Proposal workshop: unit stacking ideas

[usadefcon1]

A smaller front means a longer war. It's only not longer in this case because the superunit counts as 2 units of supply and 2 units of production.

I was under the impression that these were going to cost 3 units of supply. Come to think of it I'm not really sure where I got that impression but I think it's a good idea to limit them somehow.

 

[Rekk]

Does the melee line still get its innate +HP from promotions?

 

[pineappledan]

Does the melee line still get its innate +HP from promotions?

Yes

 

[Stalker0]

Unit carpet isn't hard to fend off. It requires a small battlefront (choke point), takes a very long time and is tedious. Eventually the AI just stops building buildings and economically falls behind, since unit cost ramps up and they aren't getting any yields from successful warring.

A superunit squad will just plow through your choke points, UNLESS you also fortify them with superunits, which will require you to have an even smaller battlefront. Overall no changes to war duration and tedium, just a reduction of maximum battlefront size for the humans to hold off an invasion.

If we were increasing the healing this would be true, but in theory armies will be doing more damage to each other with blitz but healing the same. So in the current scenario that unit in the citadel heals 25 health a round with his medic 2 buddy, and the enemy takes an extra 30 damage to his 100 health.

In the new scenario that unit still heals 25 health but might take near double the damage. meanwhile the enemy unit is still taking 30 damage from teh citadel but on 150 hp. So ultimately offense is favored here, so the classic human defensive positions will not be as good.

To PDANs point, perhaps this is exactly what we want in the long run, but bear in mind it would be a complete gamechanger. It might make any kind of defensive play obsolete, tall play would be obsolete, etc.

 

[pineappledan]

perhaps this is exactly what we want in the long run, but bear in mind it would be a complete gamechanger.

Well the idea of changing things is to change things, so yeah.

 

[ma_kuh]

I'm happy to challenge my original assumption. The first model I put together was deliberately simple to avoid a lot of the complicated maths, but I'll try and do a more realistic comparison. I'm going to continue to ignore most circumstantial CS modifiers, because that just feels like getting into the weeds.

Spoiler For all of these scenarios, I'll be assuming :

• The attackers and defenders have exactly equal base CS
• Terrain is providing +15% to defenders (open terrain bonus from Formation, or average between hills and forests/jungles)
• Flanking is providing +5% when "two units" are present
• In pure-defense situations, the unit is fully fortified (which I believe is still +50%, correct me if I'm wrong)
• Defenders heal 10 health each turn, and do not retreat (neutral terrain, or enemy terrain with Medic 1, however you want to see it)
• Attackers attack at full health
  • If the attacker is melee, their even-numbered attacks are at 60/100 health, after which they are rotated for a fresh unit
• The main comparison: Defenders are two generic melee units, or an army of the same with 200hp, or an army of the same with 150hp
• Every 1pt of damage applies -0.3 CS%
• The damage formulae are:
  • f(R)=((((R+3)/4)^{4})+1)/2 (for the defender, whose CS will always be higher thanks to Fortified)
  • g(R)= 1 / f(R) (for the attacker)
    • where R = the ratio of their CS
    • For example, the basic ratio is (10 * 1.65) / (10) +0.50 from Fortified, +0.15 from terrain
  • Attackers' CS:
    • Single, full health: 10
    • Flanking, full health: 10.5
    • Single, 40 damage: 8.8
    • Flanking, 40 damage: 9.3
  • Defenders' CS:
    • Single, full health: 16.5
    • Flanking, full health: 17
    • Single, 40 damage: 15.3
    • Flanking, 40 damage: 15.8
    • Single, 80 damage: 14.6
    • Flanking, 80 damage: 15.1
    • Single, 100 damage: 13.5 (for armies)
    • Flanking, 100 damage: 14
    • Single, 150 damage: 12
    • Flanking, 150 damage: 12.5

Spoiler Scenario 1: Two enemy spears attacking one target at a time :

Assumption: It's better to focus-fire one spear than to spread the damage between both bodies.
Assumption: Both attackers have flanking on all of their attacks, but the defender duo only has flanking against half of them. (Only one of the attackers is adjacent to the defender's buddy.)
Assumption: Both attackers have flanking against the army, and the army never flanks.

Two spears defending:

Defender HP (start of turn plus healing)	Damage dealt to defender	Defender HP at end of enemy turn
100 + 100	22 + 22	56 + 100
56+10 + 100	21 + 21	24 + 100
24+10 + 100	25 + 25	DEAD + 100
DEAD + 100	20 + 19	DEAD + 61
DEAD + 61+10	23 + 23	DEAD + 25
DEAD + 25+10	21 + 22	DEAD + DEAD

Survives 6 turns against two attackers.


