Actually going into those specifics would of coure be quite over the top. That is where some abstract comes in...

Why would we ever need that many categories really? We can all imagine that our weapon is a special, slightly longer *insert sword type* because it was made in *enter region of the map*, which is, as we all know where the *insert some special blacksmith race* lives... Etc. Some one with more imagination than I ave can finish the story.

I never really understood why we should have these things spelled out for us and why we cannot think for ourselves. Does it really matter what a sword looks like and what the measurements are? In the end it is the stats that count. the basic fantasy weapons cover the niches quite nicely and there is really no need to go any further. The hardcore weapon freaks can mod the rest in.

Plus most of the crap were going to put on the swords is just eye candy. It doesn't matter if it has a gap down the middle, or if the blade edge is wavy shaped. The hitbox for that sword will be the same. In fact it doesn't even have to be a box, you could get away with two points and a line between them.

I agree with Tamren, this ain't a problem. I myself could make 300 sword varieties using Paint program. all i need to do is to make pre-made components like cross guards, hilts, pommels and blade types. heck i could make several different varients of Rapiers alone.

READY?

GO!

I forgot the math used to calculate variations of a set. But even if you only have 10 types of blades, 10 types of crossguards and 10 types of hilts. That adds up to a heck of a lot of swords variations, even with 3 components. And that number will skyrocket if you get the ability to resize them.

Systems like this can be found all over the place. Mount and Blade has a system of generating faces that relies on a set of sliders. At the start of a new game you get to create a face using a huge number of sliders. (I never counted them but there are probably 50, maybe more) Everything from forehead size to lip thickness.

When you get into a fight with 149 other humans the game goes and randomizes a face for every single one. I see no reason why you couldn't use a similar system to create randomized looks for swords and other weapons. Considering that each and every one is made by hand they would vary greatly in size, shape and design. In fact you could extend this randomization to everything from body size to shield shapes.

No more armies of clones ala rome total war. And since the whole system operates on rules, not examples it wouldn't hurt in the processing department.

And as soon as the player doesn't care, building the whole system is a waste of time. There's lots of things the players DO care about that need developer attention.

It might be interesting to have a History of Medieval Weapons supplement for the game, but even though I 'like micro,' I'd be pretty annoyed at having to think about 300 different instances in just a single class of weapon. Maybe two years after I've been playing the base game in a few different modes, but out of the box? My eyes glaze over at the thought.

I'll be more than entertained if we just get a modest step up from the weapon-defense complexity in GC2, say the ability to carry more weapons in inventory than we can have 'ready' at a given time, enabling units to essentially reconfigure without going through an upgrade cycle like they have to in GC2.

That reconfiguring is actually a nice idea. I would also like to see units carrying bows for when the enemy approaches, and they can drop them and enter melee once the enemy closes in. Allowing units to carry miltiple weapons like that would be cool I think.

Another idea that comes to mind is allowing for something called a pilum like the romans had. It is essentially a spear that was tossed into the enemy rank before charging. Even if it was not the most useful of ranged weapons it did infact allow the romans to take out some enemies before they were in melee range. Even if the chances of being killed were slim you would still be seriously hurt if you got hit, and for all purposes you were out of the battle.

That wasn't my point. I was saying that all of this super detailed complexity would not be modeled in large scale battles because they have little to no effect on the battle as a whole. If you are playing MMORPG style which a single character who is YOU, breaking your ankle in a fight is a terrifying prospect. But when you are throwing blocks of 500 soldiers against others, what does it matter if one soldier out of 500 stubs his toe? Now if those 500 soldiers run over a patch of ground festooned with pit traps, THEN you get a warning that pops up about decreased movement speed.

Well... like I said before those 300 swords would condense into very few actual "types". You have to change a longswords quite a bit before it no longer works like a "regular" longsword. So what we should be aiming at instead is having maybe 10 types of swords, and 300 skins and models for each. The mechanics of how they work in combat do not change.

Hell yes. More weapons on a single soldiers comes with tradeoffs like weight, extra training and logistics. But the benefits can be worth it, especially if you configure your army into a combined arms force.

They were also used to repel enemy charges. A shield with a pilum in it is too ackward to be useful. Everyone wounded or tripped by a pilum or one stuck in the ground is another obstacle. And so on. It was also pretty genius that they couldn't be thrown back.

