Transcending the SoD

[Zeiter]

Is it just me, or does Civ4 need to transcend the simplistic stack-o-doom style of warfare? As it currently is, I have no way of, for example, fighting a WWI type of war. In WWI, each side did not just throw together SoD's and march to the other's cities. There were strategic fronts. There was strategic defense-in-depth.

I know that the motivation behind the SoD was so that the programming would be simpler and more easily grasped by the AI...but then again, checkers is easily programmable and easily grasped by AI as well. That doesn't make checkers a better game.

And as it currently stands, SoD warfare is basically the equivalent of Starcraft warfare. You put together a big SoD with some siege and cleanup troops, and hurl it at the enemy. Part of the problem is obviously siege, but siege wouldn't be so bad if the mechanics privileged the stack-o-doom in the first place. (The saving grace of Civ4 remains the complexity of the economy. Without this complexity, Civ4 really would be the TBS-equivalent of Starcraft).

*Zones of Control:

I don't think the answer is to implement hard zones-of-control like in Alpha Centauri.

Instead, I'd like to see a small amount of damage done to units that move through ZoC, depending on how many units are contributing to the ZoC and depending on the types of units (such as for every melee unit, 1*(strength ratio) hp damage per move through ZoC, for every archery, siege, mounted, gunpowder, and armored unit, 2*(strength ratio) hp damage per move through ZoC).

*Siege:

I'd like for siege to *potentially* damage all units in the attacked tile, but for the damage to depend on how many regular combat rounds it survives against the top defender (so if the catapult got killed off solely by a longbow's first strikes, the catapult would not get an opportunity to do any collateral), and I'd like for the collateral damage to be a lot less in general. As an alternative to this sort of head-on siege, I'd like there to also be a "mission" type of lesser bombardment (with a higher cap, such as 25%), just like how aircraft do damage. Siege should also be able to bombard units' fortification bonuses in the field (8% per turn for cats, 16% for trebs, etc.)

This would encourage the use of more "mini-stacks". These would be combined arms stacks, but with the incentive to be smaller and split up so that more counter-siege would be needed to do the same damage. Would it be impossible to program the AI to split its stacks into smaller combined-arms stacks once those stacks reached certain size ranges?

*More nuanced terrain/fortification dynamics:

I'd also like to see the defensive bonuses of forests, for example, to be decreased to +10%, but I'd like for the max fortification bonus to be adjustable based on the type of terrain (and the type of unit you are using--fortifying in forests would be more useful for melee than for archery units, and vice-versa for hills). (For example, fighting in a forest hurts you almost just as much as the attacker if you aren't fortified in the forest--the French found that out the painful way when trying to regroup and counter the Germans in the Ardennes during WWII). So here are the new bonuses I'd like to see:

Desert/Ice: no inherent defensive bonus, 10% max fortification bonus

Tundra/plains/grassland: no inherent defensive bonus, 25% max fortification bonus

Forest: +10% inherent, +50% max fortification bonus for melee, +40% max fortification bonus for archery, gunpowder, +60% max fortification bonus for trench units (in the Wolfshanze mod, this includes machine guns, anti-tank infantry, SAM infantry, and mobile SAM).

Hills: +20% inherent, +40% max fortification bonus for archery, gunpowder, +30% max fortification bonus for melee, +50% max fortification bonus for trench units.

Forts: extra 20% inherent, extra +20% max fortification bonus for melee, extra +30% max fortification bonus for archery and gunpowder, +40% max fortification bonus for trench units.

Cities: +25% inherent. Walls/castle/cultural defense % = max fortification bonus for all units that can fortify. This % can be bombarded down (units would gradually lose their fortification bonus down to zero). The 25% inherent bonus will always remain (you could still hide in the rubble).

Thus, there can be a situational advantage to defending from cities, except that...city bombardment should also have a small random chance of destroying a (non-wall, non-castle) building. So you might want to defend along a front out in the field if you can gather enough troops to do so. (And you may not want to defend a city and risk destruction of its infrastructure and architecture (just like how the French didn't want to defend from Paris and risk having all their beautiful buildings destroyed), but rather give it up and make a strategic retreat, defeat the enemy army in the field, and then swoop back to retake the city with minimal fighting.

So, to have the AI take advantage of this, program it to station more archery units on hills, more melee units in forests, having it occasionally use smaller stacks, have it concentrate those stacks to punch through fronts on a narrow span...even if the AI programming to take advantage of this tactical nuances was sub-optimal, I'd be more satisfied with this than the braindead SoD warfare that we have now.

