posted 12-03-2003 03:09 PM            

There is one thing that bugs me, though, and I must preface this by saying that I don't play GGXX nor am I wholly aware of its ins and outs, but from what I've gathered, there is a set of base defensive (and potentially offensive) tools afforded to all players. Thus, all characters start at the same level of ability before a single individual move is considered. In addition, each character has a very unique set of moves and properties (in addition to character design) that makes the cast feel very diverse and original.

My problem is that this is essentially how most fighting games are constructed, and their level of balance doesn't seem to benefit in the slightest. In Street Fighter, every character is allotted a base set of universal functionality, but the true measure of character worth lies in their individual moves and properties. If we were to apply an artificial value to the abilities intrinsic to GGXX, we would find that the relative effectiveness of each character is determined completely independent of the base set of systems (UNLESS - and this is potentially where my argument falls apart - individual characters use the base set of defensive abilities with varying levels of efficacy. I couldn't determine from the article if this was the case or not).

For the sake of argument, let's say that the sum value of the base defensive abilities in GGXX is 50. Now, let's (again, for the sake of argument) say that the total intrinsic capability of Sol minus these defensive capabilities is 50, and that Potemkin's is 25. This means that Sol is better than Potemkin, and he will always considered to be, regardless of the base set of abilities.

I would say that the only way to really achieve character balance is to 1. have a rich set of base abilities and to 2. reduce or increase the individual character's proficiency in utilizing these abilities to compensate for any natural imbalances that present themselves due to the vast and varied differences that each character will ultimately have. This is what Soul Calibur II did (albeit unsuccessfully), and is what Virtua Fighter has done for quite some time now (although VF's characters are also much less diverse than any other fighting game mentioned thus far).

posted 12-03-2003 06:01 PM            

Also, whatever happened to Keep It Simple Stupid (particularly regarding GGXX)? It seems like every layer of complexity you add would make a game more imbalanced rather than balanced. I get the impression that certain things were only added to compensate for other things, but every time you add a new meter you add a new potential loophole for the player to find, so you have to add a new counter, etc. Are they simply trying to overwhelm the player so that he/she can't hope to conprehend all of the weird interactions, e.g. compliance through ignorance?

posted 12-04-2003 12:06 AM               

I do not know what you meant by

"2. reduce or increase the individual character's proficiency in utilizing these abilities to compensate for any natural imbalances that present themselves due to the vast and varied differences that each character will ultimately have."

Maybe I can answer you anyway though. You seem to be saying "Let's say all characters have the same base defense stuff, and Sol's unfair stuff is worth a 7 and Potemkin's is worth a 5, so Sol is still better than Potemkin. The answer should be in making them each worth a 6 in their unique strengths."

Well...in pure theory, sure. But the entire problem in the first place is that it's pretty near impossible to give everyone wildly different stuff and balance that. So a good solution is not "well, just balance it anyway." haha. It takes years of play testing with all the best players to ever even figure out how unfair Potemkin's stuff is compared to Sol's. So yeah, the designers should do as good a job as they can there, but no one should realistically expect them to create wildly diverse and balanced characters. So...

That is where the shared defense abilities come in. Imagine a game where we don't have that base set of help to the defenders. Now imagine Sol is worth 7 for his b.s. and Potemkin is worth a 5. Sol is waaaaay better than Potemkin in this game. But when we add in the base set of defenses, it's a lot closer match. No matter how unfair Sol's stuff is, Potemkin has ways out, and vice versa. It's a way a smoothing out all the rough edges without having to know exactly what those rough edges are going to be ahead of time, as a designer.

So fighting games in general do this, it's a question of degree though. Soul Calibur's parry is a great example of shared defensive system that can get everyone, even slow Astaroth, out of trouble. But SC2 has it's own problems (knock down has fewer protections than any fighting game ever made, not to mention numerous bugs). Also, it, like vf, has very low character diversity. I'm sure all the 3d fans will write hate mail over that, but there just is no match as extreme as Dhalsim vs. Zangief (ST), Chipp vs. Potemkin (ggxx), or even Sentinel vs. Cable (MvC2). Chipp has to land like 10 combos to Potemkin's 1 combo to win!

