Fighting Games

20080212064049!Ryu-nxcfixConsidering that I’ve mentioned that I classify execution as an “ambiguity sauce” – meaning, a cheap, bad way of introducing ambiguity/uncertainty into your system, you might be surprised to know that I have an extensive background playing fighting games, as well as other real time games.

In fact, I’ve played far more hours, more seriously, of real time games in general. The games I’ve played more than any others, hours-wise, would be Super Smash Bros. (N64), Super Smash Bros. Melee, Team Fortress 2, Doom, Doom 2, Quake, Warcraft 2, Starcraft, Warcraft 3, Starcraft 2.  I don’t think any turn-based game has come close to to catching up with those yet.

I’ve also had numerous 2D fighter and 3D fighter phases, but these have been steadily declining as I’ve gotten older. When I was 12, I thought that the huge move lists and difficult-to-execute inputs were “cool”. In fact, this is the whole reason that anyone really liked Mortal Kombat over Street Fighter, the latter being a vastly better game but the former having “fatalities”, a sequence of the game that simply allowed you to show the other player that you had memorized and drilled a sequence of button presses. That brings me to the state of the genre as I see it.

I see the 2D fighter and 3D fighter worlds as very stagnant. Now, people who play these games seriously and have never stopped – these are “fighter people”. To them, the differences between 2012’s Marvel Vs. Capcom 3 vs SNK vs Tatsunoko and 2013’s Capcom 2 vs Marvel vs Pokemon Mega Battle vs Little Mermaid are huge. To them, the addition of a new bar, or the subtraction of a bar, or the addition of some kind of move you can do that swaps an on-screen bar with another bar – these are the kinds of changes that most people would expect to see in a sequel, but this group has been conditioned to see additions/subtractions as something akin to from-the-ground-up innovation. It’s just the same as a “metal-head” being able to listen to hundreds of nearly-identical heavy metal albums – when you reduce your perspective to a small, narrow field, small differences pop out at you. This is fine to do sometimes, by the way; we should all be able to do this to appreciate small differences, because sometimes the small differences do really matter.

But if you take a step back and look at fighting games as just another kind of game, they’re… kind of all the same. One of the best ways to point out how they’re mostly all the same is to point out a few examples that are actually different. The two that quickly spring to mind for me are the well-known Super Smash Bros. and the less-known PSX game Bushido Blade.  A big thing that makes these games really, fundamentally different is that their goals are different.  Their goals are not to “reduce the HP sack to zero”.


The Formula

Firstly, what is the formula?  I’d say it’s one really solidified by Street Fighter 2:

  • Hit Points (Health Bars)
  • Flat, rectangular stage
  • A certain ratio of character size to stage size
  • Jumping (specifically, high enough to clear another human being’s head)
  • high and low attacks
  • a LARGE number of moves
  • Special moves that require complicated button inputs
  • Asymmetrical forces

Later additions to the Formula included “more bars and counters” for special, super special, and super ULTRA special moves, and a number of other features such as Street Fighter 3‘s parrying system.  The 3D fighter added, basically, a “side-step”, that’s roughly equivalent to a special kind of dodge (a circular kick or other circular attack might still hit you, for instance) and a few other proclivities, but the basic gameplay kernels are the same.  (EDIT: I should also mention that most 3D fighters do add a ring-out thing, which is an excellent addition… it just needs to be supported by more of the system)

I mentioned above two games that I see as really having broken the formula – Bushido Blade and Super Smash Bros. In Bushido Blade, the goal is to land a single killing blow.  Most of a match will be trying to either knock his weapon aside to get a clear shot, or moving around and zoning. Some matches take several minutes to complete, but many are done in under five seconds.  Further, the game has some other very interesting concepts, such as disabling an arm or leg – crippling the opponent for the rest of the match.  I’m not really endorsing this feature, as it does have a power-snowball issue with it, but it’s interesting, at least. This is a neat way for health to actually mean something to the current game state beyond just “who has more health”. Finally – and I’m not sure this is a good idea, but it’s certainly innovative – there are a number of weapons in the game, and characters, and you can pair any weapon with any character. Certainly interesting, although probably difficult to balance, especially given that the weapons range from a small knife to a super long polearm weapon.

(Before I go on, there are also some creative entries that could sort of be considered in the fighting-game genre and do things differently. One of my favorite examples would be a PSX game called Poy-Poy, a game that I’m always hoping I’ll have a reason to mention).


