Balance of text-based combat system

Started by
13 comments, last by Kylotan 21 years ago
I just want some feedback on a potential idea I may use in my MUD (text-based internet game): - You can acquire a variety of skills that correspond to ''fighting styles''. Each style comprises a small number of techniques that describe and define it. (eg. ''Basic swordfighting'' might include ''basic slashing weapons'', ''basic stabbing weapons'', and ''parrying''.) These techniques award fairly significant bonuses to the person who uses the styles that own them. Some are geared towards defense, some towards attack, others for speed, damage, etc. Crucially, many of these techniques are shared across different styles. - Each character is able to use a small number (eg. 1-3) of these styles simultaneously, thus getting cumulative benefits. So you might generally have 3 to 10 techniques active at once, with most people hovering around the 5 or 6 mark. - However, if your opponent knows a style containing a technique you are currently using, the benefit you get from that technique is reduced by 50%. Additionally, if your opponent is also actively using a style with that technique, then you lose the other 50%. (In other words, the bonuses for that technique would cancel out on both sides.) So yes, it''s glorified rock-paper-scissors, except with the random aspect reduced. The main benefit of this system is that due to the cancelling-out effect, it avoids the spiralling statistic problems of many RPG-type games in that evenly-matched opponents will always have ''average'' stats when fighting each other. Yet a master fighting a newbie can have very high stats due to being able to use esoteric styles that the newbie does not know (and therefore, cannot cancel out). Secondly, it adds a degree of strategy (not usually found in this type of game) in that anticipating and reacting to an opponent''s strengths and weaknesses will really pay off. (eg. You could take an adventuring party equipped with flails to a land where they never fight with flails and are thus less capable of defending against them.) The main disadvantage I see is that, if there are too many techniques, then everybody has a load of bonuses, nothing cancels out, and the system is largely pointless. And if there are too few techniques, then everybody will have the same ones and choices are pretty much made for you. I suppose the key is to spread out the techniques across styles that are progressively harder to obtain, but I''m pessimistic about being able to get this right first time. The second disadvantage is that it is a fairly passive system - apart from changing styles to match your opponent, you don''t do much. I had once toyed with a system that worked on a blow-by-blow level, but it seems hard to do in the semi-realtime environment of online text-games (too slow for Soulcalibur style moves, too fast for most detailed turn-based systems). Mainly I''m interested in comments on the pros and cons listed above, and any ideas for or amendments to the system. [ MSVC Fixes | STL Docs | SDL | Game AI | Sockets | C++ Faq Lite | Boost
Asking Questions | Organising code files | My stuff | Tiny XML | STLPort]
Advertisement
Instead of just losing half the bonus a style provides when the opponent knows the style, it should depend on how well the oponent knows the style relative to how well you know the style.

A master of TechFu would lose little of its benefits when fighting a novice that also knows TechFu, and the master might know an advanced technique that helps counter the novice even more, so the novice is actually at a disadvantage because he is using a style his opponent knows better than him.

Or, maybe the master of TechFu also knows TekFoo, which is a defensive style developed in response to TechFu that has several counters for each TechFu move, so it gives him double the bonus over the novice because he can both anticipate the attacks and knows special techniques to counter them. When fighting someone less skilled in either TechFu or TekFoo, the master can anticipate their reaction and use it to his advantage (not only nullifying their bonus, but possibly turning it into a penalty).

Having styles that are naturally more effective vs other styles would be an interesting addition and would make it even more strategic. Instead of having mirror matches, you could use a primary style of your choice and for your secondary style choose the natural counter to your opponents primary style. It would help vary things up a little more at least.
"Walk not the trodden path, for it has borne it's burden." -John, Flying Monk
quote:Original post by Extrarius
Instead of just losing half the bonus a style provides when the opponent knows the style, it should depend on how well the oponent knows the style relative to how well you know the style.

