#64 Combat Stances

[Quigg]

When I think of the difference between GUI-based and text-based interfaces for combat, the first gives me something literal and exact to look at during play, while the second requires more imagination.

I think a key advantage of the text-based medium, if used, is it's flexibility in allowing not only developers, but players to engage in improvisation during combat. Stances and emotes need not be tied directly to the combat resolution code to add to the fun.

For instance, you mention swinging on a chandelier. Yes, such an action could be hard-coded into a combat simulation as one of N choices for the player in a given situation. In a rigid GUI-based game, this would imply a programming task to generate the corresponding visual effects. In a rigid text-based game, this would require an appropriate text message be displayed "Pirate Bill swings from the chandelier, slashing from above."

But there is another possibility in a text-based game. Code up a generic set of mechanics options such as "berserk attack, normal attack, sneak attack, hide, duck, defend myself, defend Pirate Bill, balanced combat, special defense, magic defense, grapple, escape, pummel, spear attack" etc. etc. as appropriate ... then add a second ability to emote during combat and even to position your character in a way independent from the mechanics. Thus I might "avoid combat" as a mechanical choice while "Pirate Bill ducks behind the large barrel" as an emote - just for RP flavor, but no direct tie to the combat system.

Depending on how it is done, this could shift the emphasis more to the role-playing in the text-based medium, allow for greater degree of improvisation and also of variety of combat and stories ... while still keeping some concrete resolution system in place for players to utilize.

This duel approach can either be left to the player's, or might be moderated by a GM in a small game. A GM (or audience) might give bonuses to particularly clever improvisation, for instance.

- Quigg

[Nom]

Interestingly enough, this neatly summarizes two of the three complaints I have with the current CM duelling system:

(1) It's too fast.

Cycle times are slightly longer than most online for a twitch-based games I've played. However, they are way too short to allow players time to type in more interesting emotes allong the way. For example, the game can move through four to five states in the time it takes me to 'wave my sword wildly "Yaah!"'.

(2) It locks out commands during recovery.

After I (for example) 'slash', there is a timeout before I can take another action. This lag time could be spent emoting, if the game did not prevent me from doing so. It would be better if the timeout only locked the 'loaded' commands.


The third (reasonably unrelated) complaint is that the engine is too simple. Once you know how it works, the system is almost degenerate. For a text based storytelling game, it would be better to have something that worked a lot slower, so it was almost a lightweight strategy game. Or alternatively go even simpler, and have a consent-based way to force the issue.

[Gareth]

Stance can reflect issues of:

1) Risk-taking -- such as taking tricky or recklessly risky moves for hopefully positive results, or prudent avoidance opting instead for more traditional, safe or conservative approaches

2) Tempo of battle -- rate of energy consumption -- leading to exhaustion or mishap because of mis-timing.

3) Force -- also adding to energy consumption is the strength or force of blows. Some forces are strong, direct and focused, and some are weak, diffuse and yielding.

4) Dynamics -- motion or statis of position, and accomodation or refusal of opponent's way.

Let's just do some quick examples:

a) "None Shall Pass!" -- I want to block a door so the criminal does not escape the room or the enemy does not get past me. I will accept risks to do so to ensure the dynamic of escape is denied the opponent, I will meet force-with-force and not yield but I will not advance to leave the passage point, and I will conserve my energy for as long as possible so I am not overwhelmed.

b) "Let's Get It On!" -- High risk-taking, furious tempo, powerful and vigorous blows, and strong dynamics of movement. We're going to mop the room with each other. Expect things to get knocked over.

c) "Rock-'em Sock'em Robots" -- Low risk-taking, strong exertion of force and fast tempo, low dynamics of movement -- it's a blow-for-blow battle until someone gets their head knocked off. We move forward, engage, and no one moves other than some foot-shuffling until someone keels over.

d) "Get Me Out of Here!" -- Low risk-taking, evasive maneuver, disengagement, and high "trickiness" is needed to skirt past enemies and find your way out of a troubled location. Consider a footballl running back needs to stay disengaged, and will do anything they can to not get stopped or overwhelmed. This is useful for fighting escapes too. I also think about Aliens and that famous line "Marines, we are LEAVING!"