One army defending:

Defender HP (start of turn plus healing)	Damage dealt to defender	Defender HP at end of enemy turn
200	22 + 23	155
155+10	21 + 22	122
122+10	25 + 26	81
81+10	24 + 25	42 (Army is dead with 150hp)
42+10	28 + 29	DEAD

Survives 5 turns against two attackers. (And would survive only 4 with 150hp, a single extra turn compared to one unit.)
The overkill damage not being wasted is a huge deal here!

Spoiler Scenario 2: Two enemy spears attacking two targets at the same time :

Assumption: Two defending spears are facing off without flanking bonuses.
Assumption: Both attackers have flanking against the army, and the army never flanks.

Two spears defending:

Defender HP (start of turn plus healing)	Damage dealt to defender	Defender HP at end of enemy turn
100 + 100	21 / 21	79 + 79
79+10 + 79+10	19 / 19	70 + 70
70+10 + 70+10	22 / 22	58 + 58
58+10 + 58+10	20 / 20	48 + 48
48+10 + 48+10	23 / 23	35 + 35
35+10 + 35 + 10	21 / 21	24 + 24
24+10 + 24+10	24 / 24	10 + 10
10+10 + 10+10	22 / 22	DEAD + DEAD

Survives 8 turns against two attackers.


One army defending:
Same as Scenario 1. Survives 5 turns.

Spoiler Scenario 3: An army attacking with Blitz :

Assumption: The attacker with Blitz will attack until taking more than 100 damage, then retreat.
Assumption: When the attacker retreats, the replacement army only attacks once in that turn.
Assumption: I'll use real-hp values for the attacker.
Assumption: The defender duo has flanking until one of them dies.

Two spears defending:

Defender HP (start of turn plus healing)	Damage dealt to defender	Damage dealt to attacker	Defender HP at end of enemy turn
100 + 100	21 + 19	41 + 48	60 + 100
60+10 + 100	15 (retreats) + 22 (newcomer)	58 (retreats) + 40 (newcomer)	33 + 100
33+10 + 100	21 + 18	44 + 49	4 + 100
4+10 + 100	(retreats) + 24 (newcomer)	(retreats) + 38 (newcomer)	DEAD + 100
DEAD + 100	19 + 16	48 + 56	DEAD + 65
DEAD + 65+10	(retreats) + 22 (newcomer)	(retreats) + 41 (newcomer)	DEAD + 53
DEAD + 53+10	20 + 18	45 + 51	DEAD + 25
DEAD + 25+10	(retreats) + 24 (newcomer)	(retreats) + 38 (newcomer)	DEAD + 11
DEAD + 11+10	22	41	DEAD + DEAD

Survives 9 turns against an army.
Having to cycle out an army and only getting a single hit on the swap really affects things.


One army defending:

Defender HP (start of turn plus healing)	Damage dealt to defender	Damage dealt to attacker	Defender HP at end of enemy turn
200	21 + 19	42 + 47	160
160+10	17 (retreats) + 23 (newcomer)	54 (retreats) + 39 (newcomer)	130
130+10	21 + 19	43 + 48	100
100+10	(retreats)+ 25 (newcomer)	(retreats)+ 37 (newcomer)	85
85+10	23 + 21	39 + 42	41 (Army is dead with 150hp)
41+10	(retreats) + 27 (newcomer)	(retreats) + 33 (newcomer)	24
24+10	35	26	DEAD

Survives 7 turns against an army.

Summary of results:

• Two attackers versus:
  • two defenders, focusing fire: defender survives 6 turns
  • an army with 200hp: defender survives 5 turns
  • an army with 150hp: defender survives 4 turns
• Two attackers versus:
  • two defenders, spreading the damage: defender survives 8 turns
  • an army with 200hp: defender survives 5 turns
  • an army with 150hp: defender survives 4 turns
• An army versus:
  • two defenders, focusing fire: defender survives 9 turns
  • an army with 200hp: defender survives 7 turns
  • an army with 150hp: defender survives 5 turns

My big takeaways:

• The 1pt damage => -0.3% CS really hampers armies. If we implement armies with extra health then the wounded penalty ought to be based on percentage of missing health, or armies should have a countervailing +0.15% CS per damage taken (a mini Japan passive) to offset it (possibly only on defense?).
  • Modeling with only -0.15% CS raised the army's expected turns to live in Scenario 1 from 5 to 7. I might investigate some theoretical values for this later.
• The army really needs the full 200hp to keep up with two individual units.
• As soon as an army needs to reposition (like cycling out after getting low), you start to lose damage in a big way. Two individual units can swap and still get two attacks in; armies only get one.
  • This to me is great emergent gameplay because it means that having a handful of "vanguard" units to follow up an attack or otherwise reposition is still a useful component of your military. Having all of your melee supply in armies means you'll be slower to push forward.
  • Also, Mobility goes up in value with armies.
• Armies really only shine in the specific case we're targeting: late game when the land is full.

