Stacked vs Single Unit Combat - The Battle Continues

[ThePlagueRat]

Originally posted by wrylachlan
If a fast enemy has say 10 stacks all within range of your city, instead of being able to concentrate all your units in the city you now have 1 in the city and a bunch of defensive stacks surrounding. This is actually MORE micromanagement.

That reminds me of a game where I got killed by a rather small stack. He had several stacks and combined them to one as he put them in between my borders within range of 2 cities. So to confuse me. Also he sacrificed one unit to pillage a road between them. These 2 cities where placed so it took one turn to move defenders from one to another, so when the road was cut, I could'nt. My city closest to the border had a huge stack of defenders and the other had less. So he attacked the innermost city with only a few defenders. Since it was an elimination game, I got killed.

These confusion strategies is IMO the good thing about stack movement, but it sure is more micromanagement. (I got nothing against that though) But in turnless mode, it's about being the quickest... IMO that is the downside of it. Think about 2 players with each stack of 50 cavalry and both see eachother at the same turn. The slow gets it, the fast moves on... Also I have seen players moving a few jaguars and such pillagers in towards my poor defended capital past the outer ring of cities in a end-turn double move making them attack before I can organize defence, but I have been quick and perhaps lucky on those occasions.

[patcon]

I don't think I'd like being forced to use stacked combat, but I think I'd like having the option to combine stacked combat with single unit combat.

example: I have 12 units available to attack a tile with 4 defenders with a 50% chance of winning any 1-on-1 battle. In single unit combat, I might get lucky and only need to use 4 attackers to kill all 4 defenders (leaving 8 units to do other things), then again I might need to use all 12 and still not kill all 4 defenders, but on average I would need to use 8 attackers to kill the 4 defenders. I wouldn't mind the option of saying "I need to use at least 4 attackers to do the job, so let's commit 4 right now and see what happens with them and hit 'em with more later if they didn't get the job done". I wouldn't mind the option of saying "Let's hit 'em with 6 units (especially if massed attack gained some benefit), even if I might be wasting the movement of 2 units, so I can save the tedium and exertion of extra mouse clicks". But I will not tolerate being told "You only get one chance to attack these defenders this turn, so how many attackers do you want to commit?" or "These 12 units joined together last turn to attack 8 defenders, so now they must stay together even though this turn they're attacking a single unit."

What I think is feasible is group attack options similar to the group movement options currently available. "Attack that tile with all units of this type from this tile" or "Attack that tile with all units from this tile." But I seriously doubt I'd use those options except in huge battles where I could be reasonably sure of not wasting unit movement on unnecessary attacks, unless there was some combat advantage gained by committing excess units or disadvantage for using a piecemeal attack.

What would be cool would be a method by which an attacker could commit units to a simultaneous attack on a certain tile in order to gain a combat advantage.

Say a tile defended by 4 units is attacked from a single tile by 6 units - the concept of massed firepower should give a boost to the attackers. If, on the other hand, these 4 defenders are attacked by only 2 units, the defenders should receive the benefit of massed firepower. Call it he Massed Firepower Combat Advantage (MFCA). The effect could be as simple to calculate as
MFCA = (# attackers - # defenders) / (# attackers + # defenders)
effective attack strength = total attack strength * (1 + MFCA)
The reason for dividing by (# attackers + # defenders) rather than just (# defenders) is to keep the modifier from getting out of hand and to simulate the problem of coordinating huge numbers of attackers.
So a 6-on-4 attack would gain a MFCA of 20% {(6-4)/(6+4)}, while a 2-on-4 attack would suffer a MFCA of -33% {(2-4)/(2+4)}. An even 4-on-4 attack would have a MFCA of 0%. Yes, the arithmetic might get a little messy - 6 warriors attacking 4 unfortified warriors on grasslands would have combat odds of 7.2 to 4 - but computers can handle messy arithmetic.

This idea could even be extended to include the combat advantage of flanking attacks from multiple directions. A single tile with 10 defenders is simultaneously (that's important) attacked by 12 units from 3 different tiles as 6 units from tile A, 4 units from tile B, and 2 units from tile C. The defender must spread the 10 units to cover the 3 fronts to match the threat as well as possible. I'm not quite sure how to implement this- there are lots of different ways (ideas?), but I'm leaning toward a defensive distribution which minimizes the MFCA-modified attack strength on the various attacking fronts starting with the strongest attacking front (as a defender you would need to try to stop a breakthrough on any front). This would involve massive number crunching until some reasonable algorithm could be established, but computers are good at that sort of thing. The thing to be careful of is a "front factor" which gives too much advantage to a sham flank attack with a single unit which creates a "second front" but doesn't really divert the defender's attention from the main point of attack - 8 defenders wouldn't necessarily split 4-4 to cover two fronts if the attacking forces were split 9-1.