 

[Tholish]

Another idea to reduce SOD is to give cities a negative defensive bonus. And if units had facing, were weaker on some sides than others, they would have to support each others flanks and form fronts.

 

[Skallagrimson]

You'd see less stacking if field artillery attacks weren't so unrealistically suicidal. Artillery should act similar to bombers for bombardment, but with adjacent-tile range. During their attack they should take no damage, but they should have no defensive value, requiring escort at all times from other units.

As a result of this you'd see more line formations allowing concentrated attacks at 3:1 at tactical weak spots, and attempts to exploit the opening, more like the way real war is fought in the field.

 

[rysmiel]

You'd see less stacking if field artillery attacks weren't so unrealistically suicidal. Artillery should act similar to bombers for bombardment, but with adjacent-tile range. During their attack they should take no damage, but they should have no defensive value, requiring escort at all times from other units.

Civ III artillery works pretty much like this, and what that leads to in practice as an optimal tactic is stacks of huge numbers of artillery with some heavy defenders trundling from city to city taking out all their defenders one at a time, rather than anything line-of-battley.

While I am not opposed to the notion of some mode of strategy not leaning on stacks of doom, I am very much opposed to any approach to such that involves tactical minigames, flanking, or any other of that kind of small-scale thinking; I play primarily Civ III because I enjoy the logistics and grand strategy level and don't much like even the degree of tactical-scale focused Civ IV is (like individual unit promotions) and would very much like this not to get any worse.

One thing that would seem to me to work to mitigate the utility of stacks of doom is to limit the effectiveness of defending them. You could, for example, say that one defender can only protect two artillery units, so that if you want to drag forty catapults to battle in a stack, if you have any fewer than twenty defenders sitting on them, any attack gets to take out one of the catapults as if it were entirely undefended. This does not remove SoDs entirely but it would make them a sight more expensive and difficult to maintain.

 

[Skallagrimson]

This is why it's also important that artillery units don't annihilate units they attack. They can only weaken, and weaken down to a minimum combat strength. With that in mind, a Civ3-ish gigantic stack of artillery would be meaningless and you'd have to supplement with other units.

I'm the opposite when it comes to getting down to tactical scale, but I agree that it should be an optional level of military MM at all times. In MTW you can lead a battle yourself, micro-ordering each unit where to march, whom to attack, etc., like a field commander, OR, you can auto-resolve unit battles and get on with the higher-level campaign. Pure flavor choice. In Civ your choices are just two different layers at the higher level, with no true tactical layer of choice, and I think that's a gap.

In a way stack defense is already limited because when attacking a stack you can always suicide a few initial units to weaken the defender, and then slam in the cavalry who shred through all the siege units.

 

[rysmiel]

This is why it's also important that artillery units don't annihilate units they attack. They can only weaken, and weaken down to a minimum combat strength. With that in mind, a Civ3-ish gigantic stack of artillery would be meaningless and you'd have to supplement with other units.

Only very few and late units in Civ III have lethal bombard, so you do in fact need to back them up with something to kill the nearly-dead enemy, but not in any huge numbers compared to the number of artillery, IME.

I am inclined to think that artillery units might as well be represented as regular units with very high attack strength and very low if not zero defence, myself, (zero defence for a catapult makes more sense to me than zero defence for a modern missile-bombardment type artillery unit) if one does not want to have ranges greater than a single square for bombardment.

In Civ your choices are just two different layers at the higher level, with no true tactical layer of choice, and I think that's a gap.

How is promoting individual units in different directions for different specialties and combat roles not a tactical-level choice ?

 

[Skallagrimson]

How is promoting individual units in different directions for different specialties and combat roles not a tactical-level choice ?

Core characteristics of a unit is a strategic question (beyond the scope of a single battle). It's low-level strategic, but still strategic.

Movements and goals within the battle itself are "tactical".

 

[rysmiel]

Core characteristics of a unit is a strategic question (beyond the scope of a single battle). It's low-level strategic, but still strategic.

Movements and goals within the battle itself are "tactical".

This may just be a difference in semantic usage, but to my mind something like the "Finer Points of Tank Warfare" article in the War Academy (http://www.civfanatics.com/civ4/strategy/tank_warfare.php) is Civ IV doing tactical thinking, and I just don't find that particularly interesting compared to the challenges of building the logistics and winning through strategy on a large scale.

 

[Skallagrimson]

Do you mm the economy of a single city?

