Before we can talk about fun stuff like spiked flails we need to go over some basics. This is so that everyone involved can work from the same page. I promise it won't be boring and hey, you might learn something.

For some semblance of realism as it pertains to warfare in a computer game, you have to understand 3 things:

1. Damage is not just a number, and for that matter neither is Health, Stamina, Armour or Morale. In fact they are a combination of many numbers and no numbers at all.
   
2. Computers run on math and everything can be quantified. Something quantified can in turn be simplified.
   
3. The ratio of complex data to simple data is the difference between a game thattries to emulate reality and a game that doesn't need to.

Those might seem to be complex and arcane statements that few people but me would understand. Mostly because they are! But a sentence is more than the words that build it and you do not have to understand what you read, only what it means. So lets get right to it.

--

I could go into exhaustive detail about modeling systems of combat in a computer game. In fact I actually did, before I realized that I was including far too many examples and various redundant blods of information that would bore you guys to death or else melt your brain. So I deleted that and started over. It all boils down to some very simple principles.

Damage is not just a number. It is a series of numbers working with and against one another. The key is that none of these numbers are visible. If I punch something I apply force. This force can be measured in newtons, or perhaps pounds per square inch. Lets say I punched a glass window and it broke. Your logical mind can realise that the force of my punch overwhelmed the tensile strength of the glass thus causing it to shatter. We do not need to know the numbers involved, and those numbers do not actually exist until we measure and record them. Computers work differently. If you want a computer to model the effects of a fist striking glass it has to know the numbers involved, It can make these numbers up but it can not function without them.

To translate anything to a computer we need to quantify it in some form of math. In a computer game, when the ork hits the elf and cuts the elf in half, you might not see any visible numbers. At that moment the computer had a decision to make, does the ork's X beat the elf's Y. To determine whether or not anything happens, X and Y have to interact, what number they actually represent is not relevant. In this case its apparent than X beat Y and now the elf is in two pieces and the ork triumphant.

When this data is made visible to us we assign common names to different parts. Normally X represents damage and Y health. These are concepts familiar to all of us. But have you ever stopped to consider other things like material strength, bone density, deflective armour and force multipliers?

--

To first step to making your weapons more realistic is figuring out some basic physics. Weapons do not generate energy, they only store and transfer energy. This transfer can't take many forms and more importantly many shapes. Impacts can be divided into 4 general forms: Slashing, Cutting, Crushing and Piercing.

- Slashing and cutting are not the same thing, but any wedge shape is capable of either. The difference is how the edge interacts with the surface. Lets say you were cutting an onion in your kitchen. Take a knife and press it straight down onto the onion until you separate it into two pieces. That was a cut. Now do the same thing only press the knife forwards as well as down. The knife will slide forwards as it cuts in a saw action. That was a slash. The difference may seem insignificant, after all they both have the end result. But this distinction will become extremely important when you start to model the effects of these weapons against armour.

It also helps us differentiate weapons which otherwise would seem very similar. Most of you would tell me when asked that the longsword is a slashing weapon. In fact the straight edge of the blade makes it far more effective at cutting. In order to slash something the blade must travel across it instead of directly into it. For a straight edge this means drawing the edge back and forth as one would a hacksaw. Not very efficient. This is much much easier for a curved edge. Next time you have a dinner plate in your hand place the edge against your opposite palm. By rotating the edge you can draw it across your palm without moving your arm back and forth. Most swords aren't as curved as a dinner plate granted but they take advantage of this effect to amplify the slashing effect on a target.

-Given that a cut is defined as a seperation of material caused by a foreign object. Piercing weapons could be described as inflicting two dimentional cuts. Hitting some jello with a sword results in a straight cut, stabbing the jello produces another straight cut. Now think of what would happen if you took a coffie mug and pushed it down into the jello until it hit bottom. Instead of a line we now have a shape pressed into the material. The jello under the mug is forced to the side and out. When you remove the mug the jello has been disturbed so much that it has trouble flowing back into the empty space as it did with both sword cuts. This is something common to all piercing weapons regardless of shape.

Because of this the wounds inflicted by piercing weapons bleed far more than cuts and take longer to seal themselves. Many stabbing weapons in the middle ages had a triangular cross section for this very reason.

--

Next up, crushing weapons, force multipliers, realism ratio, and the conclusion. After that things will be in much more manageable chunks, I promise. If this works out I will consider tackling Armour the same way.

Hate to say this but its just not that simple. If I listed every weapon and piece of armour as you described and explained things piece by piece then the list would be 10 times as long. To understand the specifics you must understand at least part of the whole.