I don't know what I'm even talking about anymore, so I'll sign off.

posted 12-04-2003 09:49 AM            

I am in complete agreement that a large set of basic skills, whether defensive or offensive, is good in raising the total sum of a character's worth higher than it would normally be in an environment where every aspect of the cast is dissimilar. Where I think this system breaks down is when every character can use the shared skills in the exact same manner; the worst character in the game can pushblock at the same level of proficiency as the best character in the game, etc. Basically this means that proportionally, the character balance is the same irrespective of the base set of abilities. To use my earlier numeric examples, Potemkin's 50 base skill plus his unique potential is 75, while Sol's base skill is also 50, but his unique potential is 50, yielding a total value of 100; his total sum being greater, he's still a stronger character.

But what if, as is the case in the aforementioned Soul Calibur II, the designers were to take a look at the base set of skills, and adjust those so that all characters did not use them to the same efficacy? Talim, for example (a note to anyone combing too finely over the details: I'm only using these examples as a "for instance," and not as a starter for debate on any particular game) has on average a very low damage output - let's give it a value of 10. Nightmare has a very high damage output; let's say a 15. So when Namco looks at the base set of skills, like a sidestep, they give Talim a very good one with a value of 15, and Nightmare a lesser one, with a value of 10. If these were the only two factors involved in gauging character ability, we would see that they are equal in total "value." This is becoming a popular trend in 3d games to create character balance while maintaining a level of diversity, albeit a dramatically lower one than you have in 2d fighters.

Virtua Fighter does it too. The standing jab - performed with the same input for all characters - is not the same speed for all characters in the game. Typically, the girls have the fastest, followed by the normal sized guys, with Jeffry and Wolf bringing up the rear with the slowest jabs (Vanessa doesn't fit into this example very well - her jab is on the slow side but she has another standing punch that is faster than even the girl's jab). To compensate for this imbalance, among other things Jeffry and Wolf float much lower than the other characters, making them intrinsically tougher to damage, while the girls launch particularly high, with most of them susceptible to particularly damaging combos only possible on them.

In this way, both games are able to make their best attempt at ensuring the characters stack up well against each other. These uniform systems allow them to hem and haw, tweak and balance without having to actually touch the individual character moves unless it's extremely warranted, presumable maintaining their diversity in the process. I think if the goal is to create both balance and diversity, then this is the best way to go about it.

Notes:

Itsatrap: I’ve never been a big fan of Namco’s seeming unwillingness to truly ensure a balanced and proper fighting system. Maybe it’s me being overly judgmental and harsh, and I don’t judge fighting games anymore by their “fun factor,” but rather their worth in competitive play, but all that chaos (particularly in TTT, TK4, and SC2) just doesn’t seem like an appropriate setting for true, fair and measurable play. Just my two cents.

Virtua Fighter is constantly updated based on player feedback. Virtua Fighter 4 and its sequel have had something close to six different iterations (although not all of them released to the public), all dedicated to finding the right character balance. This too is a crucial element to achieving this goal.

posted 12-04-2003 10:36 PM            

Then again, I've played Marvel Vs. Capcom 2, 3, and SF already, so maybe that's a silly point. Anyway, Sirlin, now that you (and I'm sure many others) have mapped out what I can only see as every possible variety in basic moves in 2 and 3D fighting games, do you have a desire to do the same for other genres? Like, say, the Super Mario 64/Sunshine one? Or more importantly, would mapping the basics of any other more "open" or "interactive" genre have any merit?

posted 12-05-2003 08:01 AM               

"...but from what I've gathered, there is a set of base defensive (and potentially offensive) tools afforded to all players.

...

(UNLESS - and this is potentially where my argument falls apart - individual characters use the base set of defensive abilities with varying levels of efficacy. I couldn't determine from the article if this was the case or not)."

I think I can help clarify this a little. The purely defensive base skills: Burst, Green Blocking and Instant Blocking (not covered in the article) are the same for all characters. If I pick Zappa (arguably, the worst character) I know that I'll be able to Green Block with the same profeciency as Eddie (arguably, the best).

However, most of the other abilities (the forward+p moves, super meter, Roman Canceling) are very character centric. Each character's f+p moves have very different invulnerabiltiy windows, some gain meter faster or slower than others, and only a certain number of moves can be False Roman Canceled. These are what are tweaked to complement each character.