Super Smash Brothers

Super Smash Bros. for Nintendo 64 is probably the most sophisticated and advanced game design approach to “a fighting game” that has ever happened. Before I explain why this is, know that this is not a shining endorsement of this game on the whole.  I’m fully aware of some major higher-level problems with the game such as infinite combos (combos that, once you get the other player in them, you can hit them repeatedly and they have no way of getting out), some character balance imperfections, and Z-cancelling (the cooldown on a few big moves can be cancelled by tapping Z at the right time;  most probably a bug). Also “little hops” are bad in that they’re very hard to execute and extremely important to use.

But those are higher-level issues, most of which could (and really, should) be fixed in a patch of some kind (perhaps a 2nd Edition could be released on the Wii store or whatever).  Even if you hate Super Smash Bros. though, that’s fine. What’s important to recognize, though, is it is a game that had a tinge of a design philosophy behind it.

Super Smash Bros. is a game all about positioning.  Everything you do either makes you, or your opponent change position (the one exception being the “block” move).  If you attack, it knocks your opponent back. If you dodge, it moves you. Obviously, if you move, you’re changing position. Those are basically all you can do in the game.  The objective is to knock the opponent out of the level (basically to some lines that are off-screen).

Instead of “hit points” that you have to reduce to zero, there is a resource that is explicitly tied to positioning – “Damage %”. I don’t know why they used a “%” symbol, because nothing in particular happens when you reach 100, except that I guess it gets quite likely that you’ll die around 100%, but it can go way higher than 100, so it’s not really a “percentage” of anything. Regardless, what it is is a number that increases the amount you’ll get knocked back when you take a hit. Further, taking hits increases this number.

What this means is that you have a very dynamic health system which means a dramatically different thing based on where you are on a given level. If you’re already up high, you could be killed with maybe just 60% damage if the other player knocks you upward.

Instead of a rectangle for a stage, stages have some level of variation to them (unfortunately, most of them have random hazard effects, and even more unfortunately, they can’t be turned off, which is something to look at for the 2nd Edition!).  Levels have platforms and ramps and other shapes, as well as moving objects that you can stand on or bumpers or other environmental factors.

This chart shows how I think the series progressed after the original game came out.  They basically started “backing off” this design and moving towards being a regular fighter.  Granted, they’re still more different than most MvsCvsMvsM 2D fighters are from each other, but they’re moving in the direction of sameness.

smash brothers

When they released a sequel to Super Smash Brothers for the Gamecube, all of the moves became a little less about “changing the opponent’s position”, and a little bit more about “just increasing his damage %”. Further, positioning simply matters less because of air-dodging (in place or in a direction of your choice!) and more “directional influence”, where you can tilt the controller in a direction you’d like to move even if your character is technically out of your control. Hitting characters feels less like hitting a baseball with a bat and more like hitting a beach ball with a wiffle bat.

And then the next sequel, Brawl for the Wii, just totally went all-out on ruining that original vision.  Now the game is best described as “trying to push a bowling ball towards your opponent by whipping it with a shoelace”. It’s spammy, flailing nonsense, and just to make sure that everyone knows that the developer is only interested in making a great babysitting tool, they included random “tripping”, where your character falls down on their face some small percentage of the time that you try to run. This disables any would-be apologists from being able to defend this game, which is actually pretty nice.

The Potential Fighting Game is a direction that I personally want to explore with a future game. I don’t have a design document or anything, but I do have a way of thinking that will guide me to eventually write one.



To me, all of the Smash Bros. games, while much simpler than even Street Fighter 2 (one of the simplest 2D fighters), are still too inherently complex (mass inherent complexity being another ambiguity sauce).  There are too many different unique moves you can do and too many characters.  In short, to play well, you just have to know a massive amount of numbers and shapes and basically just rules.  This, more than anything else, turns me – a person actively interested in playing a fighting game – off from wanting to learn to play new ones at all.

What I wish designers of fighting games would realize is that you don’t need to do that!  You have continuous space and continuous time!  That means you already have a huge  possibility space to work with, right away.

