Liverpool, UK
My personal impression is that ranged units are in most cases a lot better than melee units. Nowhere is this as obvious as in city sieges - I'd rather have 3 archers and a catapult than 3 swordsmen and a catapult (apart from the unit actually taking the unit, of course).

This is partially because city counter-attacks are very strong and combined with their ranged attacks, they can decimate melee attackers with ease. This isn't necessarily all bad, but it's lopsided and seems to affect the threat from barbarians as well (which love melee units).

I have the impression that this is the general consensus (ranged is too good on the offence), but, of course, it's interesting what the discussion spawns.

Personally, I think changing city combat would go a long way towards fixing this (the interplay on the field seems to be okay, apart from the AI being a bit stupid):
  • City strength should scale less with population, boost defence buildings to match (hey, a reason to build castles apart from Neuschwanstein!)
  • Less overall strength, more hit points for cities - this encourages melee attacks while penalising ranged attacks
  • Give siege units some sort of "cover" promotion for free, at least against city ranged attacks - they get chewed up too easily.
  • Give cities an automatic counter-attack against ranged fire (possible? might be too much)

Finally, if I remember correctly, city strength doesn't factor into military strength at all. If you have strong cities and gold reserves, that can cause a major derail in an AI attack.
 
Most of this was in CEP as unit changes (and changes to building defences).

There was a minor reduction in population and tech to city strength, and a decrease in base strength, with an increase in the effects of garrisons and base hit points went up 100.

City defences were increased in hit points. Palaces also gave the city more HP (helped City States). That definitely helped castles+, without making cities stronger in hitting power.

Siege units have a ranged defence bonus and are stronger (city attack bonus was lowered to compensate, they have a penalty against units instead).
Archers had a city attack penalty.
Swords had a city attack bonus.

I don't think the counter-attack against ranged fire is needed.
 
Just throwing it out there, but there is a feature in the DLL that allows variable city bombard damage and range (including LoS) to be set from Lua via an event
 
Palaces also gave the city more HP (helped City States).
Oooh, I like that, that's clever. My worry about lowering base strength was making city states and civs in the early stages vulnerable to rushes and barbarians. Palace buff negates that.
Just throwing it out there, but there is a feature in the DLL that allows variable city bombard damage and range (including LoS) to be set from Lua via an event
Yeah, I remember, we had this discussion earlier in the original thread - but the more I've been mulling over it, the less I think it's necessary (as neat as it is) to make ranged less powerful.

On another (and somewhat related) note: what about forts? Where cities do too much, I rarely find forts do enough (citadels are wonderful, though). Again, ranged can pick off units in forts easily but now they have even less chance to counter or lose their defence bonus by leaving the fort - whoward's "No Followup From Cities" could somewhat help with it, though.
 
The one aspect of the city bombard range that is of note is the possible need to allow cities to attack out to 3 in the later game (either from a tech or from say, the military base building). Arty and BBs are really good simply because they're effectively invulnerable.

I believe CEP also added a ZOC effect to forts for damage like the citadel. If that helps.
 
The one aspect of the city bombard range that is of note is the possible need to allow cities to attack out to 3 in the later game (either from a tech or from say, the military base building). Arty and BBs are really good simply because they're effectively invulnerable.

I think that is something to consider, though I would want to weaken city attacks by that phase to compensate.

May be too radical, but you could consider the 3 ranged attack at castle instead of military base. The medieval period was the period of defense warfare. A castle wasn't breached without some serious effort. A 3 ranged attack (that out ranged other units of its era) would certainly highlight that.
 
The main differences between ranged and melee units is that melee units take damage every time they attack and that ranged units can attack from afar and thereby concentrate force. In a game where the difference between losing units and retreating to heal and with constricted movement from 1 Upt ranged units have a huge tactical advantages. If nothing is done to these fundamental differences you have to compensate with huge differences in strength to the point that melee units kill ranged units in 1 attack to compensate for the tactical advantage. If you want true core balance I propose that you eliminate the damage melee units take in close combat. Then your units do damage in your turn and enemy units take damage in their turn. Then they can actually attack cities instead of just absorbing damage and being necessary for capturing the city. A warrior might then actually win a 1 on 1 with an archer.
 
