CIV 7: A Return to Unit Stacking (Limited stacking revamped system)

[AntSou]

First of all: I do not think the system of stacked units is better than the one used in Civ 5 or Civ 6. These are different systems with their own pros and cons. I do believe, however, that Unit Stacking has a lot of potential which certainly did not reach full maturity in Civ 4.

Purpose of this thread: To survey support and discuss a revamped system of Limited Stacking years in advance of a new civilisation title. I propose some ideas below.


I would like to start with, what I believe, are some of the shortcomings of the current system:

1 - Everything looks cluttered, particularly on smaller maps. This issue has been made even more evident in Civ 6.

2 - AI can't quite handle it. The constant mass of troops moving around aimlessly looks ugly enough, but the manner in which Civ 5 and Civ 6 handle stacking (or lack of it) means an extra layer of complexity to combat which AI often struggles with.

3 - Basically, it is already present in Civilization 5, which was a great title. But Civ 5 did not have districts. I believe Civ 7 should retain the districts from Civ 6, while attempting a limited unit stacking system.
---


The core of a revamped unit stacking system is this: limited stacking, linked to policies, tech or ideas. Developments in ideas/tech permit unlocking Preset formations and/or Command Limits (as used in Stellaris) which are also linked to Land Attrition and Force Supply.

Battalions / Command Limit: At start of game, civs start with a command limit of, for example, three units. A Command Limit is simply the amount of military units that can be stacked and grouped together as a battalion in a single tile. Units stacked into battalions have advantages depending on the formations that have been researched or which are currently part of your civs policy (A bit like formations in Through the Ages).

Command Limits can be increased as the game advances. Generals lead battalions and give bonuses to the battalions they command.

Preset Formations: Combat style/war is intrinsically linked to culture. In previous Civs this was expressed in the form of unique units. This principle could be expanded upon by creating Military Tactics/Formations which actually require to be researched/developed to take full advantage of them.

Example: Two civs may have reached Tank tech, yet one of them employs Blitzkrieg as one of its Military Tactics. This gives it bonuses to movement and attack in Tank only battalions. The other Civ, who controls a large contingent of Cannons, has not yet developed Blitzkrieg. Even if it did, it couldn't really take advantage of it. Instead, this Civ employs, as one of its Military Tactics, Fixed Battery, which gives bonuses to artillery fortified in that civ's cities or forts.

The amount of Preset Formations any Civ may have is also capped. (E.g. Two formations at start of game). This cap can be increased somewhat as game advances.

Preset Formations, unlike Military Techs, are mainly connected to culture rather than science.

Land Attrition: Certain land types set unit caps (Think EU4). The size of any battalion in any given tile is only limited by that civ's command limit. However, the size at which a battalion receives attrition is determined by other factors, mainly land.

Example: Your battalion is on an enemy Tundra tile. Tundra tiles allow two military units to stack on it without receiving attrition. Your Command Limit is 6. You have a battalion of 5 units on that territory. 3 of those units receive attrition every turn. Each additional unit beyond the land limit has an exponential increase to its attrition. so the 3rd unit above the limit takes more attrition damaged than the 1st unit.

Force Supply: A force supply is, essentially, a 'trade route' between your battalions and your cities. These can be pillaged by enemy forces, like a trade route, breaking the supply. A Force Supply diminishes the effect of attrition. Like trade routes in Civ 5, it also has a range, beyond which its effects are reduced/null. Like trade routes, their number can be increased and their ranges expanded through tech (E.g. Canned Food Tech).
---------------------

TLDR: I believe Civilisation 7 could try a system of Limited Stacking + Districts, which would allow the best of both worlds. Terrain could still be used for its tactical advantages, without the issues that one-unit-per-tile cause, while on the other hand avoiding the issues of unlimited stacking.

It would also reduce cluttering on the map, and make combat easier for the AI to handle.

 

[Boris Gudenuf]

There were several Threads and GOK how many posts regarding Stacking and Stacking Systems when Civ V's 1UPT first came out, so it's not surprising that it comes up again. A lot f what I'm going to say here is based on posts made a while ago, because I think they still apply to the Stacking Problem.

There are several things to consider before wading into the Stacking Swamp:
1. Any system of 'limited' Stacking means you are introducing a bunch of Stacking Rules, Limitations, Special Cases that will Complicate the Game. There had better be some durned good reasons for Gamers to put up with those complications.
2. The best thing about the current 1UPT system is the Tactical 'Feel' it gives battles: the interaction between Ranged and Melee, Flanking and Support bonuses, etc. Any stacking system has to maintain that, or it will be a bland and uninteresting (and disliked) replacement.
3. IF you are going to introduce some kind of Tactical Limitations/Rules for Stacking, for once it would be nice if the game actually got the historical terminology right. Your examples are of Inaccurate Terminology, so excuse me if I address them even though the corrections have nothing directly to do with Stacking Systems.

