How combat should work

[Squirrelloid]

Forked from the tech tree thread.

The basic CivIV chassis gives us a rather R/P/S style combat in the ancient era with axemen/chariots/spearmen - units have counters and combined stacks are less vulnerable than stacks of one unit type (throw in a few archers for defense against axemen and you have the least vulnerable stack available in that era, which happens to represent every unit type of the era as well).

Design Goal: encourage combined arms.

The CivIV concept gives us a place for all the units. Unfortunately, as you move forward in time CivIV's combat model breaks down. The medieval era is a disaster - which is exactly where we want this mod to be playing (with the addition of crazy stuff like Dragons). So we need to rebuild the medieval era to make every possible unit type playable and useful.

Enter actual historical performance of various unit types. Military historians recognize four basic types of units: heavy infantry, light infantry, heavy cavalry, and light cavalry. Heavy types use shock combat (melee), while light types use ranged combat (typically bows, but potentially slings, arquebusiers, javelins, etc...). Given relatively equivalent quality of units, the following performance metrics apply:

HC > LI
HI > HC
LI > LC, HI
LC > HC, HI

Rationales: Mounted units must use a weapon and keep control of the horse, splitting their focus. They also have less leverage (melee) or are less stable (ranged) than their unmounted counterparts - meaning that in general the unmounted equivalent should outperform the mounted troops. (Covers LI > LC, HI > HC)

Light troops, in addition to using ranged weapons, also tend towards lighter armor, meaning the man (and/or mount) tires less rapidly and is faster, thus allowing it to effectively retreat from the enemy, and turn and fire when they slow or stop their advance. Ie, LC beats HC because it can run away and turn and fire whenever the HC gives up the chase. (Covers LC > HC, LI > HI).

HC beats LI because infantry can't outrun cavalry, so they can't use the LC trick to beat HC, and they lack the weapons and armor to defeat HC.

LC beats HI for the obvious reasons.

Now, historically there have been some innovations which have altered the balance in predictable fashions. The development of the lance gave cavalry dominance over infantry lacking pike weapons - that infantry at the time were typically peasants while the cavalry were nobility in europe meant it took a few hundred years for anyone to bother handing the peasants something other than farm tools. The long dominance of cavalry in europe should be considered an aberration stemming from primarily social causes, and the perceived dominance of cavalry by nonhistorians continues long after the infantry had begun to dominate the field. (Bannockburn could be considered the end of cavalry as a dominant force on the battlefield, but if not then Agincourt.)

The development of the matchlock pistol gave cavalry the ability to be light and heavy at the same time, which led to a 50-100 year period in which cavalry were the dominant force on the battlefield. Subsequently, increasing rate of fire of rifles plus the development of the bayonet made cavalry all but obsolete in a primary combat role. Of course, the last of these is after the time horizon of interest.

One other differentiation we will likely want to make is that early pike troops are inefficient against infantry wielding shorter weapons, who can step inside the long spears and still be effective. Ie, the Roman legionaire vs. the Greek phalanx. Later medieval pikemen drilled in mobile formations to overcome such difficulties. (The elimination of a large shield almost certainly aided this adaptation).

While there are differences betwee swords/axes/maces for an individual soldier, they aren't especially notable at large scales and differentiating based on different non-pike melee weapons is silly.

Historically, light cavalry are the last to be deployed in combat. However, seeing as some cultures started their military career with light cavalry (ie, the huns, mongols, etc...), it probably has more to do with a relative lack of horses in quantity in the areas which started recording history.

So, that's the theory, next post will look into structure and implementation.

 

[Squirrelloid]

I'm also including siege engines. Note that I'm only listing units which are defined by new technology in each era - either in terms of equipment or in terms of doctrine.

Units:
Tier 0: "Ancient"
Warrior (Heavy Infantry)

Tier 1: "Classical"
Archer (Light Infantry)
Spearman (Heavy Infantry)
Ancient Cavalry (Heavy Cavalry)
Horse Archer (Light Cavalry)
Catapult (Siege Engine)
Chariot
Siege Engineer (see below)

Tier 2: "Medieval"
Unfortunately WH has this idea that the sword is actually an effective primary weapon... But I suppose we can possibly use that...

Longbow (Light Infantry)
Pikeman (Heavy Infantry)
Knights (Heavy Cavalry, lance)
Trebuchet/Bolt Thrower (Siege Engine)

Tier 3: "Late Medieval" for lack of a better term
Crossbow (Light Infantry)
Arquebusier (Light Infantry)
Cannon (Siege Engine)

-----------------

Unit stats:
Unit Types:

Archer Str 3 Mv 1
+25% vs. Heavy Infantry,
+25% vs. Light Cavalry

Spearman Str 4 Mv 1
+100% vs. heavy cavalry

Warrior Str 3 Mv 1
+50% vs. Spearman
+50% vs. Ancient Cavalry

Ancient Cavalry Str 4 Mv 2
+50% vs. Archers
-50% forest/jungle attack
-25% hill attack

Horse Archers Str 5 Mv 3
+25% vs. Heavy Cavalry
+25% vs. Heavy Infantry
25% Withdrawal Chance

Longbow Str 7 Mv 1
+50% vs. Heavy Infantry
+50% vs. Light Cavalry

Pikeman Str 7 Mv 1
+100% vs. Heavy Cavalry

Knight Str 9 Mv 2
+100% vs. Light Infantry
-95% attack forests/jungles
-25% attack hills

Crossbow Str 7 Mv 1
+50% vs Heavy Infantry
+25% City Defense
*Crossbows trade accuracy for penetrating power, meaning they lose any advantage over light cavalry. However, they also take little room to use effectively, and require little exposure of the shooter, making them exceptional for defending walls and fortresses.

Arquebusier Str ? Mv 1
+50% vs. Heavy Infantry
+50% vs. Light Cavalry

Siege Engineer Str 1 Mv 1
Blockades land tiles - see Circumvallation tech.

-----------------
So, a lot of the 'advancement' in units isn't really due to any inherent advantage of newer unit types over older units. Rather, it has to do with improvements in the same equipment. Ie, a medieval guy with a sword isn't actually all that much better than a roman legionaire. The armor is similarly protective, the sword is more or less the same, and so forth. If anything, the Roman Legionaire was actually better (the pilum was a beautiful piece of military engineering).