Do this mental experiment with me. Imagine that we have two characters who can walk left and right, attack, block, and throw, on a continuous space. Attacks beat throws, throws beat blocks, and blocks beat attacks. This is sort of the most basic concept behind most fighting games, and already, we do have a game, albeit not a terribly interesting one. But now imagine how much emergent complexity you get from adding maybe just one or two more moves (maybe a jump, a longer range slower attack, and a fireball).  Assuming these moves have interlocking relationships with the previous three, that’s a huge amount of emergent complexity. So that’s something to consider right away.

But there’s an even better approach, which Smash Bros. seemed to be considering, but didn’t really go all-out on. What if you just had one attack, but it was totally directional?  Meaning, depending on how and when you use it, it had a totally different effect?

My favorite example is Captain Falcon’s aerial-up attack in the original Smash Brothers.  It’s a reverse-flip bicycle kick style move. This is easily the most rich, interesting move in the game, because depending on when you use it, you can knock the enemy almost any direction.  So that’s just one move, but by using continuous time and space smartly, you get a whole range of effects (if you designed it right, it could be 360 different degrees of effects).  They, of course, took out some of this functionality for Melee, and the move is more of just a “I deal damage to you now” move.

Here's an image showing how Captain Falcon's up-aerial attack works

Here’s an image showing how Captain Falcon’s up-aerial attack works

This is just an example of the kinds of emergent complexity that you can get with smart design. I don’t think the original designers of Super Smash Brothers even knew quite what they had, a point which is proven when you look at the directions that they went.

Again, the original game has many flaws, but none of these flaws are because of these good qualities that I mention. They are simply flaws in a 1998 console game that had no way of patching itself, and nothing more – an anecdote. People in the fighting game community tend to believe that things will always be the way that they are now, which is probably the most self-destructive elements of that subculture.


Back to Regular Fighters

To this day, fighting games have been really held back by a desire to “simulate a fight”, when they have the potential to just “be a really great game”. They will have to be conscious of the problems of execution and of mass content to ever achieve that, though.

Asymmetry can be exciting and interesting, but it can also be a cheap tool of simply overwhelming the player with content so that it delays the finding of optimal strategies and also just keeps the “excitement of something new” alive for longer. Content is cheap, though, and all you’re doing by adding 40 characters is two things: ensuring you will never balance this game, and delaying people from discovering how bland your system really is.

There is also no excuse for complicated button inputs to execute a move. If I have a move, and it’s an option, it should be made as easy to input as possible. This is a 100% hard-line RULE of game design, not a guideline  If you need the move to have a delay to it, put the delay in the animation, or have the person have to hold the button down for however long before the move executes. Again, players should never, ever, ever “fail to input” something.  Even if it’s just one button, that’s still going to happen, but as designers, it’s our responsibility to make sure this happens as rarely as possible.

There’s also no need for 3 different punches, let alone all the ducking and jumping variations on those.  I think a Street Fighter character must have upwards of 30 different unique moves, all of which with their own hitboxes, priority, frame-count and damage boxes, which, if you want to play seriously, you simply have to just sit down and drill and study. Again, “drilling and studying” is going to happen sometimes in almost any game, but the last thing you want to do as a designer is arbitrarily increase the complexity to force players to do that. You can achieve more with fewer moves if each move has more in-system meaning. Less is more.

Anyway, thanks for reading. I hope that I can be a part of bringing the world a new generation of fighting games that are smart, elegant, dynamic, easy to learn, difficult to master, and that can stand the test of time.

  • Dasick

    Is there a picture of the Smash Brawl in the toilet, or did you do that to your own copy?

  • Nahil

    The creator of Fancy Pants Adventures (Brad Borne) made a simple smash bros variant using the Fancy Pants engine and gameplay. It’s kind of interesting since there’s only one character, one attack button, and only a few attacks dependent on whether or not you hold the attack button now and whether or not you’re in the air. There’s also a focus on momentum, and the more you build up, the farther the other guy flies and the more his percentage rises. There’s also no double jump or dodge, so there’s a lot of commitment to your positioning. Then you’ve got sliding and rolling into a all to trip up the opponent, and that’s about it. It’s just interesting to see what smash bros can be like when it’s super simplified, and it’s free at Brad’s website.

  • keithburgun

    It’s photoshopped, but it’s likely that there are a number of similar real photos out there.

  • keithburgun

    That sounds interesting. Can you find the link to this game?