First of all: I do not think the system of stacked units is better than the one used in Civ 5 or Civ 6. These are different systems with their own pros and cons. I do believe, however, that Unit Stacking has a lot of potential which certainly did not reach full maturity in Civ 4.

Civ 4 had (virtually) Unlimited Stacking, and nothing to differentiate the various types of units in the stack: it had the advantage of being really simple, really bland, and really easy to Abuse. Good Riddance to it.

The core of a revamped unit stacking system is this: limited stacking, linked to policies, tech or ideas. Developments in ideas/tech permit unlocking Preset formations and/or Command Limits (as used in Stellaris) which are also linked to Land Attrition and Force Supply.

Battalions / Command Limit: At start of game, civs start with a command limit of, for example, three units. A Command Limit is simply the amount of military units that can be stacked and grouped together as a battalion in a single tile. Units stacked into battalions have advantages depending on the formations that have been researched or which are currently part of your civs policy (A bit like formations in Through the Ages).

Command Limits can be increased as the game advances. Generals lead battalions and give bonuses to the battalions they command.

First of all, a "battalion" is a very specific unit in modern military terminology since the Renaissance, and it derives from the Medieval "Battle" which was simply 1/4 to 1/3 of the entire army, with no set limit in size. So, it's either too specific or not specific enough for Game Use. Also, a modern battalion is always a single type of troops: infantry, artillery, armor, etc. So there is no provision in this terminology for a stack containing a mixed group, making it no different from Civ VI's very limited Corps/Army units.

I suggest that, like the size of cities in the game, we use a 'sliding scale' of Unit Sizes: Ancient Armies had no 'fixed' units larger than about 1000 men (very close to a modern 'battalion' of 700 - 1000) - the Egyptian Pedjet, Sumer Sanga. In the cases we have record of, the meaning is "Host". These Hosts could consist of several smaller units (Egyptian Sa, Sumer Nu-Banda) similar to modern Companies.
I suggest, therefore, that the first Stack of multiple Units be a Host which can be of all one type (like the Egyptian chariot Pedjet of 250 chariots) or multiple types of troops.

The historical limitations on how many troops could be concentrated ('stacked') were Terrain, Supply, and Organization. Very few Ancient or Classical armies had any organization higher than the 1000 or so men in the 'Host', which meant that forces even as large as a modern Division (15 - 25,000 men) were almost impossible to control (stack). Since even a few thousand men and their animals would eat as much as a small city of the time, they were also very difficult to supply for any length of time, so the availability of food and fodder from the terrain (tile) was also hugely important. The biggest technological changes in Supply (stacking) are in supply technology - better roads, better transport vehicles, riverine/coastal boat supply, more efficient bureaucracies to control supplies. Ability to survive and fight in extreme terrains depended until near-Modern eras on size of the force and native technologies from people familiar with the terrain - there's a reason why almost every Civ hired people who lived in the mountains to fight in the mountains, or natives of the rainforest to fight in the rainforest.

So, an obvious 'basic' Stacking limitation might be By Era: in the Ancient Era, you can basically stack 3 units, but, say, only 2 in rainforest, tundra, snow, or desert. Classical Era = 4 basic units/stack, and so on.
Being next to a river or road that leads to a friendly city would increase the stacking limit, but not by much before the 'movement/transport' technology improves dramatically.

This sort of system has the advantage of being a largely invisible complication: stacking limit rules are few and related directly to the Era you are in so that for the most part the Gamer does not have to constantly mentally refer to a bunch of rules and restrictions.

Preset Formations: Combat style/war is intrinsically linked to culture. In previous Civs this was expressed in the form of unique units. This principle could be expanded upon by creating Military Tactics/Formations which actually require to be researched/developed to take full advantage of them.

Example: Two civs may have reached Tank tech, yet one of them employs Blitzkrieg as one of its Military Tactics. This gives it bonuses to movement and attack in Tank only battalions. The other Civ, who controls a large contingent of Cannons, has not yet developed Blitzkrieg. Even if it did, it couldn't really take advantage of it. Instead, this Civ employs, as one of its Military Tactics, Fixed Battery, which gives bonuses to artillery fortified in that civ's cities or forts.

The amount of Preset Formations any Civ may have is also capped. (E.g. Two formations at start of game). This cap can be increased somewhat as game advances.

Preset Formations, unlike Military Techs, are mainly connected to culture rather than science.

Glad you used 'blitzkrieg' as an example, since it is an example of a Pop History term that is meaningless. Rapid advance and attack has been present in warfare since Tomyris or Attila charged out of the Steppe and then disappeared before anyone could react. The specific German motorized all-arms tactical and operational technique labeled 'blitzkrieg' (by journalists, NEVER by the German military) was in fact Bewegungskrieg ("Maneuver Warfare") and in form dates back to the Great Elector of Brandenburg of the 1680s (the Fehrbellin campaign, go look it up), modernized by adding motor vehicles. The tactical specific is to always use combined arms, so a 'tank only' stack, by definition, is not a 'blitzkrieg' of any kind.
What you are really talking about is a formation that is a Panzer Division (US Armored Division, Soviet Tank Corps, etc.) the tactical/operational Building Block of Modern/Atomic Era Maneuver Warfare. These Formations would all consist of a combination of Units. Historically, they were Tank, Infantry, Artillery, Combat Engineers, Scouts, Antitank and Antiaircraft units. That's 'way too many for the game, so I suggest that the in-game Panzer Division consist of a maximum of 3 Units: 1 Tank, 1 Infantry, 1 Artillery, all moving at the speed of tanks (motorized infantry and artillery) .