@chumchu

What you are describing, I think, is making MELEE units 'range 1 RANGED' units, like GG/MG.
Then the only difference is WHEN the damage is taken.

MELEE vs MELEE

each take damage in their turns, player first then AI

RANGED vs MELEE

The RANGED unit (range of 1) deals damage in the player's turn and ONLY receives damage in the AI turn, the MELEE unit receives damage in the players turn AND possible counter-attack in its turn

As I see it keeping the distinction between MELEE and RANGED but introducing a 'no damage on attack in your turn' for MELEE is counter-intuitive and possibly a nightmare to implement.
 
I don't think that's actually a solution either. The ranged unit still has the risk of being attacked before it can issue damage, just as a melee unit does if it is attacked at range, and is unlikely to one shot most units to death that they could never respond (without a major tech difference).

As a further problem with that solution, instead of melee units trading damage on an even basis, melee units would be at a disadvantage when counter-attacking such that they would deal much less damage. It's better and fairer to just get it all over with in one round. Under that proposed scenario, I'd probably use no more than a handful of melee units at all. Which would be even worse than the current balance.
 
The result would be equivalent except for the the keyword which has effects for promotions/social policies et cetera, that is the distinction we need to keep. I do not by the argument that this is counter-intuitive. As it is now we have attacks with counter attacks at range 1, attacks without counterattack at range 1, 2 and 3. I propose to eliminate the first category of attacks to simplify the game.. It is merely getting all units to follow the same rules. Civ 5 is a turn based game simulating reality. Either you deal damage in your turn or you have some units dealing damage in both players turn and some only in their own or every unit dealing damage in every turn.

For comparison look at other turn based games. The combat system in Heroes of Might and Magic is close to that of Civ V except for the initiative system and it it very well balanced. Civ 5 is like that except that ranged units have full strength and no retaliation in close combat plus being able to move and shoot. This is further compounded by the melee units having so low movement (2 in open 1 in difficult terrain) that they are constantly in the way of each other and that unit heal very quickly so that retreating a damaged unit is vastly better than having it eliminated. In Civ 4 or Advance wars all units have retaliation and none have ranged attacks this is also balanced. As it is now ranged units totally dominate because of these tactical advantages and most armies consist of ranged unit plus a few token melee unit to eat city bombardment and take the city after the ranged units have bombarded it down. This is very hard on the AI which means that humans can take cities, losing no unit or perhaps one of their melee units while the AI lose entire armies trying to take city defended by 1 or 2 ranged units.

If it is nightmare to implement then the idea falls on that. Then I would argue for the second best solution, keep nerfing the ranged units. I do believe there is a role for units attacking at a distance because the map is so cramped but we could decrease their range, then they would not get first strikes at melee units as often. This would make the system more like Diplomacy with large traffic jams. We could also change it so that ranged units do not do damage to melee units when they are attacked by them. This would give ranged units an actual weakness. We could also open for counterattacks against ranged attacks which would work gamewise but would be an even bigger and harder change than the one I propose. I strongly endorse doing one of the above to correct the massive advantage of ranged units.

If you do not you will need to decrease their strength and ranged attack at least to the point were a warrior does 60 damage to an archer and the archer 15 to the warrior. Further, ranged units will still rule at taking cities because of the massive counterattack suffered by melee units. You will need to decrease the strength of cities as well in my opinion.

Regarding counter-attacking. As it is now there is a disadvantage to attacking in melee versus melee because of 1 fortification bonus and defensive terrain bonus and 2 you open your self up to counter-attack while the opponent can retreat his melee unit. With my change the balance would change to be slightly in favor of the attacker but as it is the defender has a huge bonus anyway. I do not agree with your argument that my change would reduce the use of melee units. They would be good at taking cities and necessary to defend your own cities and ranged units.
 