Yeah... differing degrees of learning will be in the system in some form, but I omitted that to keep the description clear and brief. Also see the next paragraph...

quote:A master of TechFu would lose little of its benefits when fighting a novice that also knows TechFu, and the master might know an advanced technique that helps counter the novice even more, so the novice is actually at a disadvantage because he is using a style his opponent knows better than him.

Yes, however, I am trying to move away from the traditional model of comparing 2 values that correspond to character skills. Or to look at it another way, I''m transferring some of the gameplay from character skills (represented by the percentage attached to a skill name) to player skills (represented by the player''s intelligent choice of what to use).

In the above case, the ''master'' would maybe have 10 styles to choose from and the ''novice'' would have 2. So although the master''s first choice of TechFu might be largely countered by the novice, the master can keep switching styles until he finds one that the novice cannot handle, which will result in the novice being quickly overwhelmed.

[ MSVC Fixes | STL Docs | SDL | Game AI | Sockets | C++ Faq Lite | Boost
Asking Questions | Organising code files | My stuff | Tiny XML | STLPort]
Rock paper scissors may be uninteresting, but loading things slightly and putting it in a larger picture is much more interesting. Kung fu style example (without too much knowledge of actual kung fu):

Crane
4 Damage
Cancels opponent tiger damage

Tiger
8 Damage
(doesn''t need to cancel, already advantagous over monkey)

Monkey
2 Damage
Cancels opponent crane damage


having multiple styles at once seems illogical to me.
These attacks could have more special things with training (when very good at monkey it would halve the damage of the opponent the next turn, when good at tiger it would double or multiple by 1.5 every time after a tiger and crane could grow per succesful hit)

Each of these specials introduce new thinks to brood on. Someone very good at crane for example would want to use it in succession, but could also use tiger to be ahead and prevent opponent monkey.

Anyways that''s just an example that a slightly modified rock paper scissors can be VERY interesting (and psychological so more fun for PvP)


Your strategy seems to be based on preperation and the way the character has developed though, which is uninteresting in a way, because if 2 characters fight 10 times in succession, it''s likely that the same character would win at least 9 times.
I hate to bring this up, but isn''t think kind of like Pokemon on gameboy? You have different elemental types (styles, if you will) that are much stronger against others, or perhaps weaker. For example, a fire monster takes a lot more damage from water, but does far more against grass. To counter an enemy "style" (elemental type), I can bring another Pokemon. I can even use abilities to change the environment to suit my "style" (again, elemental type)

Anyway, I don''t play Pokemon anymore (thank God ) and I guess this elemental system is really true of most RPGs.

Your idea seems similar, though it has a nice feel to it, I think. Perhaps what you need is a limited number of skill points, with a tech tree progession scheme. Players could only invest so many skills. For example, a given character might be able to learn 50 skills, but a tech tree might have 200. This gives a lot of variety to pick from, but limit the number they could pick.

Goign even further, the tree could have four main braches or something -- mind, speed, strength, and endurance (whatever) that have moves that correspond to that type. These four trees could be handled in Paper-Scissors-Rock fashion. Players could decide if they wanted to be well-rounded, or train to be an expert in a given field, and hope not to find an opponent using techniques of their "opposite".

But whatever, I like your idea; hope to see more

Peon
Peon
The best martial arts system I've seen for PPRPG's (which should translate firly easily to MUD's) was in Champion's supplement called Ninja Hero. Basically, a martial art style was composed of techniques. A "black belt" would generally have about 5-6 techniques out of donzens available in the style. Each technique offered various damage ratings, how open it left you to attack, and how open it left the defender open for further attacks amongst many others.

The Hero system used something called OCV and DCV (offensive and defensive combat values respectively) to help determine if you hit someone. Basically, you added your OCV bonus, and substracted your opponent's DCV value to your to-hit roll. A difficult attack like a flying twisting jump kick may be a minus to your OCV, but actually give you a bonus to your DCV (it's harder to hit a jumping twisting guy than a stationary one) and it would also give you a damage bonus and a knockback bonus.