Your chart for weapons and armour is waaaaaay off. The performance of armour is not something you can measure in a 0-4 scale. Let me give you can example. You list plate as having superior protection against crushing damage and the only one with no protection is chain. This is wrong, but its not enough for me to say that, I have to explain why.

Without getting too technical sharp weapons work on the principle of force multiplication. If I punched something, my fist applies X amount of energy over Y square inches. If you powered a sword with the exact same amount of energy, it would apply the same force on impact but focus it on an area thinner than a human hair.

This focused energy can be defeated by spreading it out which is why maille armour is so effective against slashing damage. Instead of pushing a hair thin blade through the meat of my arm you would be trying to push 10 square inches of chain links into my arm and out the other side, not gonna happen.

Crushing weapons don't work that way because they do not focus the energy they transfer to things you hit with them. You can't negate this energy by blunting it, its already blunt and there is a LOT of it. No armour will save you from crushing damage unless it incorporates heavy padding or can deflect the energy out and away. Without the padding, no armour on your list except plate will protect you from crushing damage. And for reasons too technical to detail here plate armour is at the same time the best and worst form of protection.

--

I will follow your reccomendation and split things up a bit. If and when we get around to armour I can divide weapons by damage type and armour into light, medium and heavy armour. Armour won't get nearly as technical as this thread because concepts that affect heavy armour simply don't apply to light armour and vice versa.

I agree hence my suggestion on using a psychological system like the one used in Close Combat. I don’t think any other game has even come close to developing such a thing. So here we have Frogboy a noted AI programmer, and an opportunity to test his mettle at branching out with such skills. It is a daunting challenge in programming methinks...

I disagree.

Imho fights should be simulated as accurately as possible to make "realistic" results. Conversely to popular wisdom I'm the oppinion that the more accurate it will be modelled, the easier it will be for noobies to understand it. A highly abstracted system is not very intuitive and can produce strange results (invincible large uberships with only weapons in GalCiv2 for example) while a more "realistic" design would bring results that would be more expected. (Expected as in: I saw that in a fantasy film; read it in a fantasy book.)

That said, I would be ok with an armor/weapons table for the sake of time and money constraints, but plate armor shouldn't be superiour to everything. It should be weak against halberds/poleaxes and maces.

So, a weapons table based on individual weapons instead of damage type would be better imho.

Cost vs return.

What weapon / armour system offers the best return (Most useable and understandable) for the least cost (Programming and balanceing time).

This is a game about Magic, who care about realism.  It just has to be fun.  Options are good but too many options that the average user can't differenciate between is bad.

Sammual

I would just like state that  a crushing blow is very hard to block using armor in almost any type of armor.  Although, armor of this type may spread the force of a crushing blow over a larger area it is still going to be a sever hit.  Think about being hit in the leg with. lets say a slashing weapon, while the leg has plate armor.  The slashing weapon is relativly ineffective.  As long as the slashing cut does not penetrate the armor.  Now imagine a flail coming down on the same leg.  The blunt force trama will most likely break the leg.  If I was a knight going up against another knight dressed in plate I would much rather have a flail than a broad swoard.

*edit I think speed, weight, skill, and fatigue are more your elements that will make the difference in if you win or lose the battle.

I think too much details and realism could be impossible for the player to follow and understand. After all, it's not intuitive, common man knowledge, I for one wouldn't know against exactly which armour types a flail is more or less effective for example.

Most important for me would be to have some different weapon and armour types, and possibly some "special effects" of some weapons, like extra attacks or being able to hit opponents further away with a long pike.

What should be achieved in my view is a strategc dimension like: a group of pikemen can be extremely deadly against mounted opponents (at least if they charge straight on) but much less in close combat against swordmen. Just so you have to think beyond simple damage and protection numbers.

As important would be the effect the changing environment and weather would have on different weapon types (especially in a game where you can manipulate the environment...). This was well done in D3. E.g. bows are useless in storm and heavy rain, plate armour in a marsh would be a nightmare ...

What should be avoided is that a unit is always of the same strength in all environments and against all opponents.

QFT

I agree 100%.  That said please continue the weapon lecture.  I was enjoying it from a learning standpoint.

Sammual

I also agree with gorgon.  I know in gal civ 2 I was frustrated with early ship creation because I didn't understand it.

I want to add to the armor thing.   Naturally, the final form of combat mechanics would have to be more than 10 times as long.  Its the nature of things, and I fear I am going to have to say it needs to be longer.