  • EvanD

    I disagree that execution of the moves should be approachable for anyone (stated like a true old fogey). I can sympathize with the statement since I see fighting game matches where I know I’ll never be able to execute at that level. However, there is great beauty in seeing a master performing moves with a high physical difficulty, like Van Halen shredding the guitar in “Eruption”. The parallel of fighting games with music is striking: a skilled fighting game player has certain chains of moves that you stick to as your bread and butter but is capable of improvising and responding when the situation changes. A skilled musician has several sets of crowd-pleaser songs and will change up and improv depending on crowd mood. Additionally, a skilled musician can play melodies that other musicians can’t. Having the parameter space so wide gives a huge range of creative possibilities. Limiting the possible moves to a small number of easily executable options is like giving a pianist a keyboard with only 5 notes.

    There is certainly a good point in the strategic view that games should make all the moves easily producible as it makes the game more mental (no accidents). However, any “real-time” game has the time dimension which always makes proper physical execution a requirement. Not everything can be or should be chess.

    PS: I do agree entirely that Smash Bros. Melee is garbage.
    PPS: Love the blog, keep up the good work!

  • Nahil


    Click arena 1 and make sure to make it 2 players. It’s not a finished game, it’s just something he’s testing for his next game. He’s doing a thing where he lets players play stuff he’s testing out on this page.

  • Davidos

    Smash games are not *hard* to master by any stretch. A normal and a special attack button, a shield that blocks attacks, a grab that goes through shields and jumping. From there, the execution of moves is pretty intuitive; pressing left while attacking makes you attack to the left, etc. I would not say that the entire focus of the game had shifted towards racking up %; it’s more like different characters were added with different goals. Captain Falcon and Fox were focused on racking up damage, while DK and Bowser were focused on getting in one or two essential hits to kill. Positioning still mattered in every iteration, as every character had strong angles of attack that become apparent in the first minute of every match, even if you don’t take the time to memorize match-ups. The variety of options that each character has against each other character is part of the game just as much as the core shield and grab mechanics are part of the game.

    Also, short hopping aint hard. Just tap on the jump button instead of holding it down. The % is how much knockback scaling is multiplied by; starting from Melee, every move has a base knockback and a knockback scaling. Falcon punch has an absurd base knockback, while Captain Falcon’s Up-air has near-nonexistent base knockback, but massive knockback scaling. At low %, the bicycle kick sends the opponent nowhere at it can be chained into itself for decent damage. At high %, it behaves like the smash 64 version of the move.

    There are variants of Smash outside of nintendo’s trilogy. Project M aims to make Brawl faster like Melee, while increasing the differences between the playstyles of different characters. Super Smash Flash is a PC game that takes out most of the ATs from Melee and brawl and adds non-nintendo characters like Megaman and Goku.

  • Bandreus

    Completely agree with you EvanD. Accessible controls simply don’t work in typical fighting games (of the like of SF & co.), those impact gameplay in a subtle but very negative way. In fact, most modern fighters include some kind of simplified input scheme (like being able to perform a special attack by pressing a single button), but that’s a feature usually dismissed by everyone but the most inexperienced/unskilled players.

    I can see how fighting games might be considered stagnant these days, but I tend to see that as a consequence of the fact that once very good and solid formula has been built, so it just makes sense for big companies to just stick to it for the most part. In fact the trend is not as bad in the genre as it is in more mainstream ones.

    Other than that, it’s just a matter of distinguishing the good games from the bad ones, like in every genre. Street Fighter III: 3rd Strike is a masterpiece, for instance. Also, MvC 2 is an extremely good fighter. I will also cite KoF ’97, which I regard as the best in that series.

    Fighting games (the good ones) are among the best kind of competitive electronic games, imho. No wonder David Sirlin took concepts drawn from fighting games as the base to his card games.

    P.S. I own Bushido Blade for PSX. Very unique. I especially liked the level design as well: fights unfold in unpredictable and interesting ways when the contenders can roam around huge, complex environments to try and use the surroundings to get themselves in a favorable spot. I think that’s probably the best example of a true fighting game in a 3D environment.

    BB is a simulation at its core, so I think considerations about balance are a bit airy. This gets even more extreme in the sequel, where two of the available characters are equipped with firearms.