Going back further, the advantage of the medieval guy with a sword over the early classical guy with a sword is he has steel weapons vs. bronze or iron weapons, and steel armor vs. bronze or iron armor. He isn't really that much better trained, there's no amazing new combat doctrine he's taking advantage of, and his equipment is basically the same. Why introduce a new unit type to cover what's essentially the same guy with better equipment? Really, that guy with a sword is the basic warrior you started with.

Now, this will probably take more work than I imagine, but rather than needing to keep introducing new unit types, allow upgrades in weapon/armor technology to be handled via unit upgrades. And rather than the automatic upgrades to bronze/iron weapons like you have at present - make them pay to upgrade their weapons/armor, and let new units be built with them. You'll probably want to advance unit graphics some of the time, but not necessarily all of the time.

Ie, our basic warrior is Str 3 with clubs. You get bronze working and can upgrade him to bronze weapons/armor, for +1 str each. Later you get iron, then steel. Handle weapon/armor upgrades separately.

--------------

New or modified technologies for the tech tree:

Military Discipline: Rename of Warrior's Code, allows regular troop formations (ie, most military units), Archer/Warrior Str +1
Stirrup: All Cavalry Str +1
Lance: Allows Knights
Formation Drill: Allows Pikemen, +20% Fear/Terror Resistance to all units produced with Barracks
Siegecraft: Warriors gain +20% City Attack and +20% City Defense, + put Trebuchet/Bolt thrower here. Req Circumvallation.

Circumvallation: Allows Siege Engineer. Siege Engineer may blockade land tiles (radius 1) and prevents enemy units from entering a blockaded enemy city tile. (I don't know if this is possible or not, but i'm trying to model actual warfare here). Req Mathematics, Military Discipline.

-------------

What to do with Chariots?
WH makes chariots into awesome platforms of destruction. Real military history shows chariots as little more than conveyances for melee warriors, whose poor mobility, high cost, and low effectiveness relative to real cavalry doomed them to obscurity well before the end of the classical period. I suppose we have to side with WH over history here, and we can take some cues from the WH rules on how Chariots should be awesome.

Chariots in WH do impact hits, which is where they get most of their awesome. A chariot can thus often break a unit on the charge, forcing it to retreat, but will rarely destroy a unit itself. Beyond this, even the WH chariots rely on having a hero riding around on them to make them decent. I therefore suggest:

Chariot
Causes Fear
Causes Collateral Damage
Carrying Capacity 1 (hero)
(no access to withdrawal promotions)

 

[Squirrelloid]

Army rebuilding Example: Druchii

WH Druchii Army List

Infantry:
Warriors
-Spears
-Sword/Shield
-Repeater Crossbow
Corsair
Shades
Witch Elves
Executioners
Black Guard

Cavalry:
Dark Riders
-Repeater Crossbow option
Cold One Cavalry
Cold One Chariot

Creatures:
Harpies
Dark Pegasus
Manticore
Hydra
Dragon

Siege Engines:
Bolt Thrower
Cauldron of Blood

Other Army Advantages/Limitations:
No Heavy Armor
Fast

----------
Implement full Drill promotion line for all Druchii units. Start most Druchii units with Drill 1.

Warriors w/ Sword/Shield -> Warrior
Warriors w/ Spear -> Spearman
Warriors w/ Repeater Crossbow -> Repeater Crossbow (UU Crossbow +2 first strikes)
Corsair -> UU (warrior chassis) +20% City Attack (stacks with Siegecraft bonus), amphibious, and the naval ability they currently have
Shades -> UU Archer +25% forest defense
Witch Elves -> UU (warrior chassis) +3 poison, +2 first strikes (tech?)
Executioners -> UU (warrior chassis) +2 Str, +50% vs. Heavy Infantry (tech?)
Black Guard -> UU Pikeman, +50% vs. Heavy Cavalry, 100% Resist Fear/Terror, +1 Str, +1-2 first strikes

Cavalry:
Dark Riders -> UU (ancient cavalry chassis) +1 Mv, +25% withdrawal chance, +1 first strike chance, +1 Str (tech?)
Dark Riders w/ Repeater Crossbow -> UU (horse archer chassis) +2-3 first strikes, +2 str (tech?)
Cold One Cavalry -> UU Knights, +1 Str, Cause Fear
Cold One Chariot -> Chariot UU, +1 First Strike

Creatures:
No real opinions - except there should be a Hydra unit. Harpies could be the Ranger UU.

Siege Engines:
Bolt Thrower -> UU Trebuchet, +2 first strikes
Cauldron of Blood -> ... don't bother

Other Normal allowed units:
Ancient Cavalry, Horse Archers, Catapult, (maybe longbow?)

-----------

Obviously a lot depends on how metals end up working, and I have some ideas regarding the Druchii (who don't generally wear heavy armor, for example, so shouldn't be able to get the benefits of a 'heavy armor' upgrade).

It may also be obvious that I don't believe in "National Units" nonsense. You should field a diverse group of units because its advantageous to do so, not because you exhausted your limit for particular unit types. Especially as the limits don't seem to scale with world size, which has... problems.

 

[Psychic_Llamas]

A lot of very good observations and ideas in here, im interested to see what you come up with for the Druchii (and how much work will be involved in making such changes...)

 

[Masada]

More elegant than the unit proposal I had floating around in my head, good work. Just what I was after

Except for the techs , I'm fairly sure they could be wiggled in, and get played around with (its not hard to have a rename for existing techs which would be redundant under that scheme).

 

[Ahriman]

I don't think that this works either in this mod or in the civ stack-based combat system, and I recommend that we do not try to implement this. I also recommend that we do NOT try to completely rework the whole combat tree and redesign every faction.
You can't keep re-creating the entire mod from scratch or nothing will ever get done. I would much rather move forward with incremental changes to the current build than a complete unit redesign.
Any gains from a massive change like this I think are outweighed by the effort required to implement it.