Do not forget this is fantasy in more than just how much damage things do.  Realism is great, but don't limit to realism when you have a game about throwing magical spells and raising armies of dragons and stuff.  We should have mythical weapons too, or things that would not have actually been seen on the battlefields of medieval earth because they arn't very functional.  Sammual had a good idea with the different types and I liked the chart he created, but I think it should also reflect the meterials of which it is made.   Since mithral chain mail should protect better from a orc's spear than iron plate.  Oricalcum spears would be better at piercing the plate than regular bronze ones.

That being said,  I think that a more advanced chart should be used similar to sammual's.  Charts are good to help with visualization, so if there is a mechanic that can have a chart applied to it in the help screen somewher, then that is what we should have.

I like vieuxchat's abilities idea.  That encourages custome item creation.  Maybe things that bypass other things, but we want to be sure the abilities are simple and still magical in nature.   Mccracken's rock paper scissors note is where to look.  In Master of Magic I was always sad at the item creation because I didn't know what I wanted my things to have.  If there is something like an ice enchantment that made armor impervious to almost everything other than fire weapons, it would really change the dynamic and encourage magic weapon creation.

Something I wanted to bring up is fire arms.   In fantasy games people tend to lean away from fire arms, and I understand why.  But at the same time getting primitive fire arms can sometimes still work.  Firearms would do about the same damage to everybody but would be really expensive to create.   I'm both a fan of classic pirates with cannon mounted ships and mage gun slingers that shoot magical bullets.    I'm not sure if it would fit or not, but it could.  I know when I was little I thought the sequal to master of magic would possibly include "technology" tomes that featured technology abilities rather than just the regular 5 magic tomes.   People with technology tomes would effectivly use mana to drive magic-based tanks and missles and stuff.

This thread is quite interesting from a learning standpoint, however I think in order to be intuitive things have to be dumbed down a bit more... that being said, I can't understand why anyone would want a rock-paper-scissors model in a Turn Based Strategy game.

I would also like to note that a large amount of a weapons effectiveness is based upon how the weapon is employed.  Macedonian Phalangites were extremely effective to the front because of the abovementioned reach (which the said formation allowed them to utilize), however their massed pike formations lacked flexability, and were thus cut to pieces by the Roman Legions (who used light javelins and stabbing swords).  However, it was only proper use of the legions which enabled them to beat the Phalangites.  Had the Romans fought the pikemen head on, instead of utilizing their flexability to break up and envelope the pike formations then they would have lost.  Much the same can be said about almost any weapons type.  It is only when used with proper doctrine that it becomes effective, and a formation armed with almost any practical weapons can beat another formation armed with almost any other type of weapon if they are used properly (example, cavalry would utterly destroy a massed Pike line if they were hitting that line from the flank or rear... it is only if the pike line is actually facing the cavalry that you would have a lot of dead horses).

You do realize that a lot of TBS games use a rock-paper-scissors model. Units are designed to have hard and soft counters against one another. There's a reason why when you highlight your pikemen you see a little tab that says 'effective against Knights'.

Now TBS' add in other elements like terrain and battle conditions but all these systems are still based around R-P-S. Like I said before, there's a reason for that; it's simple. No dev is going to spend years trying to program a system that attempts to take into account every aspect of realism. That's time that could've been spent on other aspects of the game rather than wasted on developing the most convoluted battle system ever.

Edit: This is in response to Iwarmonger of course.

Yes. But dumbing down or simplifying the system is not the way to go. You have to have as much detail as possible and simplify it for player consumption. The detail must be there to handle calculations in the background but you can't overwhelm the player by feeding it to him all at once.  So by picking out the most relevant information and putting it up front you give them as much as they need when they need it.

As for tactics. Simply put, good leaders do not micromanage. They teach those under thier command so that they can handle themselves with minimal imput. You don't need to tell them what to do because they already know. Its a bit more complicated than that of course but you should be able to tell your troops how to act in certain situations.

Some things should be automatic common sense unless you give orders to the contrary. A pikeman who gets his pike broken should draw his sword even if he can't immediatly use it. A group of swordsmen standing in the open with no outstanding orders should spread out and lift shields overhead. Stuff like that.

Oh, I agree.  I just don't want them to oversimplify things, as is the habit with a lot of games when they go down to the tactical level.  No point in even having it if they leave no room for strategy and tactics.

However, I think that this will have to be a bit more complex than you are thinking, simply to make designing your own units worthwhile.