There were even skills that let you analyze your opponent's style so that you could gain either an offensive combat bonus or defensive combat bonus after watching them fight for a few turns. So the idea was that you tried to figure out what kind of attacks he had available to him. You could also set aside so many character points to be put in general skill levels which modified your OCV or DCV (for example, if you bought 3 levels, you could put +2 in your PCV, and +1 in your DCV) and the cost of the levels was determined by the range of techniques it was applicable to (a 3pt/level might be restricted to only arm techniques, while an 8pt level could apply to all melee attacks including with weapons).

The idea was to learn a set of techniques that covered a good range of possibilities. And techniques weren't just offensive. They could be something like Evade, which gave you a +5 to your DCV, but no offensive capabilities, or Parry, which gave you a +3OCV, and in the next turn, gave you a +2 OCV to attack. There were also moves like disarms, grabs (joint locks and immobilizations) and takedowns/throws. And you could also buy skills that allowed you to use your unarmed techniques with some kinds of weapons.

Maybe what you can do is have tons of techniques, and then let the players several techniques to put into their combat style. This way, other players can't go, "aha, he's using Southern Lightning Long Fist...I'll counter that with my Shattering Omei Thunder Leg style!!". This way, player won't really know what the other player is capable of until he starts fighting with him for awhile. And while most styles require you to learn a certain way of performing techniques....in reality most styles teach the same basic methods as virtually every other style. Virtually every style has kicks, punches, elbow, knee, head, joint locks, grabs, and takedowns/throws. The difference is in how they are executed, as well as the philosophy of combat. And some styles specialize in some techniques over others (Savate for legs, Shuai Ciao with throws, Aikido with joint locks, and even some specialized styles for unique circumstances like Capoeira which would have no penalties for being manacled, or iaido for fighting while sitting).

This way, instead of encapsulating the fighting capabilities within a set style, you can "build" your style through various techniques. What you may want to do is have each style contain maybe 20 techniques, of which the player starts off with 5-6, and can gain more. The player can also pick up skills from other styles once he reaches a certain level (and I wouldn't restrict it to just basic techniques...just like most programming languages have many features in common that allow you to pick up another programming language somewhat easily, martial techniques are the same).

[edited by - dauntless on April 14, 2003 11:18:52 PM]
The world has achieved brilliance without wisdom, power without conscience. Ours is a world of nuclear giants and ethical infants. We know more about war than we know about peace, more about killing than we know about living. We have grasped the mystery of the atom and rejected the Sermon on the Mount." - General Omar Bradley
If you wanna rethink that blow-by-blow scenario I suggest you check out the Dragon Ball FE clone DBZ: Anniliation. It''s still in beta but the basics are there. You attack opponent times the appropriate counter (parry high/low, deflect, dodge). Count to 3 for normal blows, 5 for specials (elbow, rougafufuken, kamehameha)
quote:Original post by Taco
having multiple styles at once seems illogical to me.

With your example, yeah. But imagine "basic karate + advanced dagger" or "one-handed sword + shield". Many combinations would make sense. I'd obviously name the styles to try and make this so.

The circular RPS example you gave would well exist in some form.
However the directly opposing aspect where both players employ style X so that neither get the benefit is important to the game balance by keeping the stats around a certain base level.

quote:Your strategy seems to be based on preperation and the way the character has developed though, which is uninteresting in a way, because if 2 characters fight 10 times in succession, it's likely that the same character would win at least 9 times.

In most RPGs, if 2 characters fight 10 times, unless they are amazingly evenly matched, the same one wins 10 times Or it comes down to a random factor, meaning you only succeed through luck. By introducing more player skill, you make that 1 time out of 10 possible, and you make it possible through a player choice rather than a dice roll.

quote:Original post by Peon
I hate to bring this up, but isn't think kind of like Pokemon on gameboy?