This is a Good Thing. The basic ability to get yourself out of trouble really should be the same for all of the characters. This lets the player focus on the strengths and weaknesses of the character itself, rather having to relearn a basic subsystem every time.

posted 12-06-2003 06:19 PM               

Another very competitive game that never becamse successfull because of its high learning curve is Tribes. In high level Tribes 1 and 2, offensive coordinators in teams ranging from 10-16 people per side continually instruct the offensive players in role assignments and the players organize small coordinated strikes amongst themselves. Similarly to counter-strike, defensive strategies are usually just role assignments such as light flag defense or heavy generator defense. A armor role is given to give a general idea of the type of tactics to use, but the execution is still left up to the player.

Both counter-strike and tribes have a diverse amount of weapons and tactics available to the offensive players while defensive play is pretty standardized among players. It is interesting that your concept from competitive 1v1 games holds true for the strategies involved in games played 16v16 at their highest levels.

posted 12-12-2003 03:17 AM               

There would definitely be merit to breaking down other games such as platform games like Mario64/Sunshine into the design components that make them what they are. That is a hot topic among game designers these days called "design patterns". It's the idea of taking elements of games, and describing patterns of those elements that work well together. For example, "collecting coins, exploration, no time limit, central level leading to many other levels" are elements of Mario, to name a few. It turns out that "collecting coins" and "exploration" form a pattern that works well in lots of games. Anyway, yes it would be a great thing to write about, and someday I will do just that.

I'm not sure what you're asking about yomi in other games. It exists in practically any multiplayer game. Poker, Magic, chess, counter-strike, tennis, etc. Any time you have an opponent and ways to counter that opponent, yomi should be a pretty key ingredient in the mix.

posted 12-12-2003 03:23 AM               

The assumption all this balance stuff rests upon is that it's impossible to forsee how character balance in a fighting game is going to pan out before you release the game (assuming you have any kind of character variety at all). Arguing with that assumption is pretty crazy, so let's assume you agree.

In the balancing phase (before the game is released) the designers will do the best they can to tweak this move or that to balance things, but since they can't really know for sure who is the best character, for example, my theory here is that it's better that they just leave all the defensive abilities the same unless there is some overwhelming reason not to.

What if the designers thought Sol was the best in ggxx before it was released? Certainly a plausible thing for them to have believed. They could very well have taken away his ability to green block, or something. Luckily, they left defensive systems alone, and it turns out Sol isn't even as good as Slayer, Eddie, Millia, or Faust. When faced with an unknowable problem like this, I think it's best to leave safeguards in place, such as the universal ability to green block, double jump, increased gravity/decreased hitstun during a combo, etc.

posted 12-21-2003 02:13 AM            

quote:

---

Originally posted by Sirlin:
dupin,
that make them what they are. That is a hot topic among game designers these days called "design patterns".

---

quote:

---

I'm not sure what you're asking about yomi in other games. It exists in practically any multiplayer game. Poker, Magic, chess, counter-strike, tennis, etc. Any time you have an opponent and ways to counter that opponent, yomi should be a pretty key ingredient in the mix.
--Sirlin

---

I disagree with that, as a blanket statement about yomi being present in any multiplayer game, well specifically with the example of chess at least. It's definitely present in poker (heavily dependent on yomi skill in fact), but not games like chess, and go, where there is perfect information (both players can see what's going on). I think if you try the yomi tactic in chess, and start hoping your opponent makes a certain move, you can be in for severe disappointment. These perfect information games are still interesting without yomi, due to the computational intensity and strategy in positioning the pieces alone, and so when you try a little yomi, it's more likely that you wind up _assuming_ that your opponent will play right into your traps because you try to weed out as many possibilities and nodes as you much as you can but then suffer for it when you're wrong. And in chess and go that's 1) going to be more often (they can see whatever you can, and have the same distractions that you do), and 2) more problematic for the one who started trying to yomu in the first place, because the consequences are usually an undesireable position/structure that you now have to deal with the best you can- no chance to fold a hand to get another deal, until the game is over. The best tactic is to prepare and organize a strong offensive strategy, with some heuristics like going for control of the center of the board, etc. and while you don't exactly want to announce your gameplan outloud of how you hope to accomplish these, you are spending a lot of processing time watching out for weaknesses in your structure that your opponent can exploit. You can take a risk and hope your opponent doesn't see it, but if he does, you tend to lose a lot more than other options.