  • Paul Spooner

    As a musician, I take exception to your comparisons. If the purpose of fighting games was to create beauty, elicit emotion, or offer a flexible platform for the player to express themselves, then the metaphor would be apt. But fighting games are not all that. They are a game where two players compete for dominance. Anything can aspire to artistic expression through excellence, but that does not make such expression justify game mechanics or input difficulty.

    Even if the comparison between fighting games and musical instruments were apt, the comparison would have to be to that of some theoretical instrument which would produce an entire pre-recorded chorus in response to a complex and difficult to perform sequence of strums. The elegance of instruments (and the skill to play them) lies chiefely in the simplicity of the input, and the limited range of notes, combined (by the player, not the designer) in novel ways.

    The long-echoed refrain of “but I like it!” holds no sway to that point. No one is calling for banning input-contest button press fests. If you like that kind of thing, fine. The point of the article is that fighting games, as games, have much to be improved on. In this respect, music too begs simplicity and player empowerment, not input obfuscation.

  • Bandreus

    Of course they can be improved upon, that can be said of basically every genre. Also, I guess the discussion about fighting games has very little to do with them being art. Personal expression is indeed a very strong part of good fighting games.

    But, as a musician, I found the comparison very fitting.

    First of all, both music and fighting games are indeed based on simplicity. Being able to press a button in order to strike a punch is as simple as pressing a key to have a note played.

    The complexity obviously comes into play when you have to perform many inputs (commands or notes) in rapid succession AND in meaningful ways: Anyone can mash keys on a piano like mad, but what usually comes out is noise rather than music. Similarly, “Button Meshing” can too happen within fighting games. Interestingly enough, well designed fighting games are expected to make Button Mashing (mindlessly pushing buttons in a not meaningful order) a sub-par tactic.

    Moreover, like some fighting games require you more effort to learn/execute all the input commands, some musical instruments are way less accessible than others. Getting acquainted and subsequently able to play the recorder, or the guitar, is substantially simpler than doing the same with, say, the Harp or a drum set.

    Ultimately, both playing fighting games (or, for the matter, other competitive, skill-based games) and playing a musical instrument require significant passion and effort, especially if you want to achieve high skill levels.

    The patterns in the learning process also are surprisingly similar, leaving aside the obvious differences: you first learn the basics (how notes, octaves, scales and tempos work for music, basic commands and differences between playable characters in fighters); then lots of practice quickly becomes the focus of both (for instance, playing famous songs, or mastering bread-and-butter combos and techniques for your main character); Lots of studying and daily practice is required to achieve perfection in both music and fighting games.

    Actually, if you think about it, getting to being very good at playing a fighting game takes WAY less time then doing the same with playing a musical instrument. You can learn all but the most complex input strings of a Street Fighter character in a matter of minutes, maybe a few hours if you never ever played a fighting game before. Put anybody in front of a Piano and maybe (MAYBE) he’ll be able to memorize/execute the whole Jingle Bells theme in the same amount of time. Obviously, the scope and possibility space of music/instruments are far broader then what’s possible in even the most complex fighting games, so that was an expected conclusion.

    And this was only regarding the technical aspects of fighting games. The true struggle with being good at them has more to do with learning to read your opponent’s patterns while mixing up yours, improvising, punishing them when their defense shows an opening, resource management, etc.

    More accessible fighting games would be very welcome, if you ask me. If anything else, they would help more people to get into the genre and understand what they really are about.

    Also, it’s not like a lot of experimentation isn’t going on. Simply put, you need to search in the indie scene to find that. For instance, check BariBariBall (, which is an interesting mix of sports, fighting games and controlling space and your opponents.

  • Padi

    BB has a very high level of inherent complexity. For example, different characters are weak and strong with different weapons, so a player who has not memorized the good combinations can cripple his or her self before even getting past the character and weapon select screens; this isn’t even a balance issue, someone at Square just decided it would be a neat feature, apparently, otherwise you’d think characters would just be limited to weapons they are skilled with (which would be equally justifiable from the simulation standpoint, why would someone ever choose a weapon they are weak with for a fight to the death?).

    But still, I love that darn game. I’ve taught it to countless apprehensive visitors to my home, including people who don’t really like video games, and they end up having a great time. It’s actually much easier to tell people “press forward-forward circle” to do special moves than try to explain what a quarter-circle forward motion is. And, thematically, it just “makes sense-” you chop a guy with a huge sword, blood pours out and he loses, rather than all sorts of Dragonball nonsense occurring on screen, which probably intimidates players with no idea how to input combos or specials.