First, a system of hard-counters doesn't work in civ, because the "best" defender always defends the stack. If you give every unit a hard counter (ie a unit that gets +100% or +50% strength vs it), then it becomes far too difficult to attack a stack.
You can create a mechanism that counters stacks (like spells that damage every unit in a single stack), but this will just screw the AI, since stacks are all the AI knows how to do - it won't know how to use the mechanic well, and it won't know how to avoid suffering from the mechanic.

Second, warhammer army design and tabletop mechanics (from what I understand - I am not familiar with tabletop) is not really based on a system of hard counters. Dwarves are basically all heavy infantry and some ranged units. They don't have heavy cavalry or light cavalry, so its going to be very hard to pigeonhole them into such a system. Similarly for many other factions (eg Skaven).

Third, I like the differentiation between separate defensive and offensive combat strength, which Civ4 removed and FFH restored.

Fourth, there are other roles for units; city attack units, city defense units, rangery wilderness terrain units, monsters, anti-monster units, etc.

Fifth, I don't see what the gains are from having a system where you have a single unit that "pays" to buy new promotions that increase its strength, as opposed to the existing system where you have higher level units which are stronger, and you can pay gold to upgrade the lower level unit to the higher level unit. You can imagine that this is just upgrading the same unit with better equipment/training if you like, but there is no need to make a mechanic like this.

The AI knows how to build higher level units and upgrade lower level units to higher level units. It doesn't know how/when to pay for promotions that increase the strength of an already constructed unit.

The siege engineer is a nice idea.

 

[Ajidica]

I like the idea and is meshes well with real military history. Only comment I have is what happened to Militia Spearmen/Archers/Swordsmen? Without them you loose alot of the diversity in war. In WArhammer (please forgive me for saying this if you already know how Warhammer armylists go) there are basicaly two lists, common warriors and uncommon (there may be more, but I have never played WHFB, only ancient battles). You have to choose between more numerous but weak common warriors or rare but powerful uncommon warriors. By retaining militia you can show that difference.

Side nit-pick: You said that Greek Phalanxs did poorly against the Legions, they actualy did very well against them if proper cavalry support. If you took Alexanders army and a general who only understood the theory, they would give the legions a very cloose battle. Lets try and not discuss who was better against who in this thread, lets try and keep it on topic.

 

[Squirrelloid]

I don't think that this works either in this mod or in the civ stack-based combat system, and I recommend that we do not try to implement this. I also recommend that we do NOT try to completely rework the whole combat tree and redesign every faction.
You can't keep re-creating the entire mod from scratch or nothing will ever get done. I would much rather move forward with incremental changes to the current build than a complete unit redesign.
Any gains from a massive change like this I think are outweighed by the effort required to implement it.

Most of these changes are actually rather easy to implement. The only hard part is the metals, and I'd actually encourage some discussion on that since I've mostly hinted rather than fully explained an idea. That and the Siege Engineer is going to be hard, and you liked that.

First, a system of hard-counters doesn't work in civ, because the "best" defender always defends the stack. If you give every unit a hard counter (ie a unit that gets +100% or +50% strength vs it), then it becomes far too difficult to attack a stack.

You can create a mechanism that counters stacks (like spells that damage every unit in a single stack), but this will just screw the AI, since stacks are all the AI knows how to do - it won't know how to use the mechanic well, and it won't know how to avoid suffering from the mechanic.

The mechanic is called collateral damage and the AI knows how to use it. If you make collateral damage dealing units unlimited the AI will have no problems. (National units are still a bad idea).

Second, warhammer army design and tabletop mechanics (from what I understand - I am not familiar with tabletop) is not really based on a system of hard counters. Dwarves are basically all heavy infantry and some ranged units. They don't have heavy cavalry or light cavalry, so its going to be very hard to pigeonhole them into such a system. Similarly for many other factions (eg Skaven).

Actually, WH does have hard counters, its just not nearly so obvious, and its slightly differently focused. Lets use the Druchii as an example. These are some of the hard counters:

Executioners are a hard counter to heavy armor - sure, they also work against light armor, but you're wasting their potential. And they die horribly to anything with a larger charge range (ie, anything they can't charge).

Witch Elves are a hard counter to massed blocks of lightly armored infantry, because they get so many attacks.

Spears are an obvious hard counter to cavalry - sure, its not a guaranteed win, but the cavalry are much more likely to lose if they charge spears than not.

Dark Riders effectiveness in the actual game depends on teamwork (they rear charge a unit that gets charged by something else in the same turn for maximal effectiveness). However, they are also a hard counter for light infantry and siege engines.

Repeater Bolt thrower warriors are a hard counter for heavy infantry - they can run away and shoot.

Third, I like the differentiation between separate defensive and offensive combat strength, which Civ4 removed and FFH restored.

You talk about poor models, but this is and always has been a poor model. There's a reason why CivIV removed it. When you attack a stack it isn't necessarily only defending - you're simulating an entire battle in a few rolls of the dice, it may well advance as well as entrench and defend. The game is insufficiently granular to make differentiating attack from defense worthwhile except on a very gross level. (A few units that get +% attack/defense). Further, the strength of a unit represents its base strength - CivIV recognized that it was better to just multiply unit strength by appropriate multipliers instead of tracking two numbers.

Just because FfH did it doesn't mean its a good idea.

Fourth, there are other roles for units; city attack units, city defense units, rangery wilderness terrain units, monsters, anti-monster units, etc.

(1) This is what unit promotions are for, and (2) Non-pike melee units eventually become this in the proposal (City attack/defense). Not to mention crossbows. There are no specialty city attack/defense units in the ancient/early classical era because no one fought battles attacking cities. Either you laid siege to a city (thereby causing them to fight or surrender) or you bypassed the city. There is not one famous siege pre-Rome which involves an attack on a city (at least in the Western world - don't know about China offhand) - and the the dedicated siege specialist units were the roman siege engineers and catapults. Even then, sieges still generally ended in the surrender of the garrison rather than physically taking the town.

I also didn't talk about rangers because they are an irregular unit - I don't care and don't mind if they continue in their current functionality.

Monsters are just going to be big beefy number sacks, and anti-monster units should be heroes. Which is mostly how it is in the real game.