In go, one of the oldest games ever, there are lots of proverbs, one of which is, when deciding where to place your next stone, choose the location where your opponent would place his if you don't go there first. This is different from yomi, where you have to make an assumption. The go/chess move is more of a comprehesive one, in the sense that you are playing against your own comprehension of the game thus far, regardless to whether or not your opponent would have any thought of going there at all. If you start making assumptions, it compromises your strategy and you have to live with the resulting position and structure from that point on to the end of the game.

I guess tennis definitely has yomi... at least, when you can make the ball do what you want it to do, and go where you want it to go, at what speed and spin but it's all dependent upon (and limited by) your technique. I guess that's not too different from fighting games but if tennis had some kind of "energy meter" like in fighting games, ie each player could know exactly how much energy the other had during a match, it would have a little less yomi to it all? You often get a sense of this though... as wonderfully portrayed in the movie The Royal Tenenbaums, where the tennis player decided to take off his shoes during a match... and, when going for a return, listlessly lobbed his racket at the ball, practically hoping his tool in the game would do his job for him, even to the point of utter irresponsibility of letting go of the damn thing and dropping his racket altogether.

I have actually considered a technique of throwing the racket at the ball in tennis, eg a return that you couldn't reach in time... do the rules even allow that? Of course, your technique and mastery of the physics involved in such a maneuver would have to be supreme, and if playing doubles, you could risk taking out half your army if you accidentally hit them. Also it would all have to be planned as the final hit for that serve, say, best employed in a yomi situation where you get them to think they have to be at the back of their court, but you hit it to bounce just barely over the net in the front. (The real trick wouldn't even necessarily be doing this trick shot- it would be doing it to get the ball over successfully, and _then_ having enough time to pick up your racket again before they can return it.) Also, do the rules state that you have to hit the ball with the wire mesh area of your racket, specifically, or can it be like the metal rim? which happens by mistake a lot, technically resutling in a subtle yet accepted "double-hit"... Because, you could do like a "pool shot" return or something, using the butt of the racket as the tip of your pool stick, to make them _really_ intimidated. Practicing that one would usually mean getting hit with the ball a lot though, I guess. Which, I believe, is _their_ point!

In Japan, they're really good at fighting games _and_ tennis because of mastery of these kinds of techniques. Here is some footage of a Japanese ping pong match to give you an idea:
http://www.ntv.co.jp/channel/asx/hkzkt10.asx

posted 12-22-2003 07:30 PM            

quote:

---

I disagree with that, as a blanket statement about yomi being present in any multiplayer game, well specifically with the example of chess at least. It's definitely present in poker (heavily dependent on yomi skill in fact), but not games like chess, and go, where there is perfect information (both players can see what's going on). I think if you try the yomi tactic in chess, and start hoping your opponent makes a certain move, you can be in for severe disappointment. These perfect information games are still interesting without yomi, due to the computational intensity and strategy in positioning the pieces alone, and so when you try a little yomi, it's more likely that you wind up _assuming_ that your opponent will play right into your traps because you try to weed out as many possibilities and nodes as you much as you can but then suffer for it when you're wrong. And in chess and go that's 1) going to be more often (they can see whatever you can, and have the same distractions that you do), and 2) more problematic for the one who started trying to yomu in the first place, because the consequences are usually an undesireable position/structure that you now have to deal with the best you can- no chance to fold a hand to get another deal, until the game is over. The best tactic is to prepare and organize a strong offensive strategy, with some heuristics like going for control of the center of the board, etc. and while you don't exactly want to announce your gameplan outloud of how you hope to accomplish these, you are spending a lot of processing time watching out for weaknesses in your structure that your opponent can exploit. You can take a risk and hope your opponent doesn't see it, but if he does, you tend to lose a lot more than other options.

---

I agree with you in saying that yomi existing in all multiplayer games may not be a completely correct statement, but yomi certainly exists in chess. While chess players have the disadvantage of having all their moves exposed and known to the opponent, the game is all about yomi. To be successful, you have to anticipate your opponents moves, but you have to do it for as many different possibilities as you can. Hopefully then, you can make a move that will prepare you for the most of these possibilities, which of course leaves you open to some risk. It's more like yomi to the extreme, since you have to not only anticipate what kinds of moves your opponent will make, but you have to do it for an almost limitless ammount of possibilites, as well as doing it in a manner which will outmanuver your opponent's yomi.

I guess instead of a 'paper/rock/scissors' idea of yomi that some fighting games have, chess is more of a 'paper/rock/scissors/mace/gun/sword/etc/etc...,' but yomi still exists, if in a slightly different form.