  • Bandreus

    Yeah, BB has its good deal of charm.

    It’s certainly more accessible than other most known games in the genre as far as input commands are concerned. Obviously that means the combat system is not that much deep, but BB wasn’t trying to compete in the competitive fighting game department to begin with, so I like it.


  • Hamster

    Maybe Divekick would be up your alley?

    It’s a ‘fighting game’ with only 2 buttons as input. A jump button, and an attack button. Because your attack is a dive kick that moves you at an angle, you can get around using only these 2 buttons.

    It gets rids of any execution requirements to play, and boils down to the core essence of any good fighting game, which is reading your opponent’s mind. Or making them think you’re reading their mind, but it’s just pattern recognition and conditioning.

    The game’s been well received so far, and a PS3/Vita version of the game is being developed.

  • Bandreus

    I LOL’d. Thanks for the heads up!

  • Kdansky

    I find your music comparison wrong. There is nothing improvised about complex inputs and combos.

    I completely agree with Keith that difficult inputs are garbage design: They make the game difficult to pick up, and very unsatisfying to play if you cannot spend huge amounts of time in training mode. Go and watch the EVO 2012 finals for Marvel vs Capcom 3. Touching the enemy usually kills him, because you can learn to do three-digit hit combos. If you don’t know them, you can’t win a tournament. As a side-effect, the game has become very spammy, where everyone tries to cover the screen with projectiles, hoping to capitalize off a random hit. All the top tier character are top tier because they can either “convert” (which means: “convert a random hit into a deadly combo”) easily (e.g. Wolverine), spam projectiles well (e.g. Morrigan), are good spammy assists (e.g. Doom), AND deal 100% average enemy life in combos that are easy enough to do under pressure. Characters that fall into more than one of these categories are truly top tier: Vergil is spammy, can convert well and can kill 90% of the cast in a single combo because of his broken damage scaling. Doom is spammy, has high mobility to convert and a really easy combo that kills everyone in two applications, but often only needs one.

    There is very little combo variation because either your combo sticks, or it doesn’t, and then, no amount of improvisation will make it work. And that’s not even the worst. In BlazBlu, you have to press a button to break combos, but you can only do so during specific points where the enemy’s combo isn’t a true combo. So you have to either learn every combo in existence, or spend your game time wildly mashing buttons. Neither of which is interesting.

    I long for a competitive fighting game where you don’t need to spend thousands of hours in training mode, and with a decent online multiplayer. Currently my best choices are Bloodline Champions (zero input complexity, completely disproving your “fighting games need complex inputs”), Puzzle Strike and probably Starcraft 2.

  • Bandreus

    It doesn’t take one (or a few, or a ton) bad game to completely disrupt a genre or a game mechanic.

    As I mentioned in some other comment, I think it’s far more valuable to be able to put the “bad” games apart from the “good” ones. So yeah, I 100% agree with you on MvC3 being flawed for a whole lot of reasons.

    Which doesn’t at all means the whole skill-oriented fighting games genre is to be dumped. Having complex controls and hard to figure out gameplay doesn’t really say much at all about how good a game is per se. The most important question would have to be “Is the game good or not?”. Certainly a lot of shitty fighting games are out there (and a good slew of those feature very simple commands too).

    I can see the issues with having to learn lots of stuff about the mechanics of a game (like in your example, having to learn most or all combos in a game).

    But you say “If you don’t know them, you can’t win a tournament”. So what? You probably can’t achieve the best score (for a song) in guitar hero without passing all the song’s notes to mind, you can’t set world record in pacman without memorizing ghosts behavior, you can’t win GSL without being a micro/macro beast plus studying the Metagame to death. Hell, you can’t even hope to beat your average roguelike before knowing almost everything about the hundreds of monsters and items in the game.

    Very bad point there. Winning any serious competition of any game (including ones like chess) always is about being the best (or pretty darn good) at both understanding the inner mechanics and being able to execute at top levels.

    I still have to witness a game for which you can win a major tournament while still not being an expert of said game. And if such a game, I would hardly regard it as a “good” game because of it.

    Also, Puzzle Quest and SC2 are not fighting games at all. Moreover, SC2 has very major skill requirements on par with the most complex fighters, if not more. Can’t really talk about Bloodline, as I never played it.