 

[CppMaster]

If an army have to have 200 hp to make it worth, then how can this be?

In comparison, in Civ 6 combined units are about 25% stronger and have 75% higher maintenance cost, but usually it's still worth it.

 

[pineappledan]

Also post pikes we’re not talking about two 100 hp units merging into one 200 hp unit. The discipline/fieldworks/etc promos exist still and it doesn’t look like we’re talking about removing those.
Pike army: 2x100 HP => 1x200 HP
Tercio army: 2x110 HP => 1x210 HP
Fusilier army: 2x115 HP => 1x215 HP
Rifleman army: 2x120 HP => 1x220 HP
Infantry army: 2x125 HP => 1x225 HP
Mech Infantry army: 2x125 HP => 1x225 HP

 

[ma_kuh]

Civ 6's damage formula is quite different, I don't think +25% CS is the same kind of thing. Also, the movement in that game is scuffed and you can't use your last movement point to enter difficult terrain, meaning relevant choke points are more prevalent and stacking power in single cells would matter more.

On top of that, it's using a CS modifier rather than an HP modifier to achieve the power difference. The point of just stacking the HP and giving a double-attack is that by and large, the damage considerations will remain somewhat similar. We aren't trying to make armies more useful, we're trying to concentrate units into stacks for ease of movement. The biggest problem I actually see with 200hp+Blitz is that you still need to issue two attack commands each turn, which isn't much of an improvement over having two units attack. But at least melees are sitting still soaking damage while Fortified most of the time?

Also this post seems to suggest that the army vs. multiple units calculation is basically the same things we're talking about here, I don't think they're saying "usually worth it".

 

[pineappledan]

I can see this being a massive buff to Zulu, the French Musketeer, and the Danish Berserker. The ability for armies to move once and attack twice is huge. Mobility promo could end up being the premium choice for armies.

 

[Anarcomu]

Changing blitz promotion for an increase to damage dealt and taken on offense would solve the problem, remove one more click, and reduce the insane xp gain from double attack...

 

[Tekamthi]

concept for discussion: when army drops below 50% hp, it reverts to normal unit

this provides opponent some means of 'popping' the armies and mitigating any advantage they confer, and also increases their effect at resolving the problem of too many units in late game, as player will not just create army once and forever have it til death, but perhaps have to re-merge units multiple times...

 

[azum4roll]

concept for discussion: when army drops below 50% hp, it reverts to normal unit

this provides opponent some means of 'popping' the armies and mitigating any advantage they confer, and also increases their effect at resolving the problem of too many units in late game, as player will not just create army once and forever have it til death, but perhaps have to re-merge units multiple times...

So the defending unit would stay on spot forever while you keep supplying fresh units onto it.

 

[Tekamthi]

fair point, though this could mitigated by a cooldown on the resulting unit... recall as well that back a few pages ago we conceptualized that these would require full HP to create, so resulting unit would likely not be eligible for at least a turn as our theorycraft stands (ie if it converted back at less than 50% rather than less than or equal to 50%...)

On other hand may just be unnecessary complexity

 

[usadefcon1]

My concern is that the AI will be able to build more of these than me, and that I'll be able to hurt them but not kill them. Then they run off and heal.

I still think a fixed +50HP is the way to go.

 

[Phosphoraptor]

Any ideas on UI to make armies distinguishable from units at a glance? This is needed to actually reduce late game war tedium.

 

[CppMaster]

Any ideas on UI to make armies distinguishable from units at a glance? This is needed to actually reduce late game war tedium.

Maybe a star on unit icon. Or a gold border of the circle or sth.

 

[CppMaster]

Also this post seems to suggest that the army vs. multiple units calculation is basically the same things we're talking about here, I don't think they're saying "usually worth it".

Overall good points, but I think in that reddit post and some others, they do say that. Maybe not in 80% cases it's worth it, but like in 50-65% maybe? Anyway, it still means that just 25% more damage is reasonable and not out of a question, because it's too weak.

 

[pineappledan]

Could just reduce the number of units on a tile and increase the member numbers for armies

 

[CppMaster]

Could just reduce the number of units on a tile and increase the member numbers for armies

Do you mean the 3D model? That wouldn't be enough, I don't even look at it.