Which, however, brings up the other Big Problem of stacking: how do you include the Tactical interaction of Ranged and Melee factors, Flanking, Support, and Terrain bonuses that the current 1UPT system has now?
I suggest that a Combined Arms Stack or stack of Formations reduces any Malus from enemy forces or terrain to that of the most advantageous unit in the stack. That is, a city center or district tile represents a built up area, so has a defensive bonus against any type of Unit, but has a Major defensive bonus against Heavy or Light Mounted Units, which includes Tanks. A Unit or Stack of Unsupported tanks attacking a unit in a District will see its melee factors reduced by at least half. But if it is stacked with infantry, or in a Formation that includes infantry, the malus is that of the infantry for all melee units in the stack/formation and the tanks' melee factor can be added to the total with very little reduction. The ranged factor of any Artillery/ranged units in the stack would be applied as Support Fire before the stack (or before calculating defensive factors if on the defense).

That would make Combined Arms Formations very powerful, as they were - but also expensive, since the Panzer Division and its equivalents required that Artillery and Infantry units included in it be motorized to the same degree as the tanks, with all the resulting requirements in Production and Maintenance Costs and Resource Costs. For example, one refill of fuel for a Panzer Division was calculated as 220 cubic meters of fuel, for a German Infantry Division it was 6 to 9 cubic meters - a huge difference in Supply/Support requirements between the 'mobile' panzers and the 'marching' Infantry.

Land Attrition: Certain land types set unit caps (Think EU4). The size of any battalion in any given tile is only limited by that civ's command limit. However, the size at which a battalion receives attrition is determined by other factors, mainly land.

Example: Your battalion is on an enemy Tundra tile. Tundra tiles allow two military units to stack on it without receiving attrition. Your Command Limit is 6. You have a battalion of 5 units on that territory. 3 of those units receive attrition every turn. Each additional unit beyond the land limit has an exponential increase to its attrition. so the 3rd unit above the limit takes more attrition damaged than the 1st unit.

Force Supply: A force supply is, essentially, a 'trade route' between your battalions and your cities. These can be pillaged by enemy forces, like a trade route, breaking the supply. A Force Supply diminishes the effect of attrition. Like trade routes in Civ 5, it also has a range, beyond which its effects are reduced/null. Like trade routes, their number can be increased and their ranges expanded through tech (E.g. Canned Food Tech).

Strictly limiting the types and numbers of Units that can survive/move/right in certain types of terrain and climate would extend Exploration into the late game, which would be a Good Thing.

The types of terrain that were historically Deadly to unequipped, unacclimatized armies and units were Tundra, Snow, Desert, Rainforest, Marsh/Floodplains, and Forest, in varying degrees. Forest largely made certain types of units harder to use tactically: Heavy Cavalry in dense woods is pretty Counter-Intuitive, for instance. The other types of terrain killed men and animals, either because they were associated with disease (Marsh/Floodplains, Rainforest) or because there was nothing to eat and killing Climate associated with them (Tundra, Snow, Desert).
So I would put a strict Stacking Limit on Tundra, Snow, and Desert terrain tiles from the start of the game, and perhaps, except for certain Civs, an attritional factor for any unit in Snow, Tundra or Rainforest except Scouts. Unequipped, unacclimatized troops simply cannot survive there without technologies that are not available until later or available from the people living in that terrain/climate.
Stacking in Marsh/Floodplains or Rainforest tiles would also be dangerous: you can do it, but the more troops to put within range of the mosquitoes and tsetse flies and other 'natives' the more you will inevitably lose until technologies like Germ Theory or Antibiotics or Sanitation give you counters to the conditions.