    I still regard the SC series among the best ever, but I think that mentioning it was a very poor way to back your arguments about accessibility and lo-execution-barrier to entry.

  • Kdansky

    You underestimate SSBs number of moves.

    You got three basic attacks (side/up/down), three strong attacks, three smashes, three special attacks, four air attacks (back/front differ), four throws, one dashing attack, plus a few variants per character, such as Link’s up-special behaving differently whether in air on on the ground.

    That’s a total of 3+3+3+3+4+4+1 = 21 very different attacks at the very least, more than most 2D fighters.

    Add to that blocking, dodging, and movement (sprinting, running, walking, jumping, ledge-hanging) which is vastly more complex movement than any SF4 where you have a single jump arc, fixed dash distance and no flying.

    As a comparison, take KoF13 with its four buttons. You got two kicks and two punches, each in air, standing and crouching. That’s 4×3 = 12 basic attacks. Then you get about four special moves and three supers, and you’re at 15. Sure, all of the special moves come in two to six variants that have slightly different damage and frame values, but they are all but identical. 3/4 of them are also only used for combos, and never raw (and some, like Kula’s crouching jab have literally zero uses).

    If you can’t make an interesting game with 20 attacks per character, you cannot make one with 50 attacks per character. Why? Go has one kind of move, and it’s the deepest game on the planet, proving once and for all that complexity doesn’t correlate with depth. If you watch Extra Credits on this topic, they even go so far as to state that complexity *limits* depth.

  • Kdansky

    I was very unclear on that.

    I didn’t mean “you can win a tournament without becoming an expert”. I meant “you cannot become competitive (at all!) without learning a bunch of pointless bullshit”.

    SC2’s execution requirements are what irk me most about it, but execution only really matters at a high level. You can easily play in Diamond league with 30 APM, provided you use that click every two seconds wisely. In essence: You can enjoy playing it, you can enjoy improving your play, and you get better at execution while playing it.

    The problem with UMvC3 is that you cannot _play_ online without learning at least the basic 20-hit combos. If you can’t convert your jab into a launcher into a 50% combo (that’s way easier than what the tournament players do), you are so far behind even the lowest of players that you will never win a round. I’m fine with losing to MKP. I’m not fine with losing to a player whom I can outplay ten times, but then get blown up by a single footdive loop.

    In SC2 (and chess) you practise opening execution after you’re good enough to enter tournaments. In MvC3, you must practise combos before you ever play a human. The drilling should not come first.

  • Bandreus

    Your figures for KoF XIII are way underestimated. Just check any faq for the game and you’ll see you’re off by a whole lot.

    Not only you have 1 to 3 command attacks (usually performed by tapping a direction plus a specific button) per character in addition to “normal” standing/crouching/jumping attacks, but you also get throws, knockdown attacks (strong punch + strong kick), dashing, rolling, recovery rolling, short/normal/long jumps, Guard Cancel Rolling, Guard Cancel Attacking, Special moves, Desperation moves, Neo Max.

    So I would say your average KoF XIII has more like 18 to 20 normal attacks (way more if you count all the different jumps and dashing variants as well), to which you add 8 to 15 special attacks (not counting in weak/strong versions).

    I don’t get your comparison with go vs fighting games. That’s oranges to apples really and has very little sense.

    Also, Complexity is too vague a word in this case. If you specifically assume “complexity” = “depth”, then I agree. In fact good fighting games emphasize breadth over depth.

    Guess what, your average Street Fighter HD Remix character has well more than 20 different attacks and that’s probably among the best balanced/playable/enjoyable fighting games ever. The barrier of entry (execution wise) isn’t that high even.

  • Bandreus

    What I dislike is you use your arguments about MvC3 and Blaz Blue to imply something about a whole class of games.

    Not liking a game specifically is perfectly fine with me. Also, if you say “I long for a competitive fighting game where you don’t need to spend thousands of hours in training mode, and with a decent online multiplayer.” that’s fine with me too.

    But the discussion (i.e. the topic of this blog post) is about fighting games in general, and if the degree of complexity in execution is or isn’t something good or required at all in the genre.

    Starcraft isn’t a fighting game, like it isn’t Go and Chess nor Puzzle Strike or what have you. You can possibly achieve plat (or higher) on SC2 ladder by cannon rushing every single time, but that bears no meaning regarding fighting games, nor it implies a good cannon rusher automatically is a good SC player because he stands in a specific league.