 

[rysmiel]

Do you mm the economy of a single city?

Sure, to some extent, but I expect in an ideal Civ-like game to have significantly more units in play than cities, particularly late in the game. Easily more than an order of magnitude more if I'm in the middle of a major war.

 

[r_rolo1]

May I suggest a simple solution for the mega artilery stack with something to mop up?

Limit the number of siege units per tile. Same as it was done with super-bomber stack.

 

[Antilogic]

In all fairness, "stacks of doom" is a far more accurate way to represent warfare before the 20th century. Armies typically massed and fought within a relatively small region, or conducted extended sieges. I haven't actually done the math, but if you take the circumference of the earth and divide by the number of tiles horizontally across the map, I think you will be able to encompass most battlefields within that region unless you play on the most gigantic of maps.

If anything, a complete overhaul may not be necessary. Just a focus on changing the 20th century warfare to allow for trench-style warfare and larger fronts.

 

[Slaughter]

A way to stop the "Artillery Stack of Doom" problem would be artillery duels, like in Alpha Centauri. Two stacks meet each other. Both Stack A and B have catapults. Stack A bombards Stack B and begins a Artillery Battle. They duke it out until one of them withdrawns or is destroyed.

A limit to siege units is a good idea, too.

Besides the SOD problem, another thing that bothers me is the strange way battles are fought. Two units enter, one leaves. Its like freakin' Thunderdome. Sure, I don't mind if a crushing defeat kills everyone, but why EVERY skirmish between two units must end with one of the units totally dead? Cavalry can retreat, but its an exception rather than the rule. Why?

I would fix it by giving every unit a retreating chance. Say, Melee Infantry has 5%, Cavalry has 25%, Ranged Units have 10% and later 15% with gunpowder, Mech Infantry and Tanks 30%. Retreating chance can be increased with promotions. A unit can retreat IF:
- There is at least one nearby square with no enemies to retreat (can't retreat to peaks or water, unless there's a transport there).
- It can't retreat to a square near a garrisoned enemy fort or into a enemy fort with troops, but it may do if the fort is empty.
- If the attacking unit has lower or same speed of the retreating unit. (this means that infantry can retreat from combats with infantry, but it can't retreat from cavalry or vehicles that can outrun and destroy it)
- If there's a faster enemy unit (faster infantry excepted) nearby, it engages in a atempt to pursue the retreating unit. If the retreating unit suceeds, then the retreat suceeds, though it may sustain damage. If the retreating unit fails, then its destroyed.

 

[Zeiter]

I like the idea about giving siege defensive penalties. In other words, not only can siege not receive defensive bonuses, but they have like -50% defense as well.

Hmmm...yes...perhaps, the SoD is appropriate for the pre-modern era of warfare...but as you noted, not for the modern era.

Well, in order to figure out how to change warfare in the game to emulate this change, we need to figure out what factors in real life warfare actually changed to give rise to massive fronts...

I think it had to do with the massiveness of logistics and supply lines. Food, fuel, ammunition...if you allowed an enemy army to dart around you and get behind you, between you and your territory, you were toast. Whereas in earlier times armies could just live off the land through pillaging, I guess. Something like Sherman's march to the sea, where he cut off his communications and supply lines with the north and set off on his own from Atlanta to Savannah as a single concentrated army...that would not have been possible in WWI. Past a certain army size, that's not possible. You need railroads, horses, and transport vehicles to get through to your army to keep it on its feet. So you needed a front to protect your supply lines, to make sure the enemy did not get around or behind you and threaten those supply lines.

So here's an idea: you can have as many units as you want on a tile as long as there is a path back to one's trade network that is free of enemy units' zone of control. If a group of enemy units does get its ZoC between your army and your trade network, then your units in those tiles will take damage at 10% per turn in grassland, 15% per turn in plains, 20% per turn in tundra (the Russian winter strikes your army), and 30% per turn in ice or desert, with a max damage from this mechanism of 90% (it will not kill off your units, just weaken them gruesomely). The only way to prevent this, once your supply lines are cut, and assuming you cannot counter-attack to re-establish supply lines, is to split up your stack so that it can live off the land (no need to formally pillage, we can assume that they are commandeering local food and supplies from locals, but not burning farmsteads and whatnot). The limit will be 10 units in grassland, 8 in plains, 6 in tundra, and 4 in ice or desert. (Maybe this could scale with map size and/or gamespeed).

