Its been a common complaint that weapon systems in games (especially RPGs) are not realistic enough. Sometimes its necessary to reduce complex cause and effect relationships down to numbers, but sometimes defense values and +3 to damage equations simply do not seem good enough. NWN incorporated a parry and riposte system, but in KOTOR the mechanics behind swordplay (even with lightsabres) were reduced to simple attack and defense numbers. The Thief series once incorporated an interesting sword combat system, but the new game in the series has left it out. Are complex weapon systems that represent both the defensive and and offensive uses of weapons a thing of the past? It seems to me there is a lot of oppurtunity here to implement something fun that can also be a channel for player skill. In this thread, I am hoping to start a brainstorm of sorts to come up with new ideas about possible weapon systems. An idea I have involves the use of multiple weapons and multiple targetting. When fighting multiple enemies, a character could target a certain amount, and by doing so gain defensive bonuses, and focus on these enemies in order to 'read' their movements. A player's array of equipment could be used in ways that spells are used in other games, except they would be context-sensitive. Any other ideas or related thoughts?

To see good, complex swordfighting you need to look no further than Japanese-style fighting games. There you can see all the possibilities of the stupid crap you can get into trying to simulate realistic swordfighting. While some of them feature fireballs and equally stupid gameplay, many of them are based on realistic weaponry (though damage is always unrealistic).

The general problem you reach is this: it becomes a super-high-speed game of Trivial Pursuit. You learn the combos that the enemy player has and learn how to block it and when you can insert a shot with which to start your own combo. Upblock downblock upbloc upblock STRIKE combo combo combo combo combo combo. Repeat ad nauseum. And that's realistic. I've fenced IRL - that's somewhat how people play - there are sequences of attacks use, sequences of defense to reply, and points of weakness in which to inject something. Fencing is fun because its exciting and real, but not because the strategy is very deep.

There are more "freeform" fighting games, particular fighting/adventure hybrids like Soul Reaver and Zelda that nicely mix hand-to-hand combat with free movement. One important feature: first person is a no-go. You don't have the depth perception needed.

If Blizzard moved away from their love of "can't move and attack at once" then maybe Diablo could have had good sword fighting, but Diablo wasnt' intended to. It was meant to be a real-time high-graphics Roguelike.

I think the best game for multiplayer swordfighting remains the new Jedi Knight games.

I think this is a two fold problem.

As others have mentioned, you can take two approaches. You can design the combat system such that you take into account player ability, or you can take into account character ability.

When designing the combat system, this should be foremost on your mind, because this decision affects every other design decision, including how the player interacts and sends input to the game itself.

In paper and pen RPG's, there is sometimes a criticism leveled at games which require the players themselves to master the combat rules in order to be effective fighters. The criticism's logic is that the character is not the player, and the player is not the character. It should be possible to play a great swordsman even if the player himself has never held a real sword in his life before. In other words, skill rests with the character, and not the player. Such game systems often rely very little on player input and choice (and skill), and essentially become probability calculators based on the experience, skill, and weapon capability of the character.

OTOH, some games want to get the player into the nitty gritty of things. This requires the player to become familiar with the intricacies of combat itself. In such games, players (and hence their characters) are rewarded or penalized by how appropriate their input is to a given situation. For example, weapon length is critical in determining what sort of attack and defense posture to take. One armed with a dagger is probably not going to launch an attack against an opponent armed with a spear, he'll more likely try to dodge/evade, and manuever himself such that the tables are turned, and he is now inside the opponents effective weapon range (spears are good at keeping opponents at bay, but once an opponent breaks past a certain range length, the advantage goes to the wielder of the shorter weapon).

These kinds of games require that the player himself is familiar with the tactics of combat, and especially how these tactical rules are defined within the rules. He doesn't have to know the exact benefits or penalties, but the player should realize that there is indeed a penalty or beneffit in the appropriate context.

I personally prefer these types of game systems because to me, like anything worthwhile in life, there is a learning curve. While the other school of thought has a point in saying that the player and the character are not the same thing, I think they miss the point in that part of the entertainment of a game IS to get into the shoes of your character, and to learn and experience the things your protagonist feels and knows.