So, basically, you want an arbitrary cutoff point of troop numbers after which your system no longer applies. Not only would that mean 2X the work for whichever poor dev codes the TCAI and TC calculation engine (a different AI and calculation system for deatiled and abstracted), but it would also throw the horde/elites system out of balance: since battles using hordes of weak soldiers would use a different system than battles involving a few well-trained experts, you would either find it easier to win in one or the other, depending on which one is tougher on the AI.

So really all you want is a lot of extra graphics? That would be doable, as long as Boogie et al had the patience for it..... however, all the hitboexes and stuff would still cause SERIOUS problems.

There is no arbitrary cutoff. It always applies when it is relevant. Things just have different effects at different scales of conflict. If your channeler is leading around a posse of 9 other warriors and one of them steps on a thorn. Your "squad" is going to have to slow down, or else leave him behind. If your channeler is leading 499 other soldiers over a hedge wall the same thing applies when 50 of them get injured from the thorns.

The problem is just people stepping on thorns and the effect it has on combat changes remain the same. What changes is your perspective over the situation. In small squad combat the sitation is that one man has an injured foot. In army level combat all you need to know is that one of your units has a penalty to movement this turn.

Its really not as complicated as it sounds. If you went and googled up some images for swords probably 80% of them have straight blades. A straight line is probably the easiest hitbox you could create. It doesn't matter if we end up with millions of examples of straight bladed swords, the hitbox for each and every one could be rendered as a simple line.

This is why most of the work is graphical. Make one longsword shaped hitbox and mesh and all you have to do is stretch it in different ways to create 300 unique swords that all work the exact same way. What is really required to make them unique is a snazzy look. Even if this involves using a different mesh than the standard one the sword still retains its basic hitbox.

1. There has to be an arbitrary cutoff. In the example above, the computer needs to know that a thorn is important to 9 soldiers, but not to 499. For the system to work, there has to be an arbitrary number of units for which the computer knows to say "let's zoom out". Even if you did add some sort of gradient thingy, the imbalence I mentioned would still stand.
2. Without a full, integrated physics engine, there will not be hitboxes: the battle you see is just a graphical representation of numbers being curnched. Even if you did have full physics, the way the unit swings its sowrd is determined largely by luck and accuracy numbers that can be much more easily abstracted as just that- numbers.

I see what you mean now. Well we could add a rule that says at least 5% of the figures in a unit have to be affected before the unit as a whole will suffer a movement penalty. 20 elite swordsmen might be able to beat 100 mooks. But as soon as one of them takes an injury bad enough to make them limp the whole unit is affected. In this regard hordes are more "durable" to movement penalties. And it also makes much more sense to spend extra and outfit all of your elites with steel plated boots.

This is only one instance so it wouldn't be fair to say it imbalances anything as a whole. We don't know enough about the combat system yet anyway.

Well... define "full" physics.

Wouldnt it be easier to have pierce/bludgeon/slash/magic for weapon types and light/medium/heavy/arcane for armor types? You can make a "scissor-paper-stone" type of combination with those and could even have a morning star doing 70% of bludgeon and 30% of piercing damage. Then you can have slashing doing 60% damage on heavy armor, while pierce would do 90% and bludgeon 110%. Something like that would be good and would simplify things while still making them interesting/important during gameplay.

"Full Physics" as in the tactical battles being run like an FPS or RPG, where all objects are able to freely move in a three-dimensional world, ogverned by Newtonian motion.

Well yes and no. A straight up RPS system is simple to calculate and easy to understand, but without enough detail you start to lose the fine distinctions between weapons and the like.

No matter how detailed we get with weapon damage scores it still boils down to a single number for each damage type. That number doesn't exist until you assign the weapon to a user. The maximum damage you can do with a weapon depends on how strong the user is as well as how much force the weapon can give before it breaks. What TYPE of damage is done depends on how it is used. Its not good enough to say that a halberd for instance does 33% cutting/piercing/blunt of X weapon damage.

Ah. Well even if newtonian physics are present they do not have to apply to everything. A dragon walking along the ground is going to leave footprints. But it won't be because the weight of the dragon overwhelmed the soil with its ground pressure rating and deformed it enough to leave a mark. Its just going to leave a footprint. Sand is a lot harder to run on than stone. But we wouldn't be calculating the extra calories our soldiers require to move over it. Sand would just reduce overall movement speed by X%. You do have to draw the line somewhere. But adding a feature that requires a lot of processing power does not mean it will need that power all the time.