So this would not pose much of a hindrance for earlier wars, when stack sizes are not likely to be much larger than that.

 

[Camikaze]

I like the idea about giving siege defensive penalties. In other words, not only can siege not receive defensive bonuses, but they have like -50% defense as well.

Hmmm...yes...perhaps, the SoD is appropriate for the pre-modern era of warfare...but as you noted, not for the modern era.

Well, in order to figure out how to change warfare in the game to emulate this change, we need to figure out what factors in real life warfare actually changed to give rise to massive fronts...

I think larger armies created larger fronts. There is only a finite concentration of soldiers you can possibly have, and so larger armies mean that fronts are expanded by necessity. Think the Schlieffen Plan. Going through Belgium was a logistical nightmare. This brings me to my theory. Perhaps SoD's only seem to be more relevant to pre-modern warfare due to the fact that the same number of troops was a larger proportion of an army, i.e. A stack of 20 000 in World War One would seem less significant than a stack of 20 000 in the Hundred Years War. This is only because the scale of the war in World War One was much bigger than in the Hundred Years War. Therefore, it follows that modern warfare does contain Stacks of Doom, just a whole lot more of them than in previous times. I think this reflected in the game well, with higher population and production in the latter stages of the game leading to bigger stacks. However, these are a bit cumbersome, and so perhaps something needs to be done to reduce their size without compromising the historical principle of increasingly large armies. I don't know what.

Also, I think siege weapons should have better defence capabilities. When I see a trebuchet, I instantly think Minas Tirith and the defensive trebuchets. Also, in World War One, it could be said that artillery were used defensively in 'Deep Battle', in order to take out other artillery. I think an improvement to artillery would be the ability to bombard a stack like it does a city. In reality, artillery does not charge into battle and die as a consequence, causing collateral damage. It bombards and causes collateral damage without getting near the front. To solve this, I propose making artillery and other siege weapons like Great Generals- attachable to a unit. That unit could then bombard for it's turn, causing collateral damage. That's my two cents, anyway.

 

[Antilogic]

Uh...how about we stick with real battles and not ones involving orcs? At least for the main game--you can do whatever in your mod.

Also, artillery, the actual 18 strength unit, has a +50% bonus against other siege equipment. When attacking cities, if they are promoted with first strikes, they can defend against enemy artillery strikes...at least until tanks and marines come along.

 

[rysmiel]

May I suggest a simple solution for the mega artilery stack with something to mop up?

Limit the number of siege units per tile. Same as it was done with super-bomber stack.

I think that's unsatisfactory as an approach to bomber density and would be lousy as an approach to siege units: it get back to how big a tile is supposed to be anyway.

 

[rysmiel]

A way to stop the "Artillery Stack of Doom" problem would be artillery duels, like in Alpha Centauri. Two stacks meet each other. Both Stack A and B have catapults. Stack A bombards Stack B and begins a Artillery Battle. They duke it out until one of them withdrawns or is destroyed.

That only works if the person who sees your artillery stack has some motivation for meeting it with another artillery stack, though.

Besides the SOD problem, another thing that bothers me is the strange way battles are fought. Two units enter, one leaves. Its like freakin' Thunderdome. Sure, I don't mind if a crushing defeat kills everyone, but why EVERY skirmish between two units must end with one of the units totally dead? Cavalry can retreat, but its an exception rather than the rule. Why?

As I understand it, the underlying idea there from Civ III is that any unit with move 2 can retreat from battle with any unit with move 1 - and likewise for move 3 with move 2, I suppose, not sure I have tested that. This is not entirely a dumb idea.

I would fix it by giving every unit a retreating chance. Say, Melee Infantry has 5%, Cavalry has 25%, Ranged Units have 10% and later 15% with gunpowder, Mech Infantry and Tanks 30%.

This seems to be adding a direction of complication that achieves nothing beyond itself, and also to be focusing more on battle-level tactics and less on larger strategy, so I would vote against.

 

[Skallagrimson]

Uh...how about we stick with real battles and not ones involving orcs?

ROFL, might as well have orcs the way Civ is written at this point.

Keep most of the economy, dump all the rest, start over.

 

[Camikaze]

Uh...how about we stick with real battles and not ones involving orcs? At least for the main game--you can do whatever in your mod.

I don't mean we should include mythical creatures as units (although it could be interesting). I was merely pointing out that siege units like trebuchets, and for that matter, catapults, cannons and artillery, can be used in a defensive manner, and you should be able to build them to do so in a more effective way.