Fifth, I don't see what the gains are from having a system where you have a single unit that "pays" to buy new promotions that increase its strength, as opposed to the existing system where you have higher level units which are stronger, and you can pay gold to upgrade the lower level unit to the higher level unit. You can imagine that this is just upgrading the same unit with better equipment/training if you like, but there is no need to make a mechanic like this.

The AI knows how to build higher level units and upgrade lower level units to higher level units. It doesn't know how/when to pay for promotions that increase the strength of an already constructed unit.

Lets have a discussion on exactly how metals should work, perhaps? I certainly don't want to break the AI.

The siege engineer is a nice idea.

I am mostly amused that you like the idea which is going to be hardest to implement.

(BTW, it would also be nice if city governors could surrender a city under certain conditions to make historical-style sieges actually work - such as Governor surrenders if city has starved n times and has no trade route with capitol).

 

[Squirrelloid]

I like the idea and is meshes well with real military history. Only comment I have is what happened to Militia Spearmen/Archers/Swordsmen? Without them you loose alot of the diversity in war. In WArhammer (please forgive me for saying this if you already know how Warhammer armylists go) there are basicaly two lists, common warriors and uncommon (there may be more, but I have never played WHFB, only ancient battles). You have to choose between more numerous but weak common warriors or rare but powerful uncommon warriors. By retaining militia you can show that difference.

Well, I mostly plan on warriors/archers/spearmen upgrading into those positions. I mean, they basically do anyway, which is why I talk about Chasses. We formalize the metals as a source of promotions, and note that Military Discipline gives a bonus to the str of archer and warrior units. If you want to consider the post Military Discipline warrior and archer units 'new' units that are the militia ones - be my guest.

Anyway, the second post defines a number of Chasses to cover general unit types. We can then specify special instances of these types by adding modifiers onto them - see Corsairs, Witch Elves, Executioners, etc... in my Druchii rebuild example. They are UUs that use the same chassis as the warrior, but they have additional abilities to reflect their specialized roles.

Of course, the choice between few and many should come down to hammer investment. Specialized UUs should cost more hammers than the basic model and thus you will have less of them if you build them instead of the basic model.

Side nit-pick: You said that Greek Phalanxs did poorly against the Legions, they actualy did very well against them if proper cavalry support. If you took Alexanders army and a general who only understood the theory, they would give the legions a very cloose battle. Lets try and not discuss who was better against who in this thread, lets try and keep it on topic.

Do you have a battle in particular you're thinking of?

I'm referring specifically to unit vs. unit performance. The legionaires did have a functional superiority to the greek phalanx (two actually - the pilum was hell for the phalanx). In CivIV terms, if your sword infantry attacked a stack containing spearman and cavalry, you'd end up attacking the cavalry (who have an advantage vs. sword infantry), so we model the situation you describe perfectly.

Considering Alexander's use of heavy cavalry was as a limited strike element to quickly get additional warriors into the worst of the fighting - ie, basically a mobile reserves - it would be hard to say that greek generals considered the cavalry a 'support unit' for the phalanx so much as quick heavy infantry.

It was the roman's who came up with the innovation that the general stands back and watches the battle so he can issue orders during the battle, so the coordinated use of support units with other units is probably also a Roman invention (without a commander capable of making decisions after the battle starts its hard to exert that much control over unit performance in a battle. Pre-roman combat was mostly a plan laid out beforehand and then the general only had command of his unit and whatever units he was within shouting distance of).

 

[Ajidica]

Your version of militia sounds nice.
On the Phalanx/Legion: Lets sort of drop this debate. We each know to much and too little of what we are talking about.

 

[Ahriman]

I note that you neglected to reply to the single most important point in the thread; that this kind of unit design is completely incompatible with the way that combat works in civ; that the attacker is faced off by the strongest defender in the stack. So a system where every unit has a very hard counter fails completely and makes offense nearly impossible (and makes it very easy to whip the AI, who will suicidally throw units against your superstack).

Most of these changes are actually rather easy to implement.

This would require an entire redesign of every faction's unit roster, stats, and the tech tree. Is there anything else in this mod? Oh, art. Yeah, you want to change the art too.

That and the Siege Engineer is going to be hard, and you liked that.

Mostly I was desperately looking for something to be positive about - avoid too much negativity and all
Its an interesting idea, but would be difficult to implement.

The mechanic is called collateral damage and the AI knows how to use it.

Not really. The AI doesn't use catapult/cannon bombardment effectively (in order to be able to preserve their units) and doesn't use suicidal catapults as stack weakeners effectively either.

(National units are still a bad idea).

Eh? Unit caps work really well in this mod. They force combined arms by limiting the spammability of potential units (eg even though Estalia has awesome pikemen, they can't have them everywhere), they simulate the restrictions in the tabletop game of having limited number of special and rare unit slots in an army, they make core militia units never become obsolete, they provide a reverse slippery slope mechanic where larger empires have their elite units stretched thinner and thinner, they prevent deliberately powerful but rare units from becoming commonplace (eg we don't want giants or griffons everywhere). National unit implementation works very well in this mod.

Actually, WH does have hard counters, its just not nearly so obvious, and its slightly differently focused. Lets use the Druchii as an example.

There is a difference between "hard counters" and "soft counters". A hard counter is one that is extremely effective against one unit type but almost completely ineffective against others. Pikemen and helicoptors are probably two of the closest examples from vanilla civ, but even those aren't too close. I don't know much about the dark elf unit roster and abilities; I suspect though they have a similarity to eldar (and presumably dark eldar) in 40k. Eldar in 40k (based mostly on playing and comments I read from early versions of DOW) are a perfect example of a hard-counter philosophy. Warpsiders are great vs light infantry but almost completely ineffective against anything with armor. Fire dragons are great vs vehicles and heavy armor, but ineffective against anything else. If eldar are anything to go by, then high elves and dark elves are the exceptions that prove the rule. Spearmen for most factions aren't ineffective against anything other than cavalry. Some factions are nearly entirely melee infantry, but still aren't totally hosed by missile units. By and large, warhammer units aren't hard counters.