I mentioned Supply already above. The thing to remember here is that for most of history, Supply = food and fodder for men and animals in the host/force/formation/army, and for most of history the armies provided most of their own supply by sweeping up everything edible from the country they moved through. That's another reason that desert. rainforest, tundra and snow tiles are dangerous: nothing to eat, or nothing that the non-native recognizes as edible.
Food and Fodder being bulky, heavy materials required in bulk (rough estimate for Modern/Atomic Era armies: 1 Ton of Food per day per 1000 men minimum, 5 - 10 tons of Fodder per day per 1000 horses, depending on how much they can 'graze' from the surrounding area) the only way to transport them until railroads was by water: river or coastal shipping. That means any change in supply for at least the first Eras of the game (until Industrial) is related to access to river/coast leading to a friendly/allied City as an Ultimate Source.
Troops can be supplied overland, but the distance is strictly limited: an animal-drawn conveyance, no matter how efficient, will consume the weight it is carrying in food for the draft animals after about 100 - 150 kilometers, which means beyond that you are effectively transporting 0 supplies.
So, a Stack is limited not only by Era (basic Technology/Organization) but also by how many can be fed in a given tile. A functioning Supply Line means any number can be fed. A tile that has been organized already (Improved) can be 'Foraged' to provide Supply for X Units for 1 Turn, but only until Gunpowder starts to require Supplies not available locally (gunpowder, ammunition, replacement weapons). Foraging would be like Pillaging, but more temporary (Tile 'recovers' automatically in 2 - 3 Turns) and automatic if the Stack doesn't have a functioning or adequate Supply Line. That means, as was true historically, that an army staying in one place without a supply line is In Trouble: don't try to besiege a city without a river or coastal Supply Line!

That also means that Supply For Armies changes dramatically with the advent of the Railroad: the railroad allows any amount of supplies to be provided from any city connected to the railroad over any type of terrain: strategy and operations will become 'railroad dependent' for both Food and the increasingly important Fuel and Ammunition required to make Industrial and later era units combat capable.
This, by the way, indirectly increases the importance of the Industrial Revolution and its consequences, something else the game is currently lacking.

Last comment: Supply lines can be cut or broken by raiding units, which would also affect attrition due to Stacking Limits. This could make Light Cavalry/Light Mounted units more important in some circumstances or air/sea attack an effective 'Supply/Stacking Modifier': an enemy Bireme or Galley syitting across your coastal Trade Route/Supply line could mean starvation for your Roman Legions trying to do an end run through the coastal desert!
Later, air attack can also interrupt railroad and sea Trade/Supply routes, but not as completely: historically, railroads could be repaired within hours or days even after major air strikes, and it took a very sustained, powerful air campaign to tulle interrupt supply, like the massive Allied air campaign lasting months before and after D-Day in northern France in 1944.

 

[Mik1984]

The point of one unit per tile is to make combat more battle front line oriented instead of clashes of stack of doom. AI can't handle it because the game needs Hearts of iron 4 style unit group management methods, like designing frontlines, drawing defensive lines, creating formations from units. Once these UI tools will be implemented for players to manage multiple units at once instead of giving orders to units one by one, the AI will be able to take advantage of these tools to manage its own armies.

 

[Krajzen]

The point of one unit per tile is to make combat more battle front line oriented instead of clashes of stack of doom. AI can't handle it because the game needs Hearts of iron 4 style unit group management methods, like designing frontlines, drawing defensive lines, creating formations from units. Once these UI tools will be implemented for players to manage multiple units at once instead of giving orders to units one by one, the AI will be able to take advantage of these tools to manage its own armies.

But... AI in HoI4 is not very good.

Personally I support limited stacking at this point, at any form. A long time will pass before technological breakthroughs will allow on implementing commercial video game AI handling 1UPT level of complexity. Also, Vox Populi is not good enough (and it took insane amount of time and effort). 1UPT in the way it is done in civ5/6 is absolute hell to program AI for and I have simply completely lost hope it will ever become decent (at the current technological stage). 1UPT makes AI nonissue, which trivializes war, which trivializes diplomacy, which trivializes everything.

1UPT promises to add tactical depth to strategy game, but it also annihilates AI so tactics are unnecessary anyway, and in the meantime strategic depth is heavily damaged too.

Other issues with 1UPT are:
- horribly cluttered map, both visually and logistically, constant traffic jam and pain in the **** involving harsh terrain movement
- a lot of time spent on micromanagement of dozens of units every turn
- worsened performance (which makes AI issue even worse, because it could suck less if it consumed more turn processing time which is impossible due to already high turn times)
- it is completely unrealistic before modern warfare


Just make, for example Three Unit Stacking with more relaxed movement rules and all units having the same range (but ranged ones taking no damage when attacking). Due to less units on the map, less painful movement, less traffic jams and less ability for human player to crazy micromanage 30 units AI already becomes better. What is lost from 1UPT is recovered by the entire new layer of unit composition planning (do I make units consisting of two pikemen one musketmen or vice versa etc). 3UPT would also reduce one of the greatest problems of AI, efficient city sieges - with three times less units but three times stronger ones (so quite often two would be enough to take a city) there would be far less traffic jam and gigantic pain (for AI) with maneuvering units around city sieges.

 

[AntSou]

I'm not a fan of HOI4 UI either.

What is your opinion regarding Endless Legend unstacking system, or the need to have generals in order to command stacked armies? It certainly helped decluttering tiles in Endless Legend, but combat isn't really the best part of that game, and you never get anywhere close the amount of military units used in civ games anyway. One positive thing about combat in Endless Legend, is that the game stops once a battle starts, and you play that battle out turn by turn before proceeding with the game.