So now you have to determine the actual tactical combat rules themselves, keeping in mind what the player would want to know to maximize his combat ability, and minimize his weaknesses. If you're looking at melee combat, I'd include at least the following:

External Factors (not related to character ability)
1) Weapon Length - the length of weapons is important as noted in the discussion above
2) Weapon Agility/Speed- some weapons, though they are the same length, are more responsive than others. This for example is an advantage of a sword over an axe
3) Weapon Damage Type- is the damage cutting, impaling, crushing?
4) Weapon Durability- wooden hafted weapons have a tendency to break more often than all metal ones...a problem that can (and did) happen in real life combat
5) Terrain- What is the environment you fight in? You don't want to fight an opponent with a short sword in a confined hallway when you have a 2H axe.
6) Encumbrance- So you thought layering all that Plate mail was a good idea? Well, it would be if you know you've got a battle coming up, but you wouldn't want to wear it 8hours a day. Tell that to the Vikings at Stamford bridge in 1066 (just before Hastings) who took off their chainmail because it was a very hot day (it gets hot in England?).

Internal Factors(character action/ability)
1) Strength- Can the character wield the weapon?
2) Wounds- This includes fatigue. Very few Computer RPG systems have good wounding systems that modify the effectiveness of a character.
3) Skill- How experienced they are with their weapon
4) Ability- How agile or dextrous they are
5) Reaction- How quickly can the character react to events?
6) Positioning- This includes the opponents facing relative to each other, and their absolute location relative to one another. For example, fencers sometimes fight obliquely, such that only one side of their body faces the opponent. This is actually VERY important in real combat. Manuevering to get ideal positioning is half the battle.
7) Target Vector- how and where is the character attacking his target? Is he swinging his weapon, and if so, from what direction? If he thrusting his weapon? This is VERY important and oft neglected. This plus positioning determines many other factors.
8) Action- What action is the character actually performing? A dodge, maneuver/evasion, move, move-attack, attack, evasion-parry, counter, grapple? Note, that there is only a basic attack..all the other factors, like positioning, target vector and others will combine to form more complicated maneuvers
9) Morale- how come almost no game ever considers how frightened someone is to affect their combat ability? In many ways, this is probably the most important aspect to combat (including ranged combat) because it affects every other internal ability.

As you can see, it's a pretty lengthy list, and there's more you can implement. And this is just for melee combat. There's other things you have to worry about when dealing with ranged combat as well such as, damage, penetration, max_range, accuracy, target delta_theta (the change of the angle of the weapon over time to keep on target), recoil, aim time (how long you aim), stability (bracing, kneeling, prone, standing, hip fire, supported) et al.

Hey guys, thank you for replying. First I'd like to apologize for my absence in this thread up to now, been in a time crunch, but lets get down to business.

Lets begin by following Estok's suggestion and defining exactly what we are talking about. First and foremost, the goal of this thread is to move away from the mindset that JohnSKX has put forth, and to increase the complexity of weapon systems using realistic representation instead of simplified representation alone. In most combat systems today, there are usually only three factors: damage, accuracy, special effects. Although these three properties can cover a wide range of combat situations, they also focus combat into a predictable system of numbers and specific effects. It is my belief that games are all representations at many levels, and in any environment with multiple complex systems and several variables involved, realistic representations are key to making a game more interesting, as well as more intuitive to the player.

I think the next most important thing to discuss is the character-skill vs. player-skill problem brought up by Dauntless. I agree with a lot of what he has said, and I also believe that taking the route of making a more realistic and complex system is the prefered path. Gaming is all about mastering something, and although the focus of many RPGs is to play a character that is greatly different from you, it can be very rewarding to attain mastery of a game and is the reason why many play them. In RPGs this is also true, but usually very simplified, however even for RPGs I believe a lot of people are willing to take the time to learn a complex system and in this way enjoy it more. The goal of this thread is to come up with a combat system that is on par with the complexity of Baldur's Gate, but at the same time easy-to-learn due to its intuitive nature and hard to master because of its dynamic nature.

In his post, Pxtl's brings up the negative properties of having a focus in combat systems on twitch skills and combining those twitch skills with move-lists. I agree completely, although like Pxtl said, I also think we must also acknowledge the advantages of japanese fighting games. "High-speed trivial pursuit" is not a good way to implement a combat system, but fighting games are very good at representing many of the factors of combat by giving players a set of very different skills that have different uses at different times. The problem for these games is not that they call upon the player to think on-the-fly and react to different situations fast with the skills they know, but because they force players to react through a non-realistic move list and combo system. The more popular fighting games such as Soul Calibur or Tekken do implement more realistic controls and because of this they make more sense to players and are more fun to play.