By and large, civ doesn't use hard counters, it uses softer counters. Most units aren't specialists, they're generalists. Even most of the speciailsts are still generally useful; War elephants are very effective vs cavalry, but are still effective against other units too. This is partly because civ designers realised that hard counters = fail with the stack-based combat model.

You talk about poor models, but this is and always has been a poor model. There's a reason why CivIV removed it

Actually, there was widespread outrage that that this feature was removed. There are some units that are inherently more effective on offense or defense. Armored units are much more effective on offense, where their mobility and armored spearhead tactics are more valuable. Machine guns are much more effective on defense. Archer units are much more effective defenders (ever tried firing arrows uphill?) than attackers. Pikemen are fundamentally a defensive unit (only a few nations ever managed to develop effective offensive pikeman tactics). And attacking with units in a turnbased game represents a strategic and partly tactical level offensive, so it makes sense to allow some units to be more effective on attack than defense.

CivIV recognized that it was better to just multiply unit strength by appropriate multipliers instead of tracking two numbers.

Just because FfH did it doesn't mean its a good idea.

Civ4 implemented it, that doesn't make it a good idea. Just because FFH did it doesn't make it a *bad* idea. I think there are an awful lot of people who think that civ3 unit rosters and combat worked better than the civ4 system.

Either you laid siege to a city (thereby causing them to fight or surrender) or you bypassed the city

Civ, like most games, fails to simulate siege warfare, partly because it knows that player prefer to fight and destroy units. To capture cities in the civ engine you have to have your units attack and kill all the units in the enemy city, and the civ AI defends cities, it doesn't really defend territory.
This is a game, not a historic simulator, and we are forced to use the civ engine. That means that city assaults are inevitably a big part of the game. You can't wish away the restrictions of the game engine. I would rather have a game that is fun to play than one that is historically accurate. Hell, even the Total War games (rome and medieval 2 in particular), which are relatively more historically accurate than other non-grognard strategy games make city assaults a key part of their mechanic.

(BTW, it would also be nice if city governors could surrender a city under certain conditions to make historical-style sieges actually work - such as Governor surrenders if city has starved n times and has no trade route with capitol).

The AI wouldn't understand this. I think that creating a realistic medieval warfare mod is not the role for this mod. If you want a historically accurate mod, I'm all for it, but I don't think thats what Warhammer is about. The idea of capturing cities by passively sitting around them for a while with your unassailable superstack (with hard counters for anything they could throw at you) is anathema to warhammer; if you want the city, you should have to spill a lot of blood.

and anti-monster units should be heroes. Which is mostly how it is in the real game.

I think there is a strong argument for having hunter/ranger/beastmaster units get bonuses vs monsters, as well as dwarven trollslayers/giantslayers.
Maybe this should be removed from knights however.
A player might be able to field a dozen monsters, and the 1-2 heroes an enemy gets aren't really going to be an effective counter.

Well, I mostly plan on warriors/archers/spearmen upgrading into those positions.

They can do this now, and the AI understands how to upgrade them. Why change this?
Also, they tried a "chassis with mods" system in the Planetfall mod (trying to recreate SMAC) and it failed; they ripped it out.

They are UUs that use the same chassis as the warrior, but they have additional abilities to reflect their specialized roles.

What is the possible design reason for not just having these all be individual UUs, as they are now? Why do you want to change the system?

The easiest way to merge the current system with what you want is this;
Have unit lines of melee units (general combat and spearmen), archer units, recon units, cavalry units (knights, "light" cav and horsearchers), chariots and monsters. There are 1-3 tiers of units in each category, each a separately defined unit in the mod. The lowest 2 tiers are non-national units. The highest tier are national units with unit caps.
This basically exists already.
Now, it seems you want more in the way of rock/paper/scissors between categories. You can do this by adding some x% strength bonus vs unit type y for various categories of units. Eg: cavalry units gain +25% vs archer units. Some of these is already built into the mod (crossbows get bonus vs melee, chariots get bonus vs melee, spearmen get bonus vs mounted, etc.). These bonuses should not be too large, because of the way that stack combat works in civ.

There is no reason to throw out the current design system. If you want a little more rock/paper/scissors the current units can be tweaked a little more that way, but there is actually already a fair bit of that in the game.
Swordsmen > spearmen. Spearmen > cavalry. Cavalry > archers. Chariots > swordsmen. Crossbows > melee. Recon is good in terrain and vs monsters. Monsters = strong but limited numbers.

I think the current metal system works fine for melee units and cavalry units, though I'd also like to get meteoric iron implemented and maybe a steel weapons bonus implemented.

 

[Squirrelloid]

I note that you neglected to reply to the single most important point in the thread; that this kind of unit design is completely incompatible with the way that combat works in civ; that the attacker is faced off by the strongest defender in the stack. So a system where every unit has a very hard counter fails completely and makes offense nearly impossible (and makes it very easy to whip the AI, who will suicidally throw units against your superstack).

...

Civ IV ancient units:
Axeman +50% vs. melee. Hard counter Spearman. (Soft counter axeman)
Spearman +100% vs. mounted. Hard counter mounted.
Chariot +100% attack Axeman. Hard Counter Axeman.
Archer. +50% City Defend +25% Hill Defend. Soft Counter Axeman.

Yup, the normal game has absolutely no hard counters... wait.

Yup, the AI fails completely in the normal game... wait.

Oh yeah, neither of those statements are true. The AI can handle hard counters.

This would require an entire redesign of every faction's unit roster, stats, and the tech tree. Is there anything else in this mod? Oh, art. Yeah, you want to change the art too.

Wait, what's wrong with the current art? Really, it doesn't necessarily involve a completel unit roster redesign, as it arrives at more or less the same place (end units are the WH army roster, with some ancient units as placeholders to get there) - it does involve some mechanics redesign, which is easy because that's just changing the unit spec files. It also involves re-arranging tech requirements - also pretty easy. These are changes I could learn how to do and implement in an hour or two.

Now, playing with metals may take more work - I'm not sold on any one system yet.

Not really. The AI doesn't use catapult/cannon bombardment effectively (in order to be able to preserve their units) and doesn't use suicidal catapults as stack weakeners effectively either.