*WHACK*!!
Nah, you're probably right. Just as long as I don't have barbarians yelling "Pika pika!" when charging an orc horde, I don't mind too much. The elemental system applies in some way to things like Magic the Gathering, too.

quote:Your idea seems similar, though it has a nice feel to it, I think. Perhaps what you need is a limited number of skill points, with a tech tree progession scheme. Players could only invest so many skills. For example, a given character might be able to learn 50 skills, but a tech tree might have 200. This gives a lot of variety to pick from, but limit the number they could pick.

Skill points would be limited by a variety of factors. Combat is only one element (thankfully) so people won't just be min/maxing the combat skills. But some sort of skill tree may well exist. As for a 'maxcap' on the number of skills that can be learned, it's more likely to be a cap on the total skill percentage (so you can learn a few really well, or a lot to a lesser degree). I'm going to see if I can get away without this however, as acquiring and raising skills will be quite difficult in my game.

quote:Original post by Dauntless
A "black belt" would generally have about 5-6 techniques out of dozens available in the style. Each technique offered various damage ratings, how open it left you to attack, and how open it left the defender open for further attacks amongst many others.
...
The idea was to learn a set of techniques that covered a good range of possibilities. And techniques weren't just offensive. They could be something like Evade, which gave you a +5 to your DCV, but no offensive capabilities, or Parry, which gave you a +3OCV, and in the next turn, gave you a +2 OCV to attack. There were also moves like disarms, grabs (joint locks and immobilizations) and takedowns/throws.

I like the detail in this, although there would be problems in adapting it to my game.

Firstly, it's played by text, across the internet. This means you can't react as quickly as you might even on other internet games, as not only do you have the ping of 200 - 1000ms, but you have to be able to read and digest what's going on too. On the other hand, you can't take too long over a move, because other people are waiting. For this reason, I wanted to look at things more abstractly - ie. in terms of styles rather than in terms of individual attacks/defences. However, I would consider changing if I was convinced it could work.

Secondly, my game is largely based around medieval weapons fighting rather than elaborate Oriental martial arts. This means I have less of a body of research to draw from, meaning it would be very hard to generate enough interesting techniques for all the various weapon types.

It was also not obvious from the example you described how you would be able to effectively react to an opponent's choice of techniques. Obviously a few would counter others by virtue of the stat modifiers cancelling out, but in general I don't think you'd see that. You would also see 'black belts' having fairly high scores, even versus other 'black belts', which (although not critical) is something I'm interested in avoiding.

quote:Original post by PSWind
If you wanna rethink that blow-by-blow scenario I suggest you check out the Dragon Ball FE clone DBZ: Anniliation.

I think I would rather give up computer gaming than be at all involved with anything DBZ-related I would at least check out the website, if I could actually find one on Google... any hints?

[ MSVC Fixes | STL Docs | SDL | Game AI | Sockets | C++ Faq Lite | Boost
Asking Questions | Organising code files | My stuff | Tiny XML | STLPort]

[edited by - Kylotan on April 15, 2003 7:37:03 AM]
Kylotan-
In Champions (Ninja Heroes) there are moves which you can use as "reactionary" moves. These are almost always defensive moves like, parry, block, dodge, evade, counter throws, joint locks, escape or something like that. Champions uses a turn sequencing system based off of your character''s speed. There are 12 phases in a turn, and depending on your character''s speed, he gets to act in various phases. So for example, a speed 7 character gets 7 phases per turn (and you can sometimes do more than one thing in a phase, like do a half move with a tackle, or a half move with a dodge) while a speed 5 character gets 5 phases per turn.

Therefore it''s possible that a higher speed character will have more actions in a turn. But this doesn''t mean the speed 7 guy gets two free licks at the speed 5 guy. Whatever your last action was, that''s what your DCV value is for the remaining phases. So let''s say the spd 5 guy has used up all of his actions, and the spd 7 guy has one action left. If spd 5 guy''s last action was evade which gave him a +5DCV, then even though he has no actions, he still gets the +5DCV.