Riskiness is a matter of recklessness and prudence, and an aspect of chaotic (surprising) versus orderly (anticipated or expected) tactics.

Aggressiveness is a matter of expression of strength or force of your energy versus conserving your force or strength, thus possibly yielding or accomodating the opponent's chosen tactic.

Being "full out" aggressive might be a good tactic if you feel you have better endurance than an opponent and your wildness can wear him out. Conservation of energy or yielding might not be possible if you are being pressed hard.

At the same time, if you have room to maneuver, dynamics might be the way to win certain battles.

At the same time, if you are well-armored and holding your stance well, you might be able to hold off a wild enemy by letting them waste their energy against you.

This is where tactics should reflect optimizations based on character statistics and skills, and of course the desired goals of the combatants.

Am I fighting for life-and-death, or am I fighting for some other purpose?

There is also the concept of "order" to a battle, which the stance should probably affect.

You'd hope a "professional soldier" would stay in their battle like and fight shield-to-shield with their fellows in good order.

You'd hope the wild warrior would break ranks and hope to shatter the enemy line and turn the battle into a series of one-to-one melees in disorder, where their wilder, more chaotic tactics can be put to better effect, and where in open order the exhaustion of fighting in heavy armor can wear their opponents out quicker.

These are not first-order problems for most combat simulation systems. Yet they are the more satisfying elements that people with actual experience in fencing or sword fighting or combat situations look for.

The actual flurry of "jab, cut, feint," is interesting, but that's a button-pushing atomic and dare I say "neurotic" exercise. The concepts of controlling your avatar with more of a thought of goals, strategies and tactics -- more than just raw action -- makes it a far more psychologically interesting game.

[Gareth]

One way around the issue raised by Nom of the neurotic 'jab-jab-jab until someone falls over' -- where no real roleplaying is possible between tactical actions -- is to make the system a "two-player posted action" system. In this, both players need to commit their selections before resolution of the "round."

It is somewhat unrealistic in that it allows people to hold off on their attacks potentially indefinitely. It might be metagamed so a player who was losing simply does not respond and thus combat falters. It has the advantage of allowing people to make tactical choices and develop strategies and, moreoever, to "roleplay."

It also helps prevent problems when one person is going 'jab-jab-jab' and the other person is disconnected. Many people do not like reconnecting to find out they are wounded, or worse, dead.

This is more like the classical "Lost Worlds" system, where both people have to make a selection to "clear" the turn. You choose. I choose. Good! Once both have selected, both declare -- or the system reveals the choices to each other -- and the round is resolved.

However, this might be too esoteric and slow-paced for some people. You might put a timer so that, in absense of a choice, you do nothing, or stand defensively for a time.

It's a matter of tone and style.

What is the style and purpose of combat in your game? Are you looking for a lot of chandelier-swinging dramatics, long inter-round soliloquys, interspersed by some flashing blades, or are you here to inflict maximum damage in minimum time, such as to get through rounds of monsters in a huge flurry of action?

If we want to play Xbox for furious action, we can do that. Or are we more armchair strategists? Or dramatic hams?

Solve for a this metagame need -- why do we fight at all in this game? -- and you'll be further along determining how you want your combat system to work.

[mikedsc]

AD&D's 2nd Ed. Player's Option (Combat & Tactics) has a duelling system that reflects what you said, Gareth.

The problem about such a system is that it requires that ALL combat be one-on-one duelling. Then it falls to the fact that you simply can't use it in certain types of games where 3+ people are participating in the fight.

What do you all think is an appropriate length for a duel, anyway? Everyone seems to agree that, in general, it moves too fast for reasonable roleplaying (which might, in some cases, be said to be an immersive quality: "the heat of battle" and all that.) But when does it become too slow and tedious?