Have you played against the BTS AI? (Is the mod using it?) Because the BTS AI is amazingly good at using suicide catapults - ok, before they prohibited siege units from participating in amphibious assaults (most recent patch) there were some issues there because the AI only seems to 'see' one transport's compliment of units at a time when making attack decisions. But I've had a computer suicide 12+ cannons into me to soften up a city for its other units. If the mod isn't using it then it needs to upgrade to the BTS AI, which is vastly superior to prior versions.

Eh? Unit caps work really well in this mod. They force combined arms by limiting the spammability of potential units (eg even though Estalia has awesome pikemen, they can't have them everywhere), they simulate the restrictions in the tabletop game of having limited number of special and rare unit slots in an army, they make core militia units never become obsolete, they provide a reverse slippery slope mechanic where larger empires have their elite units stretched thinner and thinner, they prevent deliberately powerful but rare units from becoming commonplace (eg we don't want giants or griffons everywhere). National unit implementation works very well in this mod.

Monsters maybe deserve limited units, but the others don't. Combined arms should be encouraged, not forced. And militia units should remain useful because they are much cheaper than other options, not because you run out of units to build.

National units scale poorly with world size as well, because you get the same number regardless of size.

Finally, simulating army list restrictions is silly because that should happen at the stack level, not the civilization level.

There is a difference between "hard counters" and "soft counters". A hard counter is one that is extremely effective against one unit type but almost completely ineffective against others. Pikemen and helicoptors are probably two of the closest examples from vanilla civ, but even those aren't too close. I don't know much about the dark elf unit roster and abilities; I suspect though they have a similarity to eldar (and presumably dark eldar) in 40k. Eldar in 40k (based mostly on playing and comments I read from early versions of DOW) are a perfect example of a hard-counter philosophy. Warpsiders are great vs light infantry but almost completely ineffective against anything with armor. Fire dragons are great vs vehicles and heavy armor, but ineffective against anything else. If eldar are anything to go by, then high elves and dark elves are the exceptions that prove the rule. Spearmen for most factions aren't ineffective against anything other than cavalry. Some factions are nearly entirely melee infantry, but still aren't totally hosed by missile units. By and large, warhammer units aren't hard counters.

Do you own a single WH army factionbook? Have you played a game? WH has a lot more hard counters than 40k because movement is more important and weapon/armor types are more important. 40k is like warhammer for dummies (yes, I play it as well, its a lot simpler). There are no Space Marines (good against everything) units in WH.

And just about every army has cavalry or a unit which functions like cavalry. For example, Ogres move almost as fast as cavalry, giving them advantages similar to cavalry against most opponents.

By and large, civ doesn't use hard counters, it uses softer counters. Most units aren't specialists, they're generalists. Even most of the speciailsts are still generally useful; War elephants are very effective vs cavalry, but are still effective against other units too. This is partly because civ designers realised that hard counters = fail with the stack-based combat model.

See ancient era analysis above. Also: Maceman +50% vs. melee. Crossbow +50% vs. melee. Pikeman +100% vs. mounted. Grenadier +50% attack rifleman. Rifleman +25% vs. mounted. Infantry +25% vs. gunpowder. Anti-tank/gunship +100% vs. armor. Marines +50% vs. machinegun. I could go on - Civ IV is chock full of hard counters. Mounted units are even hard counters (to siege units) because of flank attacks.

In fact, its hard to find military units that *aren't* a hard counter to something (those would be warrior, archer, longbow, siege units, musketman, paratrooper, tank, mechanized infantry, modern armor), and many of those are 'hard counters' to city attacks/defense. (Warrior, Archer, Longbow are city attack counters. Siege units and armor are city defense counters).

Actually, there was widespread outrage that that this feature was removed. There are some units that are inherently more effective on offense or defense. Armored units are much more effective on offense, where their mobility and armored spearhead tactics are more valuable. Machine guns are much more effective on defense. Archer units are much more effective defenders (ever tried firing arrows uphill?) than attackers. Pikemen are fundamentally a defensive unit (only a few nations ever managed to develop effective offensive pikeman tactics). And attacking with units in a turnbased game represents a strategic and partly tactical level offensive, so it makes sense to allow some units to be more effective on attack than defense.

Units which are more effective on the defense because they occupy good defensive terrain - something which Civ IV models quite well. Archers in open plains suck defensively.

Pikemen are amazing offensive units, who were implemented with great skill by the Swiss, Germans (not necessarily as one country - different factions during 30 years war for example), Dutch (see War of the Spanish Succession), and that's just counting medieval/renaissance europeans. The Phalanx is a pikeman precursor which was used offensively by the Greeks and the Persians. And the pike was certainly known in China although I'm not nearly as familiar.

The only reason more nations didn't develop effective pikemen are generally social - France didn't want to train and arm their peasants for example. Further, they were aided by the ready availability of mercenary pikeman companies - making the capability available without the need to train or arm their own populace. Countries which did develop effective pikemen tended to have a freer peasantry population who was often required to train for military service, and generally lacked a strong chivalric tradition. Ie, those countries which thought of war as a noble's sport didn't make efforts to train non-nobles in warfare (eg, France).

I think there is a strong argument for having hunter/ranger/beastmaster units get bonuses vs monsters, as well as dwarven trollslayers/giantslayers.
Maybe this should be removed from knights however.
A player might be able to field a dozen monsters, and the 1-2 heroes an enemy gets aren't really going to be an effective counter.

Of course, the level 1 monsters are pretty weak, meaning there are 4 meaningful monsters running around. That said, some factions do clearly have units which are superb at taking down monsters, and those can be implemented as UUs. I also don't object to hunters/rangers/beastmasters being monster counters - these aren't exactly on my radar.

They can do this now, and the AI understands how to upgrade them. Why change this?
Also, they tried a "chassis with mods" system in the Planetfall mod (trying to recreate SMAC) and it failed; they ripped it out.

I think you're making a mountain out of a molehill here. By 'Warrior Chassis' I mean it has all the advantages the warrior gets. If you want to make a unit upgrade which is nothing more than a Str increase instead of increasing warrior strength, well that's really just the same thing. By chassis I'm referring to a style of unit moreso than anything else.