Being a fencer myself, I can agree that in fencing there are many sequences and the amount of movements you can do are in fact limited. However, fencing seperates itself from fighting games in that: 1 hit will make you lose, and the complete freedom you have in every move you make.

Thats why I believe that, regardless of the actual variables you use in a combat system (Dauntless has a lot of great ideas I think), the most important things to remember are: 1) Make each move matter, and 2) Make each move dynamic in its nature. Following this theory, the ideas I have to increase focus on these points off the top of my head are:

- Damage systems including exponential damage curves, with a wide range of health penalties once a body takes critical damage
- More realistic representation of a player's status. Most games include only health, stamina and/or mana. Although these systems are easy to keep track of, they are not realistic and therefore are detrimental to an intuitive combat system that makes sense. Removing this simplified representation of a character's status will eliminate a big part of the reason for why many MMORPGs are number-crunching competitions.
- A greater focus on attack variables independant of specific click-and-forget weapon powers, and the uses of these attack variables in relation to defensive properties. Many games take combat and focus it into individual 'powers' that are only effective in certain situations. Balancing these kind of systems is extremely hard and leads to an unrealistic implementation of many skills. Each action a player may take with a weapon must have different effects in different situations but a player should be able to control attack variables such as strength used or weapon direction/positioning. Combat is flowing, not a planned queue of skill-uses, and although there is a macro strategy in combat, the very important micro is not being represented at all (even if its based on character-skill, these factors should still be present).

The key is to make combat more dynamic by focusing on realistic areas for expression of player freedom (the ability to change direction while swinging and other mid-action changes) as opposed to focusing on simplified systems and then high-level balancing. Many RPGs do not unlock skills until the player has reached higher level and this is a big part of the problem, a game does not have to be completely player-based or character-based, I believe a system can be created to take advantage of the positives for both systems.

Great replies, hopefully we'll come up with a lot of new ideas here!

Well, my combat and damage systems seems to do kind of what you want. Everything is completely character dependent, except for obvious stuff like coming up with tactics and controlling it, deciding which manouvers it should attempt. The things that my combat system tries to "fulfill" are these:

Realism (damage and to-hit calculations and general feeling of the combat). The player should not need a "twitch skill" (Diablo).

First off, it's in realtime. If you want realism, it has to be realtime, because turnbased (or phasebased or whatever) isn't as realistic. Note that I like turn based games, but for my game I have chosen a realtime approach for a more realistic approach. The player does not need a twitch skill, because it features an auto-pause system and also a slowmotion system, because even with auto-pause the combat can become rather difficult to control or oversee. The slowmotion system is simply a slider of some sort with which the player can control the speed of the game (the slowest setting should probably be something like: one real second = fifty game seconds).
Anyways, that doesn't really have to do with the combat system or realism, it's merely a way to make sure that the player can be in 100% control at all times.


Second, the game is in 3D. This is naturally more realistic than 2D, and bullets flying through the air will hit their target because the target got in the way and not because some Chance To Hit roll was made. However, I do use a CtH roll for close combat (since modelling a real close up combat would be near impossible).


Third, the damage system. I do not use health points or stamina. I use Wounds, Pain and Blood Loss.

- Wounds: When damage is inflicted on a character, one or more wounds may be generated. First, the game calculates the damage rating by looking at things like the weapon, the armor and a random roll to simulate bad hits. Then, the game looks in the wound table for the corresponding body part that was hit, and damage type (piercing, crushing etc.). In the table are several wounds, and the damage rating is checked against the value for each wound. If the damage is greater than that value, that wound is inflicted.

So what is a wound? It is something which has a direct effect on the character. If you get hurt in the leg, you may start to limp or it might simply give you an adrenaline rush if it was just a scratch. If you're hit in the head, you may faint, become blind, or die, for example.
You can always die even if you were hit in the foot. The damage just needs to be higher. Also, each wound generates Pain and Blood Loss.

- Pain: I have not really given much thought to this yet, but the principle is that wounds hurt, and when things hurt too badly, the character faints. There could also be other effects, like getting an adrenaline rush (yes, already mentioned so maybe it should go here instead ;) ) or maybe even dying.

- Blood loss: Each character has a certain amount of blood in his body. When a wound is inflicted, he might lose some blood. When a certain amount of blood has been lost, the character will start to degrade in performance. When even more blood has been lost, the character will faint, and when too much blood has been lost, the character is dead.



Well, that's most of my system right there. I'm not sure if it was exactly what you wanted, but there you go. :)