The attack would be resolved as if all the attackers were a single unit with a single combined attack strength (modified as appropriate by the combat advantage formula and flanking calculations) and a single combined hit point total and all the defenders were a single unit with a single combined defense strength (including any terrain/fortified effects) and a single combined hit point total. At the end of the combat, the side with hit points remaining would have those points distributed 1-at-a-time back to the units (strongest units first) with any units not having hit points being killed - so a "winning" defending stack of 8 units which has been reduced to 5 hit points would lose its 3 weakest units and the others would have 1 hp left each.

There's probably more I need to consider, but it's late and I hear my pillow calling my name.

[Grandpa Troll]

I too would enjoy the "option" of stacked combat

just my desire is all

Gramps

[hexagonian]

Originally posted by patcon
I don't think I'd like being forced to use stacked combat, but I think I'd like having the option to combine stacked combat with single unit combat.

Basically what I proposed here a while back... - with a pic of the rough layout.

[GePap]

I reread you proposal, and it looks good, thought I was still somewhat confused.

BUt as i said in that previous thread, combat by armies is what should happen in a game of the scale of civ.

[Boris Godunov]

I'm with lotm that the Imperialism combat model is the best way to go. Tactical battles (whether realtime or turn-based) are sorely needed.

[hexagonian]

Originally posted by GePap
I reread you proposal, and it looks good, thought I was still somewhat confused.

The basic premise allows a player to either send in his units one at a time into battle (as currently in civ3) or commit up to 24 units into a battle and then gain some bonuses to be determined by the designers (either HP, attack). The defender's stats would also be increased based on how many units it has - (and add this to my post) the defender would only bring into the battle a random number that is above the number of units that the attacker brings in.So an attacker could not simply see how many units were defending by first bringing in a lone unit.

Bottom line - you can choose to fight one at a time, but you could very well put yourself at a disadvantage. You may also bring in your full 24 units and find out that you had overkill. But that's the price you pay.

This system would also feature a lock-stack ability that can help in unit management and woulld also be workable for a worker setup in regards to movement and tasks (locked group worker stacks would be more efficient over and above the same number of single workers doing the same thing.

[wrylachlan]

Originally posted by hexagonian

The basic premise allows a player to either send in his units one at a time into battle (as currently in civ3) or commit up to 24 units into a battle and then gain some bonuses to be determined by the designers (either HP, attack). The defender's stats would also be increased based on how many units it has - (and add this to my post) the defender would only bring into the battle a random number that is above the number of units that the attacker brings in.So an attacker could not simply see how many units were defending by first bringing in a lone unit.

Bottom line - you can choose to fight one at a time, but you could very well put yourself at a disadvantage. You may also bring in your full 24 units and find out that you had overkill. But that's the price you pay.

This system would also feature a lock-stack ability that can help in unit management and woulld also be workable for a worker setup in regards to movement and tasks (locked group worker stacks would be more efficient over and above the same number of single workers doing the same thing.

As someone who generally doesn't like stacked combat, I have to say that's not bad. A couple of comments:

1)If stacked combat is implemented I think we can do away with the "Army" mechanism. It seems to be a duplication of concept. But at the same time, I like the idea of "Great Military Leaders". Maybe the leader just stays as a single unit which you can add to an attacking stack for a bonus. But if the stack is killed, the leader dies too (or is captured and can be ransomed...)

2)Doesn't the concept of "Reinforcements" overcomplicate the distribution of troops? Wouldn't it be easier to have 2 lines of "melee slots" and 1 line of "ranged slots"?

3)It doesn't make sense to me that your archers would move forward into the melee line if the defending melee unit was killed. Rather the ranged should stay put and take shots at the enemy melee as they advance on the ranged units.

An interesting side effect of this is that assuming 2 lines of melee before teh ranged units, it could take 2 "rounds" for an enemy to engage the ranged units, BUT a faster melee unit could close faster, allowing the ranged units less pot shots. This provides a natural advantage to faster units.

4)I also don't think that the idea of each unit only moving vertically through its file is realistic, or necessarily a good game dynamic. I mean if an attacker is wearing down a defender, the two defenders on either side are going to compensate and try to help out.

My piggyback idea: Each line has a "Max slots" and "Min slots" say 6-10 units. If the enemy has 10 units in their front line, and you have 10, you go at it 1 on 1 with the guy in front of you. If he kills off 4 of your guys, your line spreads out so that you are now 6 across evenly spaced versus his ten. The enemy now has the advantage as 2 melee attackers can now simultaneously attack 1 of your melee defenders BUT the enemy cannot yet advance towards the ranged units.