What is the possible design reason for not just having these all be individual UUs, as they are now? Why do you want to change the system?

They are still individual UUs, they're just based on a particular chassis. I don't mean to imply that they necessarily obsolete the normal unit type.

The easiest way to merge the current system with what you want is this;
Have unit lines of melee units (general combat and spearmen), archer units, recon units, cavalry units (knights, "light" cav and horsearchers), chariots and monsters. There are 1-3 tiers of units in each category, each a separately defined unit in the mod. The lowest 2 tiers are non-national units. The highest tier are national units with unit caps.
This basically exists already.
Now, it seems you want more in the way of rock/paper/scissors between categories. You can do this by adding some x% strength bonus vs unit type y for various categories of units. Eg: cavalry units gain +25% vs archer units. Some of these is already built into the mod (crossbows get bonus vs melee, chariots get bonus vs melee, spearmen get bonus vs mounted, etc.). These bonuses should not be too large, because of the way that stack combat works in civ.

Well, I think we need to formally differentiate Light and Heavy cavalry as different unit classes - they perform fundamentally different roles and are effective against different unit types. We don't consider all infantry the same unit class with good reason. The differences between light and heavy cavalry are equally profound.

R/P/S style mechanics are (1) historically realistic, (2) represent WH rules fairly well assuming mildly competent generalship, (3) actually encourage combined arms because every unit type is valuable. National units don't encourage combined arms, they just force you to build something different when you've exhausted your cap on particular types. You won't see combined arms, you'll see generally focused stacks limited in size by national caps.

If anything, national unit limits screw the computer up the most because they can't build sufficient numbers of some unit types (ie, siege engines).

 

[Darkform]

I think we could use the HI LI HC LC thing but with soft counters just hard enough that they canoutmatch what they counter from the same tier by 10% at what they are good at that is with counting their highest score be it offense or Defense. so a (makeing these up) 3/4 spearman wolud have have a +10% against calvary if the same tier calvary was 4/3. and if the Calvary was 6/3 and same tier spearman still 3/4 he would have +60% against calvary.

this way we have the counters but soft enough that after they win they are wounded to bad to counter the 2nd calvary unit that attacks that turn.

I'm thinking we almost have this already.

what I'm saying is we should have soft counters but not so soft that they lose aginst what they counter on the same tier.

 

[Darkform]

ya Siege weapons even tho rare in WH need to be uncaped in civ mod. the AI doesn't understand spell very good so it won't use spells that bambard as good as it can use siege.

also siege is a hard counter to chariots and war chariots in WH. 1 hit by a siege weapon on 1 of those is instant death no questions no save. since we are making chariots powerfull like WH we need there counter able to counter them.

 

[Psychic_Llamas]

You can't keep re-creating the entire mod from scratch or nothing will ever get done. I would much rather move forward with incremental changes to the current build than a complete unit redesign.
Any gains from a massive change like this I think are outweighed by the effort required to implement it.

i agree with this, the number of times weve reworked stuff is kinda silly... id rather not have to do that again :/ (although i admit ive played around with units myself to see what i could come up with, what we have going atm works pretty well, and subtle changes are all that really needed i think)

im not going to bother replying to *every* little statement in this thread, i have uni in a few mins, but id just like to say that there are some good ideas here that we can take under consideration (ie the placement of different unit types in the tech tree and how the progress of military power is through finding counters to the current dominating tactic (development of the lance gave cavalry dominance over infantry lacking pike weapons))

as another note: i dont want warhamemr to be too set in real world history. afterall this IS a fantasy mod.

also, id just like to say: calm down guys getting a bit heated in here

 

[Ahriman]

Civ IV ancient units:
Axeman +50% vs. melee. Hard counter Spearman. (Soft counter axeman)
Spearman +100% vs. mounted. Hard counter mounted.
Chariot +100% attack Axeman. Hard Counter Axeman.
Archer. +50% City Defend +25% Hill Defend. Soft Counter Axeman.

Civ4 has some hardcounters in the ancient era, but mostly moves away from them after that with only a handful of exceptions (pikemen, grenadiers, helicoptor gunships)

Units in civ that aren't hardcounters for anything (off the top of my head, haven't played it for a while) swordsmen, horsemen, knights, curaissers, cavalry, archers, longbowmen, musketeers, riflemen, infantry, mechinf, modern armor, the various siege units... almost everything. The hard-counters only exist in the very early game, when; a) stacks are small b) strategic resources are limited (so in ancient era you will likely to have access to elephants and cavalry and iron weapons).

Hard counters mostly don't exist after the ancient era, once stacks would be large enough for them to become problematic. There also aren't really that many different units at each era stage in civ.

Really, it doesn't necessarily involve a completel unit roster redesign, as it arrives at more or less the same place (end units are the WH army roster, with some ancient units as placeholders to get there)

So why change anything? You can slightly increase the degree that units are counters within the existing system.

And militia units should remain useful because they are much cheaper than other options, not because you run out of units to build.

Civ as a game never really makes it valuable to build stronger/cheaper units because of the way that units that win combats can costlessly heal (and because upkeep costs are on a per unit basis). I guarantee that if you remove unit caps on longbowmen, crossbowmen, knights, war chariots and pikemen that no-one will build militia swordsmen, militia spearmen, mililtia bowmen, light cavalry or chariots ever again.

Finally, simulating army list restrictions is silly because that should happen at the stack level, not the civilization level.

There is no way feasible way to limit the numbers of particular types of units in a stack.
But the limit works at a national level; armies throughout history have always been primarily made up of grunt troops. Any game in which you remove a limit on the number of elite troops that can be made, you find that people start fielding armies composed entirely of elite units.

Have you played a game?

I freely admit ignorance of the details of tabletop rules and unit stats.

Pikemen are amazing offensive units, who were implemented with great skill by the Swiss, Germans (not necessarily as one country - different factions during 30 years war for example), Dutch (see War of the Spanish Succession)

Swiss, German states, and spanish. But only for a narrow window; mostly pikes were there just to protect other troops (particularly gunpowder units) from cavalry.

Of course, the level 1 monsters are pretty weak

Trolls are pretty strong (strength 7 vs strength 4-6 axemen). But I'm not particularly worried about this. Partly its just for flavor.