A similar system for CIV, which more advanced PCs would finally be capable of handling, would be to have a fully fledged mini tile-set map for each tile on the map. (E.g. the tile-map itself being somewhat similar to tiles in Heroes of Might and Magic or Battle Brothers). Combat only starts once all of that civ's stacked armies on the campaign map move (a bit like combat engagement in Scythe). You could bring a cavalry only army with bonus to movement to attack from a flank, for instance. The game then zooms in on that campaign tile to begin combat, showing the combat map which is a bunch of more tiles with the same land features of the campaign tile. At any point during the game the player can see the combat map. Combat maps are permanently created at the beginning of the game, just like the campaign map. One hill might have slight different features compared to an adjacent hill. Combat maps would require some unique tiles, such as slopes and beaches.

Prior to combat, the attacking and defending armies would choose how to place their armies on their side of the mini tile-set map (ambushes would not allow much defensive preparation). They would then battle it out until someone retreats due to low forces or low morale or just because they chose to. Defending fortified units cannot retreat, but they can choose to pay the attacking player to allow them to retreat.


PROS:

- Retains 1UPT tactics without the clutter in the campaign map.
- Easier on the AI. The AI usually struggles with having to constantly change its unit placements on the campaign map. However, with limited stacked armies, the AI would need to set its units only prior to combat on the combat map. Defending AI in particular would have a much easier time.
- Simplified AI vs AI stack battles (like Civ IV), but fully fledged when human player is involved.


CONS:

- Possibly makes single player slower? Not too sure about that. Combat takes longer, but turn times are shortened. Less processing is required on the actual campaign map since there aren't hundreds of units moving about.
- Definitely makes multiplayer slower and can become boring for 3+ games. This could be solved with an option for auto-resolve combat (which would essentially make it equivalent to stack vs stack combat in Civ IV).


PS: I'm fully aware that this would deviate considerably from previous civ titles, but I also think the reason this has never been attempted has a lot to do with limitations in what PCs can actually handle.

 

[Sostratus]

I think that the core mechanics civ6 has made available that weren't really there in 5 have made a return to "stacks" fairly obsolete and unneeded.

Between having a corps/army system to increase strength density, and the critically underutilized role of support units, the only thing that can't be directly represented is a combined arms stack on one tile. The seem to have chosen (and I much prefer it this way) to have combined arms represented by multiple units on multiple tiles.

But there's no reason we could not, for example, have the ability to form corps+army (call them something else) right away and either have their efficacy improved later or have even higher forms of formation unlocked later.
They can represent force concentration during a siege or defending a vital spot or any number of scenarios where you want a big push in one spot.
Then you have support units. There are so many things they can be used for. We could have much more battlefield specialization support units. We have the ram and siege tower. We could have other stuff like machine guns become support units, or implement mantlets (movable wooden walls, basically) to help shield footmen from ranged fire.

You have a two axis unit system to control quantitative power (formation density) and qualitative ability (support units.) Literally the only thing missing is being able to put horses and swords and archers into one formation, but as i said, i think the game play is better with them on separate tiles. Especially if you add the ability to decompose formations, there's almost nothing you cannot do if you are creative.
Heck, you could even have Mechanized infantry be replaced by simply having IFV support units attached to infantry to make them more effective in combat. (As an example.)
If you want to get really really crazy, you could co-opt the GDR's upgrades through the tech tree concept to apply to all units, which would allow for extremely granular scenario technology. (For example, when you get ironworking, your spearmen can now carry better equipment. Etc.)

All of this, at our fingertips, right now in civ6.

 

[Boris Gudenuf]

All of this, at our fingertips, right now in civ6.

I think it would be more accurate to say right now this is all dangling just out of reach of our fingertips, because Firaxis, as in so many things, never developed the very promising systems as far as they could have, and the Mod Community doesn't seem to have jumped on them either.

The Potential is certainly there, though.

 

[The googles do nothing]

Excellent post. I think your asking in your vote about future versions of civ and I voted for the current system. I enjoy the challenge of figuring out the optimal movement of unit more the figuring out the optimal combination of units.

Personally I don't think there should be a next version until:

• AI can handle UPT
• Handles 'pathing' - you can place units in a formation and move them together - and they move in an logical optimal way.
• Tiles are half the scale they are now - one tile now would be 2 - and add impassable terrain or terrain that causes damage to unit moving through, different movement costs based on unit types. This means bigger maps.

 

[Gedemon]

I think that the core mechanics civ6 has made available that weren't really there in 5 have made a return to "stacks" fairly obsolete and unneeded.

Except for smaller map sizes, combined with unstacked cities and still inadequate AI...

 

[Gedemon]

There were several Threads and GOK how many posts regarding Stacking and Stacking Systems when Civ V's 1UPT first came out, so it's not surprising that it comes up again. A lot f what I'm going to say here is based on posts made a while ago, because I think they still apply to the Stacking Problem.