Once your melee line is down to less than the Min slots (five units) then the enemy can push forwards towards the ranged units.

5)What function does bombard/air support play in this combat?

6)How does terrain play into the mechanics? Personally I think that terrain should play a bigger role in the tactics of war in Civ.

[hexagonian]

Sidenote: this is a combo of what I see as the best ideas from both CTP and civ...The main problem with the CTP system for many players is the stack cap, but there is precident for some sort of stack limits in a host of games (CTP, AOW1/AOW2, MTW as well and RTW, to mention a few...). And my setup can work for infinite stacks - it rewards a player to streamline his combat by running multiple-unit battles too.

Originally posted by wrylachlan
1)If stacked combat is implemented I think we can do away with the "Army" mechanism. It seems to be a duplication of concept. But at the same time, I like the idea of "Great Military Leaders". Maybe the leader just stays as a single unit which you can add to an attacking stack for a bonus. But if the stack is killed, the leader dies too (or is captured and can be ransomed...)

This will work for me - and I've done this exact thing in a CTP2 mod.

Originally posted by wrylachlan
2)Doesn't the concept of "Reinforcements" overcomplicate the distribution of troops? Wouldn't it be easier to have 2 lines of "melee slots" and 1 line of "ranged slots"?

That can work too - the basic idea is to have ranged units firing on enemy lines under the protection of melees. Eliminate the melees and they are vunerable. So it does not matter how many rows of melees the designer chooses to set up.

It works for CTP - it works for the Total War series - why not civ???

Originally posted by wrylachlan
3)It doesn't make sense to me that your archers would move forward into the melee line if the defending melee unit was killed. Rather the ranged should stay put and take shots at the enemy melee as they advance on the ranged units.

Units have 3 numbers (attack/defend/range)
Ranged fire upon melees and inflict damage without taking damage themselves, until the melees are gone. You can make the range number weaker than the normal attack and defend number
The ranged units still fire upon the the enemy melees when their cover is gone, but since the range number will be lower than the opposing unit's normal attack/defend number, they will disintergrate quickly...

Originally posted by wrylachlan
My piggyback idea: Each line has a "Max slots" and "Min slots" say 6-10 units. If the enemy has 10 units in their front line, and you have 10, you go at it 1 on 1 with the guy in front of you. If he kills off 4 of your guys, your line spreads out so that you are now 6 across evenly spaced versus his ten. The enemy now has the advantage as 2 melee attackers can now simultaneously attack 1 of your melee defenders BUT the enemy cannot yet advance towards the ranged units.

Basically what happens in CTP2 with flanking...

Originally posted by wrylachlan
5)What function does bombard/air support play in this combat?

Same as in civ3 - choose to soften up an enemy with bombard/air strikes before atacking

Originally posted by wrylachlan
6)How does terrain play into the mechanics? Personally I think that terrain should play a bigger role in the tactics of war in Civ.

Any way you want it to work...

[wrylachlan]

Originally posted by hexagonian
Units have 3 numbers (attack/defend/range)
Ranged fire upon melees and inflict damage without taking damage themselves, until the melees are gone. You can make the range number weaker than the normal attack and defend number
The ranged units still fire upon the the enemy melees when their cover is gone, but since the range number will be lower than the opposing unit's normal attack/defend number, they will disintergrate quickly...

I'm not sure if you understood my original post... I was responding to this from your discription of the system:

When melee units are destroyed, then the ranged units come up on the front line - after that, the reinforcements come in play, taking the place of those defeated units. (or ranged units do not come up on the front lines until all reinforcements are destroyed - this is up to the designer)

Range units - both attacker and defenders - inflict damage on frontline melee units without taking hits themselves until they end up on the frontline - but they have very weak attack/defend, so they will get crushed by a conventional melee unit.

You have the ranged units "coming up to the front line" which doesn't make a whole lot of sense to me. I would think the ranged units would stay put and the enemy melee would come to them.

Just for clarification, in your system do the ranged units advance piecemeal or all at once? If I'm a melee unit on the front line and I kill the melee unit in front of me does a ranged unit drop in to take his place, or is it not until all the melee units are killed that all the ranged units drop in simultaneously?

[Grandpa Troll]

as for the issue of terrain i like that conceptr wherasthe defended unit is at a bonus in a mountain or hill versus flatland

i love chokepoints and they should be a bonus for a unit defending such

[lord of the mark]

Originally posted by Boris Godunov
I'm with lotm that the Imperialism combat model is the best way to go. Tactical battles (whether realtime or turn-based) are sorely needed.

wow, id forgotten id posted that, my how time flies.

Ive since played Moo2, which has a widely admired turnbased tactical combat approach, but one which left me longing for the Civ2 approach. OTOH that tends to run to homogeneous fleets, and it seems one of the benefits of stacking could be more realistic combined arms. The Imps do better in that regard.