By chassis I'm referring to a style of unit moreso than anything else.

Why not just make this a unit class, and use all the existing units? And add a couple of units here and there if you really think they're necessary.

If anything, national unit limits screw the computer up the most because they can't build sufficient numbers of some unit types (ie, siege engines).

Removing unit caps on siege units helps the player more than the AI, because the AI uses the (very very powerful) bombardment abilities ineffectively.

Basically, I think we can achieve a compromise solution. Keep most of the units and unit lines as they are. Swordsmen/spearmen/bowman/shockcavalry/missile cavalry/chariot/recon/monster. Give some of these unit lines inherent bonuses vs other units; so spearmen line units get bonuses vs heavy cavalry and chariots, swordsmen get a cityattack OR melee unit bonus (so maybe swordsmen and spearmen should have the same base unit strength, but swordsmen get a bonus vs melee units), bowmen get hills and city defense bonuses, heavy cav get bonus vs melee units and archers (problem with cav vs archers is that it makes cavalry city attackers, which is kinda stupid), missile cav get bonus vs spearmen and maybe chariots, recon get terrain and withdrawl bonuses (see my thread on hunter/ranger redesign), monsters are just strong.
None of these bonuses should be more than 50%. Most should be 20-25%.

All the various UUs still have a unit "type", they are still a melee or archery or cavalry or whatever unit, so still get the various bonuses/penalties vs other types.

So we can increase the rock/paper/scissors system, but do so without having to increase the base strength of units by paying for a +1 strength promotion from the stirrups tech; instead just have a more powerful unit at the stirrups tech that the older one can increase to.

If we assign the current units to tiers:
0. warrior/scout
1. axeman warband, spearman warband, archer warband, ancient cav, hunter, chariot, troll
2. Ranger, militia swordsman, militia spearman, militia bowman, light cav, horsearcher, giant
3. cavalry, war chariot, knight, pikeman, longbowman, crossbowman, handgunner, beastmaster.
(3.5 Archmage, steam tank, dragon).

Then we already have a unit in almost every slot at each tier. We could add a missile cav unit at tier 1, but otherwise we basically have the units we want.

So all we need to do to the current system is add maybe a couple of units, and add maybe some soft-counter bonuses vs particular unit types to a few more units.
Many of these already exist. The only things we would need to change are:
Give axemen swordsmen a bonus vs melee units and/or city attack.
Give missile cav units a bonus vs chariots and melee units.
Give shock cav a bonus vs archer units.
Give spear units a small bonus vs chariots (to counteract the chariot bonus vs melee).
Increase spearmen so they have the same base strength as axemen/swordsmen (and costs equal too).

 

[Masada]

i agree with this, the number of times weve reworked stuff is kinda silly... id rather not have to do that again :/ (although i admit ive played around with units myself to see what i could come up with, what we have going atm works pretty well, and subtle changes are all that really needed i think)

Yes, although more than a few of the earlier changes were done against my protests... all my beautifully balanced civics were played with, etc . PL had similar woes. So all the constant changes is really just to fix a series of fundimentally badly designed concepts which worked there way into the mod...

The units kind of work as is, but I do think a small inherant bonus would be best to certain classes of units... maybe swords could be given bonuses against units and not city attack bonuses (never figured that out myself).

 

[Ahriman]

The units kind of work as is, but I do think a small inherant bonus would be best to certain classes of units... maybe swords could be given bonuses against units and not city attack bonuses (never figured that out myself).

Sounds like once again we can fit different design principles into a workable compromise. Go team!

 

[Masada]

The key is to avoid compromising on the critical stuff, one of the reasons why we don't want massive bonuses is that attacking any defending stack is going to require probably at minimun a 4 to 1 ratio... having 25% or so requires numbers of perhaps 1.5 to 1. But we want bonuses, that work in line with warhammer canon and common sense, and this works, I would also take note of the tech proposals here,

Military Discipline: Rename of Warrior's Code, allows regular troop formations (ie, most military units), Archer/Warrior Str +1
Stirrup: All Cavalry Str +1
Lance: Allows Knights
Formation Drill: Allows Pikemen, +20% Fear/Terror Resistance to all units produced with Barracks
Siegecraft: Warriors gain +20% City Attack and +20% City Defense, + put Trebuchet/Bolt thrower here. Req Circumvallation.

There's some very good stuff in there, progressive unit bonuses for a whole type of class of unit is a good idea, considering we don't have many unit upgrades to start with. And considering we are likely to end up fewer techs we can certainly rename or alter existing techs which have substantionally the same purpose and incorporate some additional techs in the middle-late part of the tech tree.

 

[Ahriman]

The key is to avoid compromising on the critical stuff, one of the reasons why we don't want massive bonuses is that attacking any defending stack is going to require probably at minimun a 4 to 1 ratio... having 25% or so requires numbers of perhaps 1.5 to 1.

Exactly. This is my concern with implementing hard counters (+100% bonuses = bad).

There's some very good stuff in there, progressive unit bonuses for a whole type of class of unit is a good idea, considering we don't have many unit upgrades to start with. And considering we are likely to end up fewer techs we can certainly rename or alter existing techs which have substantionally the same purpose and incorporate some additional techs in the middle-late part of the tech tre

I disagree; I don't want units to bet getting +1 strength "upgrades" from researching a tech. Its much better to just have that tech provide a separate unit with higher strength. So, don't have an ancient cavalry unit that gets a +1 strength promotion with stirrups; instead, have an ancient cavalry unit, and a light cavalry unit available with stirrups that has a higher base strength, as in the current design.
The AI understands how to do this.

I don't think that a fear/terror "resistance" would be easy to code or very effective. I would prefer that fear/terror work on every unit except those with a "courage" type promotion (or "immune to psychology/mind effects). Maybe the senior officer promotion could make units immune to fear?
I dislike the idea a tech that gives all units a +20% city attack/defense bonus. That is a very large effect (2 levels!), and I don't see why we should give free promotions from techs.
Instead, just have these available through the existing promotions. We could require a tech for city raider/garrison 2 or 3. That would be fine.