There are several things to consider before wading into the Stacking Swamp:
1. Any system of 'limited' Stacking means you are introducing a bunch of Stacking Rules, Limitations, Special Cases that will Complicate the Game. There had better be some durned good reasons for Gamers to put up with those complications.
2. The best thing about the current 1UPT system is the Tactical 'Feel' it gives battles: the interaction between Ranged and Melee, Flanking and Support bonuses, etc. Any stacking system has to maintain that, or it will be a bland and uninteresting (and disliked) replacement.
3. IF you are going to introduce some kind of Tactical Limitations/Rules for Stacking, for once it would be nice if the game actually got the historical terminology right. Your examples are of Inaccurate Terminology, so excuse me if I address them even though the corrections have nothing directly to do with Stacking Systems.



Civ 4 had (virtually) Unlimited Stacking, and nothing to differentiate the various types of units in the stack: it had the advantage of being really simple, really bland, and really easy to Abuse. Good Riddance to it.



First of all, a "battalion" is a very specific unit in modern military terminology since the Renaissance, and it derives from the Medieval "Battle" which was simply 1/4 to 1/3 of the entire army, with no set limit in size. So, it's either too specific or not specific enough for Game Use. Also, a modern battalion is always a single type of troops: infantry, artillery, armor, etc. So there is no provision in this terminology for a stack containing a mixed group, making it no different from Civ VI's very limited Corps/Army units.

I suggest that, like the size of cities in the game, we use a 'sliding scale' of Unit Sizes: Ancient Armies had no 'fixed' units larger than about 1000 men (very close to a modern 'battalion' of 700 - 1000) - the Egyptian Pedjet, Sumer Sanga. In the cases we have record of, the meaning is "Host". These Hosts could consist of several smaller units (Egyptian Sa, Sumer Nu-Banda) similar to modern Companies.
I suggest, therefore, that the first Stack of multiple Units be a Host which can be of all one type (like the Egyptian chariot Pedjet of 250 chariots) or multiple types of troops.

The historical limitations on how many troops could be concentrated ('stacked') were Terrain, Supply, and Organization. Very few Ancient or Classical armies had any organization higher than the 1000 or so men in the 'Host', which meant that forces even as large as a modern Division (15 - 25,000 men) were almost impossible to control (stack). Since even a few thousand men and their animals would eat as much as a small city of the time, they were also very difficult to supply for any length of time, so the availability of food and fodder from the terrain (tile) was also hugely important. The biggest technological changes in Supply (stacking) are in supply technology - better roads, better transport vehicles, riverine/coastal boat supply, more efficient bureaucracies to control supplies. Ability to survive and fight in extreme terrains depended until near-Modern eras on size of the force and native technologies from people familiar with the terrain - there's a reason why almost every Civ hired people who lived in the mountains to fight in the mountains, or natives of the rainforest to fight in the rainforest.

So, an obvious 'basic' Stacking limitation might be By Era: in the Ancient Era, you can basically stack 3 units, but, say, only 2 in rainforest, tundra, snow, or desert. Classical Era = 4 basic units/stack, and so on.
Being next to a river or road that leads to a friendly city would increase the stacking limit, but not by much before the 'movement/transport' technology improves dramatically.

This sort of system has the advantage of being a largely invisible complication: stacking limit rules are few and related directly to the Era you are in so that for the most part the Gamer does not have to constantly mentally refer to a bunch of rules and restrictions.



Glad you used 'blitzkrieg' as an example, since it is an example of a Pop History term that is meaningless. Rapid advance and attack has been present in warfare since Tomyris or Attila charged out of the Steppe and then disappeared before anyone could react. The specific German motorized all-arms tactical and operational technique labeled 'blitzkrieg' (by journalists, NEVER by the German military) was in fact Bewegungskrieg ("Maneuver Warfare") and in form dates back to the Great Elector of Brandenburg of the 1680s (the Fehrbellin campaign, go look it up), modernized by adding motor vehicles. The tactical specific is to always use combined arms, so a 'tank only' stack, by definition, is not a 'blitzkrieg' of any kind.
What you are really talking about is a formation that is a Panzer Division (US Armored Division, Soviet Tank Corps, etc.) the tactical/operational Building Block of Modern/Atomic Era Maneuver Warfare. These Formations would all consist of a combination of Units. Historically, they were Tank, Infantry, Artillery, Combat Engineers, Scouts, Antitank and Antiaircraft units. That's 'way too many for the game, so I suggest that the in-game Panzer Division consist of a maximum of 3 Units: 1 Tank, 1 Infantry, 1 Artillery, all moving at the speed of tanks (motorized infantry and artillery) .