[Yxven]

I believe that the key to fun combat is close calls. Combat is not fun if you can easily predict the outcome of the battle from the beginning of the battle. Which scenario is harder to predict? Scenario A: two combatants are fighting. Every round alternates which combatant is near death. Scenario B: two combatants are fighting. Combatant one has 1000 hp and deals 20 damage per second combatant 2 has 500 hp and deals 60 damage per second?

A good example of a failed combat system (PvP) is Everquest. A wizard trying to kill a warrior in Everquest is a joke. The outcome of the battle is almost completely predetermined by the equipment of the two combatants. If the combatants are competitively equipped the wizard will not have enough mana to kill the warrior. At no point should the wizard ever be at risk due to the 'fade' (random teleportation) spells in the wizard’s arsenal. The outcomes are the warrior dies, the warrior retreats, or the wizard retreats. There is little thinking or strategy in the whole process.

A good system was Ultima Online during the 'precast' time period. (After tab hiding and dex buff before the renaissance patch and removal of precast) A duel between most character types was often difficult to predict. A combination of the spell explosion, hit with a halberd and the spell ebolt left the inflicted with under 30% health left. The inflicted essentially had to react or die. This made duels last from 30 seconds to epic hour long battles.

In my eyes the ideal system would have to combine action/reaction, close calls, and quick thinking. The system should not favor those who play the most as they already have the benefit of more experience.

[mikedsc]

Could you explain "precast" for me? I've never played UO and I don't know where to find info on discarded features.

[Yxven]

Precast was a bug or 'feature' that allowed you to deal a good amount of damage in minimal time. The way UO's casting system worked was that you hit your macro for the spell, the spell words appear above the characters head, your character goes through the casting animation and then you have 10 seconds or so to select a target. The bug was that in the 10 or so seconds that you could target you could also equip a weapon and instantly hit (or miss) when in range due to another bug. The basic combo was cast explosion (a delayed direct damage spell) target, cast energy-bolt (a direct damage spell), equip halberd attempt to hit enemy with halberd, unequip halberd, target e-bolt. The target is most likely within an ebolt of death. All the damage was applied at the same time so you couldn’t heal half way through it. This was important because healing was very fast in that game. Ebolt casts slightly slower than greater heal and greater heal healed for more damage than ebolt did damage.

Essentially this 'feature' added an element of player skill to combat which pvpers love. Rock paper scissors and coin flipping balancing methods just don't get the heart pumping.

Now I'm not saying that you should intentionally try to bug your games. But the sort of fast paced action/reaction system uo accidentally created is what made uo one of the best mmorp pvp games to date.

My Everquest example could be better. Try a wizard vs rogue. Lets say the wizard is out hunting. The rogue catches him off guard and double back stabs the wizard. The wizard is now at 10% health. The wizard might live due to balance breaking fade line or just the general buggy melee code but he certainly can't turn around and win. The same thing works in reverse. The fight being determined by the first hit is not fun.

The better pvp mmorpgs typically have fast recovery systems that tend to create this 'close call' environment but I'm sure it can be done without that.

[mikedsc]

Very interesting. My current methodology is more a matter of detail. The idea is to slow down combat so you had a little more time to set all the facets of your attack. I've also made it so that all attacks have high damage likelihood, but it's not very easy to actually hit the other person. My current estimate per attack is around 10 seconds, and about 15 seconds per exchange between duelists. Oh, and I added attack time. (Meaning you say Attack! and it takes a few seconds for your sword to actually hit the other person, giving them some time to react defensively.)

The fight can be determined by the first hit (assuming no ambushs, which is a problematic issue for me), but because the first hit can create enough momentum so that the defender never has a chance. On the other hand, if the reactor is smart enough, he can exploit the first hitter by capitalizing on his vulnerability during the attack. So it's a matter of figuring out what kind of fighter you are: brute force or dextrous.

Actually, now that you mention it, purposefully making the combat system flawed is worth looking into... *hums innocently*