circle of arms

[elanthis]

A very similar game to Wesnoth called Fire Emblem (one version of which is on the Game Boy Advance) has a concept of a circle (triad, in that game) of arms. This is only for melee weapons, and does not affect ranged weapons like bows or magic.

The way it works is that each weapon type is strong against one other weapon and weak against another. In Fire Emblem, the three types of melee weapons were axe, sword, and lance. Axe trumped lance, lance trumped sword, and sword trumped axe. For Wesnoth you might want a larger circle, since there are more melee weapon types like claws and clubs and so on.

I actually think this is more in line with the KISS principle of Wesnoth than the current resistance system. In the current system, each unit needs explicit resistances to be set. With the circle of arms system, instead of specifying a ton of resistances for each unit, you instead base damage bonuses on the weapon used - units with multiple weapons would get the benefit of the weapon being used, for example. That's less information per unit, less balancing to do per unit, and a simpler over-all algorithm. Of course, at this stage in development, Wesnoth is rather dedicates to the current resistance system. Might be fun to try it, though.

[cobretti]

A very similar game to Wesnoth called Fire Emblem (one version of which is on the Game Boy Advance) has a concept of a circle (triad, in that game) of arms. This is only for melee weapons, and does not affect ranged weapons like bows or magic.

The way it works is that each weapon type is strong against one other weapon and weak against another. In Fire Emblem, the three types of melee weapons were axe, sword, and lance. Axe trumped lance, lance trumped sword, and sword trumped axe. For Wesnoth you might want a larger circle, since there are more melee weapon types like claws and clubs and so on.

Such 'circle' has an implicit nonsense, IMHO. For example, how, except against very thick/metal lances, would a sword be worst than an axe? And with metal lances, why is anyway better an axe? You still can't cut it. Also, every such weapons have their own fighting style, it's not so easy to compare them.

I actually think this is more in line with the KISS principle of Wesnoth than the current resistance system. In the current system, each unit needs explicit resistances to be set. With the circle of arms system, instead of specifying a ton of resistances for each unit, you instead base damage bonuses on the weapon used - units with multiple weapons would get the benefit of the weapon being used, for example. That's less information per unit, less balancing to do per unit, and a simpler over-all algorithm. Of course, at this stage in development, Wesnoth is rather dedicates to the current resistance system. Might be fun to try it, though.

Well, it also leads to different matters: Whenever two units share the same weapon type, and are attacked by the same weapon type, it must have the same effect on them, no? Imagine that the attacked units are: Heavy Infantry and Footpad against Sword (blade), Dwarvish Stalwart and Horseman against Lance (pierce), Dwarvish Ulfserker and Skeleton against Lance (pierce)...

There are many other possible examples of clear disfunctions this method leads to, by the way. With the current system, there is a lot of room to specify the behavior of units under attack. With a 'circle of arms' it's just more like a rock-paper-scissors game. I don't know if better or worse, but clearly different.

[Dave]

I actually considered a system like this in the initial design stages of Wesnoth. One of the games that was inspiration for Wesnoth, Warsong, used it (it had a infantry > archers > horsemen > infantry... system).

However I rejected it mainly because frankly, I think it's been done to death. One of the aims of Wesnoth was to be a little different and fresh compared to most strategy games, and so I wanted to use a system a little deeper than the 'rock-scissors-paper' system so many strategy games use. I wanted, for instance, some units to be strong against certain other units in certain contexts, but weak in other contexts.

I think the Wesnoth system works pretty well....

David

[elanthis]

Such 'circle' has an implicit nonsense, IMHO. For example, how, except against very thick/metal lances, would a sword be worst than an axe? And with metal lances, why is anyway better an axe? You still can't cut it. Also, every such weapons have their own fighting style, it's not so easy to compare them.

I never said the Fire Emblem triad made sense. It was just an example. Peronsally, I'd say lance is weak against polearm (a lance is a very, very specialized form of polearm that's only useful in a charge on horseback - in Fire Emblem, all polearm users got grouped into Lance, which was rather dumb IMO).

So far as style, I agree. While individual sword styles are relatively meaningless (they all were based on the same set of strengths and weaknesses of the weapon in use, after all), some larger classes of styles had very different interactions between one another. A game system I'm working on actually bases most combat calculations off of style, and seems to work pretty well. Admittedly, it's geared towards role-playing combat, not strategic combat.

Well, it also leads to different matters: Whenever two units share the same weapon type, and are attacked by the same weapon type, it must have the same effect on them, no? Imagine that the attacked units are: Heavy Infantry and Footpad against Sword (blade), Dwarvish Stalwart and Horseman against Lance (pierce), Dwarvish Ulfserker and Skeleton against Lance (pierce)...

Again, it would need to be broken up more than the triad. Long blades vs short blades (very different styles of use), etc.

If the number of weapons are too large, a circle may very well be ineffective. A "network" might be a bit better. Each weapon has, say, two weapons its weak against and two its strong against.

There are many other possible examples of clear disfunctions this method leads to, by the way. With the current system, there is a lot of room to specify the behavior of units under attack. With a 'circle of arms' it's just more like a rock-paper-scissors game. I don't know if better or worse, but clearly different.