Which, however, brings up the other Big Problem of stacking: how do you include the Tactical interaction of Ranged and Melee factors, Flanking, Support, and Terrain bonuses that the current 1UPT system has now?
I suggest that a Combined Arms Stack or stack of Formations reduces any Malus from enemy forces or terrain to that of the most advantageous unit in the stack. That is, a city center or district tile represents a built up area, so has a defensive bonus against any type of Unit, but has a Major defensive bonus against Heavy or Light Mounted Units, which includes Tanks. A Unit or Stack of Unsupported tanks attacking a unit in a District will see its melee factors reduced by at least half. But if it is stacked with infantry, or in a Formation that includes infantry, the malus is that of the infantry for all melee units in the stack/formation and the tanks' melee factor can be added to the total with very little reduction. The ranged factor of any Artillery/ranged units in the stack would be applied as Support Fire before the stack (or before calculating defensive factors if on the defense).

That would make Combined Arms Formations very powerful, as they were - but also expensive, since the Panzer Division and its equivalents required that Artillery and Infantry units included in it be motorized to the same degree as the tanks, with all the resulting requirements in Production and Maintenance Costs and Resource Costs. For example, one refill of fuel for a Panzer Division was calculated as 220 cubic meters of fuel, for a German Infantry Division it was 6 to 9 cubic meters - a huge difference in Supply/Support requirements between the 'mobile' panzers and the 'marching' Infantry.



Strictly limiting the types and numbers of Units that can survive/move/right in certain types of terrain and climate would extend Exploration into the late game, which would be a Good Thing.

The types of terrain that were historically Deadly to unequipped, unacclimatized armies and units were Tundra, Snow, Desert, Rainforest, Marsh/Floodplains, and Forest, in varying degrees. Forest largely made certain types of units harder to use tactically: Heavy Cavalry in dense woods is pretty Counter-Intuitive, for instance. The other types of terrain killed men and animals, either because they were associated with disease (Marsh/Floodplains, Rainforest) or because there was nothing to eat and killing Climate associated with them (Tundra, Snow, Desert).
So I would put a strict Stacking Limit on Tundra, Snow, and Desert terrain tiles from the start of the game, and perhaps, except for certain Civs, an attritional factor for any unit in Snow, Tundra or Rainforest except Scouts. Unequipped, unacclimatized troops simply cannot survive there without technologies that are not available until later or available from the people living in that terrain/climate.
Stacking in Marsh/Floodplains or Rainforest tiles would also be dangerous: you can do it, but the more troops to put within range of the mosquitoes and tsetse flies and other 'natives' the more you will inevitably lose until technologies like Germ Theory or Antibiotics or Sanitation give you counters to the conditions.

I mentioned Supply already above. The thing to remember here is that for most of history, Supply = food and fodder for men and animals in the host/force/formation/army, and for most of history the armies provided most of their own supply by sweeping up everything edible from the country they moved through. That's another reason that desert. rainforest, tundra and snow tiles are dangerous: nothing to eat, or nothing that the non-native recognizes as edible.
Food and Fodder being bulky, heavy materials required in bulk (rough estimate for Modern/Atomic Era armies: 1 Ton of Food per day per 1000 men minimum, 5 - 10 tons of Fodder per day per 1000 horses, depending on how much they can 'graze' from the surrounding area) the only way to transport them until railroads was by water: river or coastal shipping. That means any change in supply for at least the first Eras of the game (until Industrial) is related to access to river/coast leading to a friendly/allied City as an Ultimate Source.
Troops can be supplied overland, but the distance is strictly limited: an animal-drawn conveyance, no matter how efficient, will consume the weight it is carrying in food for the draft animals after about 100 - 150 kilometers, which means beyond that you are effectively transporting 0 supplies.
So, a Stack is limited not only by Era (basic Technology/Organization) but also by how many can be fed in a given tile. A functioning Supply Line means any number can be fed. A tile that has been organized already (Improved) can be 'Foraged' to provide Supply for X Units for 1 Turn, but only until Gunpowder starts to require Supplies not available locally (gunpowder, ammunition, replacement weapons). Foraging would be like Pillaging, but more temporary (Tile 'recovers' automatically in 2 - 3 Turns) and automatic if the Stack doesn't have a functioning or adequate Supply Line. That means, as was true historically, that an army staying in one place without a supply line is In Trouble: don't try to besiege a city without a river or coastal Supply Line!

That also means that Supply For Armies changes dramatically with the advent of the Railroad: the railroad allows any amount of supplies to be provided from any city connected to the railroad over any type of terrain: strategy and operations will become 'railroad dependent' for both Food and the increasingly important Fuel and Ammunition required to make Industrial and later era units combat capable.
This, by the way, indirectly increases the importance of the Industrial Revolution and its consequences, something else the game is currently lacking.

Last comment: Supply lines can be cut or broken by raiding units, which would also affect attrition due to Stacking Limits. This could make Light Cavalry/Light Mounted units more important in some circumstances or air/sea attack an effective 'Supply/Stacking Modifier': an enemy Bireme or Galley syitting across your coastal Trade Route/Supply line could mean starvation for your Roman Legions trying to do an end run through the coastal desert!
Later, air attack can also interrupt railroad and sea Trade/Supply routes, but not as completely: historically, railroads could be repaired within hours or days even after major air strikes, and it took a very sustained, powerful air campaign to tulle interrupt supply, like the massive Allied air campaign lasting months before and after D-Day in northern France in 1944.