Mount and blade for instance has ragdoll effects. In the options you are allowed to set how many ragdolls can be calculated at once. At first I set this to "unlimited" and never had many problems. Generally people just fall over dead and don't go bouncing and rolling very far. But then I started fighting battles on very hilly terrain. Suddenly there were 50 ragdolls all bouncing and rolling down huge hills and my computer slowed to a crawl. This would not have been as much of a problem had the ragdolls felt the effects of friction. But instead they slid down the slope as if it was greased. Given that the problem was situational it wouldn't be hard to program a situational fix.

Perhapes "full" physics was a bad choice of words. Looking at the situation now, I realsie that "integrated" physics better describes it. The fact reamins, though: the actual mechaincs of TC will be governed by math, not physics. Each "unit" is just a pretty graphical "face" for what amounts to a square with ident and location data in it.

perhaps we could have some sort of tweaking feature where you can adjust the length of weapons, reflecting in changes to reach and attack speed.  Reduce weight, sharpen edge (increase damage for slashing) and other such features.   It would be complex, but it sounds a lot easier than developing some sort of physics to calculate impact.

I think the original MoM did a reasonably good job of this. Units with a certain type of equipment get bonuses. So 'first strike', 'negate first strike', 'armour piercing', stuff like that. It could extend qute nicely I think.

Although a real first strike only comes from being invisible and airborne .

A lot of this sounds extremely complicated, but only when it is on paper. Learning that bladed weapons dull easily against hard armour is a lesson you learn once. But if you were to explain it on paper you would have to talk about material hardness, sharpening techniques and all this other hoopla.

So you're saying that if you have a troop of 500 units, and one steps on a thorn, it's just ignored? Even though that soldier can't move at the same speed as the other 499, you still have a troop of 500 soldiers with full movement?

While if you have 10 soldiers and one of them steps on a thorn, you have to either slow everyone down or leave that guy behind? Honestly that seems terrible to me - small groups would be seriously disadvantaged because a small impairment would cause major consequences to them while that same impairment is ignored for a larger group.

And what if you have 500 troops and every turn one of them steps on a thorn? If that movement penalty only lasts a turn, the larger group would remain at 500 and at full speed, while a smaller group of (say 10) would either have to leave people behind every turn or move at reduced speed each turn. If the movement penalty lasts longer than a turn, effectively the same thing occurs - with just the difference that after some time the larger group might reach a critical number of injured and bring things back into balance - but the effects would be delayed by however many turns it takes for that to happen. I just don't see a mechanism like this succeeding without causing major issues.

That was just an example.

When someone at the front of a group starts moving slowy it will not have any effect if the unit as a whole doesn't get ordered to move faster than that figures top speed. If that did happen then it has an effect on the rest of the unit. Faster members will move around them like water flowing around a rock which slows them as well. Eventually they will drop all the way back to the rear of the unit.

This is the tricky part. What are you going to do now? In order to maintain unit cohesion you have to slow down the unit as a whole or else leave someone behind. If you do leave someone behind how do they escape? Do they just trail behind the rest of the unit by 50 feet? Is that even possible on a tile based map?

We just need to think of something/.

Ummm.... there's another problem: each unit, although it LOOKS like several soldiers, is in fact just a single entity. You can't divide it without a LOT of extra code.

What unit in what game? We don't know anything about elemental's combat system yet.

Units with multiple entities are nothing new. Of course they require extra code, but that always comes with added complexity.

Of course is every soldier is treated like a seperate entity we will need some solution for problems like these. If this game should be 100% realistic there should be some solution for this as one slowed soldier can and will in fact slow 499 others down unless they leave the wounded one behind.

What I am wondering however is why we would want a realistic approach rather than a more simplistic one. I greatly prefer the simplistic approach where thorns do not slow down individual soldiers in a squad, but rather have a chance of 25% - or whatever arbitrary effect would be more desirable - to slow an entire squad down.

This has an additional benefit to it as well, because showing all these effects to the player will help in the player learning the mechanics because they are all represented in ways we can understand. In the end the difficulty should not come from an obscure system for combat - Empire:Total War I am looking at you! - but from choosing an optimal strategy.