What you can do with reactive techniques is forfeit your next action phase for the reactive technique. Let''s say spd 5 guy gets to act on phase 7, but spd 7 guy acts on phase 6. Unfortunately, the last technique spd 5 guy did was a joint-lock which left him at a -2DCV. Spd 5 guy sees spd 7 guy about to attack. Originally, he was planning on moving, but spd 7 guy was too fast. So instead, he opts for one of his reactive techniques....a throw. This is where it gets tricky. A throw is both a defensive and an offensive move. The spd guy still gets to go first on phase6, however, the spd 5 guy''s DCV is now +1 instead of -2. If spd7 guy hits, any damage is resolved and it''s ill effects. Once this has been done, spd 5 guy now gets to try his throw technique against whatever DCV value his opponents DCV value was for the technique he just used.

Of course, spd 7 guy can always use a reactive technique himself (which is a good idea if he used an attack which puts him at a defensive disadvantage), forfeiting his next actions move. You can not however use reactive techniques once all your phases have been used up. Also, some techniques preclude the use of reactive techniques for the player''s next action. For example, certain jumping techniques would disallow reactive techniques...afterall, once you''re in the air, you have to follow through on what you''re doing and it''s hard to suddenly change. And IIRC, you can not "store" up your phases to use them all at the end of the turn...you use it, or you lose it (or you use a reactive technique).

I should also note that every character had brawling techniques...a sort of instinctive fighting skill. Then you had techniques which were more advanced and powerful from your fighting style. Since you are doing this from a western melee style combat, you can still borrow many ideas. Even in fencing for example, there were myriads of styles of schools on how to fight (spanish and hungarian being well noted school...or salon....styles). There are German manuscripts dating from the 16th century on the use of one and two handed swords, and supposedly there''s been a viking axe style handed down for generations amongst some Scandinavians. And I don''t see why europeans (or regular euro-style fantasy) settings can''t have hand to hand martial arts. Savate (aka Boxe Francaise) was a french martial art style started in the 1700''s amongst pirates who inhabited the Marseilles and Riveria regions. It combined sword and knife work (and adapted to the street cane) but became most famous for its leg techniques (which are the equal of any eastern style). Pankration, which is perhaps the oldest fighting style recorded anywhere (older than Greco-Roman wrestling), was a Greek fighting style that combined wrestling with boxing. Many of Alexander the Great''s troops were known to have been Pankrationists, and some speculate that when he invaded India, it may have influenced some Indian styles. I don''t see why cultures in a fantasy setting couldn''t come up with some neat and unsual fighting styles of their own
The world has achieved brilliance without wisdom, power without conscience. Ours is a world of nuclear giants and ethical infants. We know more about war than we know about peace, more about killing than we know about living. We have grasped the mystery of the atom and rejected the Sermon on the Mount." - General Omar Bradley
quote:Original post by Kylotan
With your example, yeah. But imagine "basic karate + advanced dagger" or "one-handed sword + shield". Many combinations would make sense. I''d obviously name the styles to try and make this so.

The circular RPS example you gave would well exist in some form.
However the directly opposing aspect where both players employ style X so that neither get the benefit is important to the game balance by keeping the stats around a certain base level.

In most RPGs, if 2 characters fight 10 times, unless they are amazingly evenly matched, the same one wins 10 times Or it comes down to a random factor, meaning you only succeed through luck. By introducing more player skill, you make that 1 time out of 10 possible, and you make it possible through a player choice rather than a dice roll.


ok the styles make sense to me now, somewhat, I still don''t see how similar styles should cancel each other out though... if you both fight with a shield, you both have a harder timing hitting the other, no?


And on the note of RPG 10 out of 10 wins... would you really want a game to be that way? Is it really fun to have the combat victory depend on all your preperations and still require you to do it yourself? I''m not saying preperation and skill shouldn''t have effect, but I''m getting tired of it being the factor that decides victory. Of course my example given above is most interesting against a human player, as there is no psyche of a computer to deduct.

This topic is closed to new replies.

Advertisement