Thanks for a great post again.

I'm testing limited stacking in my mods since civ5 (R.E.D. WWII, Combat& Stacking Overhaul), and I still want to build up something based on those experience for civ6 (if/when we're finally authorized to mod that...)

Depending on the era and/or organization/supply mechanism, I'd use a limited stacked combining:
1 to 2 "division" or "brigade" units representing the bulk of thet roops (ie infantry, mechanized infantry, tank, ...)
3 to 6 support "regiments" or "battalions" units representing specialized troops with specific abilities (ie artillery, cavalry, engineers, special forces, assault gun...) or "normal" troops (infantry,...) on a smaller scale (faster to build but weaker) to provide eventual stacking bonuses (I like your proposal for the combined arm stack limited maluses)

For the ranged vs melee tactics, something I think is helping the AI is to set all range to one (except for some late game very specific units) and use various support fire mechanisms limiting risk-free attacks.

For example, in my civ5 mods, ranged units could have some of all of the following abilities: offensive fire (selected ranged attack during their player turn on adjacent tiles or automated preparatory fire when a melee unit in your stack is attacking), defensive fire (automated fire on an adjacent attacking unit before resolving the melee combat), counter-fire (automated fire against adjacent enemy artillery units)

On a frontline with multiple adjacent stacks including artilleries, you had to think twice before opening fire or even starting a combat... IIRC I had set a "fire point" limit per turn to prevent an artillery unit to be able to provide "infinite" support fire on a stack vs all 6 adjacent plots during the same turn.

Supply (local and lines) as a soft limit for stacking is something I've not experimented on yet, but would love to include in my civ6 mods, still trying to keep mental notes of your various posts on the subject for the overall design (that already include food consumption and has some bases for health and diseases)

 

[Zeuxis]

I only ever played a couple of games of Civ 5 and stayed with Civ 4 because I hated the ridiculous moving around of units so they could all attack. It made the game fiddly, unrealistic and no fun for me.

The armies in Civ 6 make things a bit better but it is still a tad irritating.

Perhaps a realistic solution would be to cap the number of units of a single type on a square, such as one melee plus one ranged plus one support unit......

 

[Lonecat Nekophrodite]

Stacking. this time Pike & Shotte should also be formed by combining Musketeers with Pikemen,

also any Infantry unit should be able to add either field cannons or MGs into a stack or combo choices.

 

[Mik1984]

What Civ VI basically needs is to:

- learn unit group management from this mod:
https://forums.civfanatics.com/threads/ai.603494/page-61#post-15473036

(please do not install this mod, it is a complete failure at this point, because the dev started overhauling the entire AI and things went awry. Read my review for more info)
- implement more features known from Panzer General - literally the godfather of hex based one unit per tile strategy turn games. What combat in Civ 6 is, is a heavily abridged version of Panzer General style games.

- I believe even corps and armies with the features above can be removed from the game as quasi-unit stacks. They should rather be enchancers of a known feature of hex games (even the board based ones) called "line of command".

That is my final opinion. Civ should not abandon one unit per tile but instead should go all in, learning from a large collection of one unit per tile strategy classics. Basically don'a abandon one unit per tile, instead go all in and implement all the features that made the one unit per tile strategy classics so great.

Furthermore, there are one unit per tile games of the series "Great Battles of ..." the combat in ancient and medieval era can learn from these games. For example a fallanx, can be a unit that spans over two tiles in its battle formation and has facing, invulnerable from the front, but needs to have its flanks secured. Legions have a charge of javelin attack, but they have to take many turns to recharge it. Units may have limited ammunition, and fuel later in the game.

Every era would have something unique about the combat in that era. So on and so forth.

For now the problems with one unit per tile are mostly related to the fact that civ 6 has basically dipped its toe in this feature without going all in.

 

[epicivfreak]

Personally, I hated the unlimited stacking style of warfare of Civ4, but I also hate the 1UPT style of warfare in Civ5. I haven't bothered with Civ6, as I'm sure it sucks worse than previous editions of Civ. I'm playing a strategic game, and I want good strategic warfare to go with it, not this tactical nonsense.

So, my vote is that Firaxis loads up the code for Civ4, pretend 5 & 6 never happened, and try again. Try something, anything, different.

 

[ghost88]

My preference would be stacking similar to Age of Wonders III. Perhaps 6-8 units of your choice on the main map ( maybe even allowing multi-stack attacks and defense), with a tactical map for those who like to "fight" their battles, with auto battles for those who do not.
Of course as I have stated elsewhere for me to continue playing Civ they need to make all their "bells and whistles" like World Diplomacy, Disasters, and Global Warming optional or with a sliding scale that allows me to play Civ my way not being forced to have to micro-manage more things each time the game is "improved". But that is a different subject from the OP's