A couple of thoughts here:

City bombard range – Whoward's mention of modifying city bombard range shouldn't be overlooked. Scaling a city's bombard range with eras would be a huge boon for easily-destroyed catapults early on, and could prevent artillery from being the 'must have weapon' in the late game. I've waited to go to war many times until I have artillery, and that is not the most fun gameplay strategy. Think of it this way: we can either spend lots of time tweaking siege units to make them more survivable, or we can simply make this one change and the issue becomes moot.

Melee attrition for sieges – This is an idea I've been milling about with for a while now. What if melee units, either melee ships in a blockade or melee units around a city, reduced a city's ability to heal itself and/or slowly damaged cities passively? It would be a form of blockade/siege – starving out your opponent by cutting off supplies with melee units. Seems like this could help even out the power of ranged versus melee, as a melee-heavy army could, theoretically, slowly reduce a city to a low enough state to be captured (more slowly than with ranged, mind you), especially if a city's bombard strength was reduced somewhat. We could also use this to make some snazzy new traits for leaders – Attila or Genghis, for example, could have their 'siege attrition' rate doubled, making their armies all the more terrifying.

Avoiding Rock-Paper-Scissors – I've said this before, but I want to stress that we need to build the CBP to be AI-oriented. If units become to specialized for specific tasks, the AI is the loser. I think CEP made some great changes to the units in the game, but I also think that the more we change unit-versus-unit dynamics, the bigger the chance of making the AI perform poorly.

G
 
City bombard range

At the very least, IMHO, indirect fire should be removed until Arty. I think units should be able to hide behind that mountain/hill when approaching a city.
 
At the very least, IMHO, indirect fire should be removed until Arty. I think units should be able to hide behind that mountain/hill when approaching a city.

Agreed. Cities should be vulnerable for most of the game, not bastions of rock-throwing and rocket-launching as they are in vanilla BNW.
G
 
Agreed. Cities should be vulnerable for most of the game, not bastions of rock-throwing and rocket-launching as they are in vanilla BNW.
G
I'd like to see cities that can withstand one or two (weak) units on their own, though for a simple reason: less need for early unit garrisons. Personally, I found that you can actually use your starting warrior for something one of the better changes going to Civ5. Getting rid of indirect fire until units get it sounds sensible, though.

That is also a problem with reducing early city range, it sounds like it could do wonders for siege but will also buff archers. I don't think non-siege ranged units really need a buff.

I like the blockade idea, especially if then castle, for example, make the city more resilient against the blockade. More reasons to actually invest into defensive buildings is good. Alternatively, I'd suggest giving harsh defence penalties for starving cities, means the AI's obsession with pillaging translates into a boon for taking cities.
 
I would say that there two other things that are needed in Civ V:s combat system. More room to maneuver and less of a gap between eliminated and damaged units. The first one is harder to address. Decreasing or removing the ranged attack of cities will do a large difference as movement is restricted by avoiding being within 2 tile of the city as the city's ranged attack can eliminate units. Spacing cities further apart would also make a difference. For instance, why not allow cities to work an additional ring of resources. It also important to have large tracts of bad terrain to get larger areas between cities to fight field battles which is why I am against boosting the yield of marginal terrain like Tundra and Ice. Then you could increase the movement of melee units and get a more flowing combat system.

To lessen the difference between damage and a solution is to make it cost gold to heal units which is both realistic and has interesting gameplay effects. I'm thinking around 1 gold per healed hammer. Healing a warrior from half to full strength would cost 20 gold (40hammers/2) and a pikeman 45). There would need to be some changes to promotions and pillaging should give less hp.

The effects are firstly that it depopulates the map as you would now sacrifice more units than before. With good balance you would want to heal some core units with high xp and sacrifice the others. Secondly that it makes wars less all or nothing, since the difference between 1 and 0 hp it not as huge as before. As it is now, the winner of a battle loses maybe 1 unit and has to wait a while for the others to heal whereas the loser loses an army. Think of the outcome failed sieges. Thirdly that it would increase the difficulty of the game with proper AI coding. As it is now the player can defeat an AI siege with minimal losses and proceed to take all their cities barring happiness problems. The AI can neither keep units alive nor pillage as it is now which leads to alot of units shuffling around your cities while you kill them with archers. The Japan AI does not do this as much however and can therefore bring its production superiority to bear and take your cities. It is probably easier to code it to grind your cities than to rotate units and avoid casualties.