Indeed, quite different. I don't think it really turns into paper-rock-scissors at all, though. We're not talking about huge differences in ability here, just like the current resistance system rarely has units with huge resistances. It's something smaller - say a 10% to 20% difference in damage or to-hit.

It also increases the strategy of the game a bit. Right now, I notice that I often care more about the sheer strength of an attacking unit than I do about its attack type (pierce, etc), since the attack types are rarely important against most units. With a circle/network of arms, it becomes a little more important to match up the type of units you use beyond just range/melee. When you see that line of pikemen approaching, you know better than to send your knights in - send in your swordsmen. From personal experience (fighting SCA-style is a hobby of mine), an experienced man with a sword and shield can easily trump a polearm user any day. Set up those polearm users behind a wall of shields, though, and you have serious trouble. (Something that Wesnoth by default doesn't handle at all, but I understand that the engine supports it for any variants people want to make.)

[turin]

Nope. No walls of shields is possible, if by that you mean have a row of shieldsmen, and behind them a row of polearmsmen on a different hex.


Also, I don't know where you got this idea that usually you can just attack and not worry about resistances... but it is kind of false. Elves and orcs have few resistances, but most other units vary quite a bit.

[cobretti]

Well, it also leads to different matters: Whenever two units share the same weapon type, and are attacked by the same weapon type, it must have the same effect on them, no? Imagine that the attacked units are: Heavy Infantry and Footpad against Sword (blade), Dwarvish Stalwart and Horseman against Lance (pierce), Dwarvish Ulfserker and Skeleton against Lance (pierce)...

Again, it would need to be broken up more than the triad. Long blades vs short blades (very different styles of use), etc.

If the number of weapons are too large, a circle may very well be ineffective. A "network" might be a bit better. Each weapon has, say, two weapons its weak against and two its strong against.

If the number of weapons is too large, either a 'circle' or a 'network' is way too complicated. You have to remember all the combinations, and it grows with every unit type.

There are many other possible examples of clear disfunctions this method leads to, by the way. With the current system, there is a lot of room to specify the behavior of units under attack. With a 'circle of arms' it's just more like a rock-paper-scissors game. I don't know if better or worse, but clearly different.

Indeed, quite different. I don't think it really turns into paper-rock-scissors at all, though. We're not talking about huge differences in ability here, just like the current resistance system rarely has units with huge resistances. It's something smaller - say a 10% to 20% difference in damage or to-hit.

Well, as Turin said, there already are many units with huge variations in resistances, that make the difference in damage much more than 10% or 20%. Some of the examples I gave have them.

It also increases the strategy of the game a bit. Right now, I notice that I often care more about the sheer strength of an attacking unit than I do about its attack type (pierce, etc), since the attack types are rarely important against most units. With a circle/network of arms, it becomes a little more important to match up the type of units you use beyond just range/melee. When you see that line of pikemen approaching, you know better than to send your knights in - send in your swordsmen. From personal experience (fighting SCA-style is a hobby of mine), an experienced man with a sword and shield can easily trump a polearm user any day. Set up those polearm users behind a wall of shields, though, and you have serious trouble. (Something that Wesnoth by default doesn't handle at all, but I understand that the engine supports it for any variants people want to make.)

Not true. Damage types usually have much importance, it's just that you are already given the final values, and you don't need to do the calculations yourself, so you just select the attack type with best 'final' damage.

SCA? sorry I don't know that acronym, but I'm just talking from a gameplay perspective only. From a real fighting perspective, well, I'm also a martial-arts practiser and would be glad to discuss other aspects of weapon fighting, but here we are just talking about making the game fun. As Dave has said some times, between gameplay and realism, we choose gameplay, and I don't think it makes more fun.

[Jetrel]

Frankly, I would like to see a distinction between "light blade" and "heavy blade"

Light blade is dealt by things like daggers, and throwing knives, and most swords.

A wose, for one example, would actually be a bit resistant to light blade.

Heavy blade is dealt by big swords and axes. The aforementioned wose would be slightly vulnerable to heavy blade damage.


This would be applied to all units - almost every movetype would be modified to reflect this.

Heavily armored troops would have increased resistance to light blade, most everyone would have no change to heavy blade, with the exception of skeletons and other such troops, which would be weaker to it than normal.

Fleshy troops (ghouls, for example) would be slightly more vulnerable to light blade.

[romnajin]

Why are resistances categorized by movetypes?

[stillnotelf]

Probably simplicity of coding, units of the same general idea (unmounted solider, for example) will have the same movement types and same resistances. I know a whole bunch of the elves have exactly the same movements and resistances as well. It lets them double up on code, so that you need fewer characters and less effort to specify each unit.

[CyberJack]

It's a quick way to define some common defaults.

Individual unit definitions can (and do) override the class defaults; for example, the Cavalier inherits from movement_type=mounted but changes some of the resistances to reflect the specific nature of the unit.

Movement being related to number of legs , medium of travel, and weight of armour makes it a convenient starting point for selecting an appropriate set of resistance values. Though as I just pointed out in another thread, the Duelist currently has the same resistance values as the Footpad!