[korn469]

I like the army concept up to a point simply because it limits how many units can attack in one stack and it also forces the player to invest in having stacked combat abilities which I think is a good thing. Though I do think that the normal army stats in civ3 need tweaking.

However if I was going to change the way stacked combat happens in Civ4 here's the only change I'd do: make all units in a stack attack simultaneously. I'm kind of tired, and I'm not completely positive about the math, but I think it would work.

For example, 4 arches attacked a tank, with each having an oppertunity to attack every turn as long as they remained alive. It seems like they would have a far better chance at winning, than if they attacked one at a time like they currently do.

In addition to that change, I'd love to see a system of units and counters instead of plain attack.defense.movement stats like we currently have. That is the only way to really encourage combined arms in my opinion.

[hexagonian]

Originally posted by wrylachlan
I'm not sure if you understood my original post... I was responding to this from your discription of the system:
You have the ranged units "coming up to the front line" which doesn't make a whole lot of sense to me. I would think the ranged units would stay put and the enemy melee would come to them.

There has to be some sort of abstraction in any computer gaming system.

My system allows a bonus when ranged units are protected by melees. Over time, this 'free shot' can be a powerful weapon too.

The general concept is that up close, ranged units lose their effectiveness (and this is demonstrated in the CTP, MTW and RTW battle systems... Engage the ranged units up close and they do rout quickly)

In a totally realistic setup, the ranged units would get some more 'free shots' as the enemy advanced on them, but this is more of a RTS gaming element. It can be set up in a TBS setup as an additional round whereby the ranged unit does get a last 'free shot' before the enemy engages the ranged units.

But I see this as somewhat unnecessary for gaming aspects. The idea is once the cover is eliminated, the ranged units are weak. But as I said, the free round of battle with a 'last shot' can be used for my system - its a minor point in the overall picture.

As for army composition, this should not hamper the AI because the designers could choose to impliment an infinite or very high unit count on a tile setup. The AI would then be able to fill its slots in a battle situation as it normally does - choosing the most powerful units and automatically designating them for battle. Since there would also be a limit for how many melees/ranged go into a single battle in my system, the AI (and the defending player) would automatically fill the slots with the types of units required, if it has them. The slotting system might call for 15 melees - and so the computer would fill those slots if it has the units. If it does not, it may have to use ranged units as melees. So be it then...

Originally posted by wrylachlan
Just for clarification, in your system do the ranged units advance piecemeal or all at once? If I'm a melee unit on the front line and I kill the melee unit in front of me does a ranged unit drop in to take his place, or is it not until all the melee units are killed that all the ranged units drop in simultaneously?

It can be either way - determined by the designer.

You could have a setup whereby the front row has to always have a set amount of units at all times - if the units are available (and the type is not important) The front row could be hard-capped at 10 units. If you have no more melees, then the ranged units automatically get moved up on the front lines.

Or it can be set up so all melees need to be used up first, and then the idea of a flanking bonus becomes more important (see below) - though in my first setup, there can be a flanking unit type that is slotted on the wings. If you do not have a flanker type of unit, that slot is empty. But if the enemy has a flanker-type, he gets a free shot against your end unit.

Melees could be set up in a two row system. (first row in battle, second row in reserve to be called into battle when the units on the front line start dying)

Kill a melee on the first row and a melee from the second row takes its place. If you manage to maintain a larger front row of melees (you are at a point when your front row has 5 units and the enemy only has 4 - and this would happen once all the reserves get used up) then the end units on your army get a 'free shot' against the corner enemy unit.

EXAMPLE

r - ranged
x - your melees
v - enemy melees

rrr
xxx(x)(x)
--------------------
vvvV
rr

both (x)s gets a shot against the last 'V' unit. 'V' only gets to defend against one of the (x)s

And if all the melees are all eliminated, the the ranged units move up and become roadkill.



Hope this clarifies things - but I'm under the impression that civ4 might be at a point in time in its development where a revision of the battle system might not be doable. The question is just what the developers have in mind.

I really hope the developers have looked at the poll on this tread, because it is overwhelming in favor of a complete overhaul of the civ-civ3 system.

I'll say this too - if it remains the micromanagement nightmare of past civ games, and if it is a single-unit vs single-unit setup without any thought for combined arms, I won't be playing it.

Not that it'll matter to sales all that much

[Grandpa Troll]

I stand firm that while a single unit represents a "compnay" or Battalion" on units...still there should be stack option..when you concentrate firepowerin a group at once yes indeed it makes a difference

I also love the idea of retreat when you attack and seem to be outnumbered then yes retreat

Gramps