For further background and discussion see this
 
I'd like to see cities that can withstand one or two (weak) units on their own, though for a simple reason: less need for early unit garrisons. Personally, I found that you can actually use your starting warrior for something one of the better changes going to Civ5. Getting rid of indirect fire until units get it sounds sensible, though.

That is also a problem with reducing early city range, it sounds like it could do wonders for siege but will also buff archers. I don't think non-siege ranged units really need a buff.

I like the blockade idea, especially if then castle, for example, make the city more resilient against the blockade. More reasons to actually invest into defensive buildings is good. Alternatively, I'd suggest giving harsh defence penalties for starving cities, means the AI's obsession with pillaging translates into a boon for taking cities.

Indeed- pillaging could reduce a city's health-recovery rate. An easy fix for archers v. siege is to give siege units cover to start, and to forbid that to ranged units. Ranged units may be strong, but they are flimsy.

With the new happiness system, most people will garrison now anyways, so that's not too much of an issue.
G
 
Melee attrition would help some problems but why create a whole new mechanic. If we want more realistic sieges we should remove the city's ranged attack instead. Then we can get situations where the city can not be taken but neither can it remove the besieging army, a real siege. What will happen is that the besieging units will pillage and place themselves on food tiles so that the city starves while they wait for siege weapons to come to their aid or a relief party comes to the aid of the city. This means that you might need a garrison unit to clear away barbarians early, which adds more interesting decisions, like going for archery early instead of pottery or how long to wait before building your defence unit/moving back the early warrior.
 
Melee attrition would help some problems but why create a whole new mechanic. If we want more realistic sieges we should remove the city's ranged attack instead. Then we can get situations where the city can not be taken but neither can it remove the besieging army, a real siege. What will happen is that the besieging units will pillage and place themselves on food tiles so that the city starves while they wait for siege weapons to come to their aid or a relief party comes to the aid of the city. This means that you might need a garrison unit to clear away barbarians early, which adds more interesting decisions, like going for archery early instead of pottery or how long to wait before building your defence unit/moving back the early warrior.

I'd say your idea and mine are both new mechanics, as both would involve a lump of code for me to write. :) That said, the key is to remember the AI. The AI already knows how to swarm a city and surround it with units, so adding the attrition effect would benefit an already-used AI tactic. The AI does not, however, do a great job of rushing to defend a city (particularly if it is engaged in an attack of its own). If we take away a city's ability to defend itself (even if limited) when surrounded, the AI would have no recourse.

Also, if a city cannot attack when 'sieged' by a unit, what's the point of giving a city an attack at all?

G
 
So. It sounds like the following could be useful. To summarize so far

Simple changes.
1) Change defence buildings to give more HP and lower city combat strength, through some combination of means (this would include the palace as a means of increasing HP)
2) Get rid of indirect fire until artillery, which would nerf archers and to a lesser extent siege units. Make it require the tech.
3) Give cities a way to range to 3, preferably via infrastructure, but tech or policy if necessary.
4) Give siege units a ranged defence bonus (CEP did 50%, not just cover 1).
5) Consider doing CEP's unit stat style changes to siege units/ranged units with the siege promotion reduced but the unit stronger. Also archers with a city penalty and maybe swords (but not infantry) with a city bonus.

6) Siege attrition (from melee units? Horse units? Pillaging? Starvation?) Siege attrition reduced by defence buildings. This would take more work to implement.
 
Sounds good. Did any of those changes useLUA functions in CEP?

Regarding #6, it wouldn't be terribly difficult to implement if there's interest. I'd do it as a malus to a city's ability to heal based on the # of hostile melee units, including horses, in the six hexes around a city. If the malus from those units exceeds a city's standard heal, it'd just remove health instead. We could then add in a trait that increases the effective bonus by a % amount. Pretty simple, and I do think it would go a long way towards making melee units useful in more than just the coup de grâce role they have now.

G
 
Back
Top Bottom