    Also, having 20 hit combos in a game doesn’t automatically say anything about the level of execution required to perform them.

    Street Fighter IV require you to hit very specific and short timing frames in order for you to pull out some combos (some Sakura combos need 1/60th of a second precision with buttons pressing to be performed), so the execution barrier is pretty darn high.

    In some other games, much longer combos can be performed while the timing to do so not being nearly as strict, making the execution requirement lower regardless of combo length.

  • Kdansky

    I have to point out that the Sirlin-designed SF2HDR is probably the SF2 version with the smallest fan-base, precisely because it throws out pointless input complexity in favour of depth. The FGC doesn’t really like it.

    KoF13’s move list isn’t as long as you make it out to be. I completely forgot the rolls (but there are just two) and guard cancels, but there are very few different jump variant attacks (some characters have one, most don’t). There are four jumps, but if you use an attack in air, you always end up with the same attack, no matter which jump you used. Four jumps sounds like a lot, but then, in SSB you have air-control during your jump, which is more than “4”. As for 15 special attacks:

    25 moves total, including pointless mentions like Press Plus (a command you can only input after your command grab, and that you never skip because it always does extra damage for no cost). Andy has 27, Kula has 29, Kyo has 28.

    Compared to the 21 that is the lowest boundary for any SSB character, it’s not a huge difference. Many SSB characters have more than 21 attacks: Link can throw his bombs in four directions, can use his chain to grab a ledge, has a rekka fotward smash, up-B is different during jumping, normal attack can be mashed for stabbing and the bow’s arc changes with charging, but I think the basic 21 gives us a good number for comparison.

    I conclude that SSBM has fewer moves than KoF13, but not significantly. They are in the same ballpark of 20-30. SF4 might be above that, BlazBlu below. In the end, SSB has more than enough.

    My point about Go was: “20 moves is enough to make a good game. Adding another 10 only adds complexity, but no depth.” and my argument is: “Go has one kind of move, and it’s deep as hell. This proves that the number of possible moves in a vacuum is not a significant factor for depth.” The minimum number of moves a good game needs is ONE.

    Complexity = BAD.
    Depth = GOOD.
    Why =

  • Kdansky

    I’m basing my discussion of fighting games on MvC3, SF4, and [insert any ArcSys game, they have identical mechanics] because those are three big games right now on the scene. Using the EVO line-up for examples is the best we have. Sure, I can point to GGXXAC+ or Skullgirls instead, but that’s just obscure to most current gamers and both games are completely in line with my arguments.

    SF4 has by far the shortest combos in all current games, but to make up for that, it has also the worst kind of execution bullshit: 1-frame Sakura links are the worst of the worst. To quote: “If you can’t do that link, play another character!” I didn’t bring that up because I thought it to be even worse than MvC3. SF4 also has the worst inputs. 720s? Really? Another unbelievable example is Ryu: Fireball and Shoryuken have nearly the same input. WHY?! Why isn’t fireball a qcBack, and Shoryken a qcForward instead? It would be so much easier to do!

    KoF13 is also very guilty of having overlapping commands, like Andy’s Zan’eiken and Shouryuudan, and inane double half-circle inputs for NeoMaxes. ArcSys generally dodges that bullet, except for single half-circles everywhere. BB pretends to be lenient with combo timing because of the 5 frame buffer, which is a half-baked solution to a problem we shouldn’t even have.

    SSB would have low execution barrier if not for the wave-dashing bugs, bullshit infinite throws, bullshit tripping and garbage balance (Meta Knight whoooo! Only 10% of the cast worth playing!). I can’t name any other fighting game without stupid controls.

  • Kdansky

    On a different topic: Why is Brawl so much worse than Melee? I have played both and didn’t find them to be much different apart from a few details:

    + Brawl has better stages, mostly because it has more stages.
    – Brawl has tripping. I give you that. Tripping is the worst feature since Pay2Win.
    + Brawl has a lot of playable characters, if you ban MK (and possibly Snake)
    – Melee only has Fox, Falco, Marth and Sheik as playable characters.
    + Brawl looks better.
    + Brawl has less wave-dashing.
    – Both could do with less “hang off the edge” for the invincibility frames.
    – You need to deactivate half the items.

    What’s so much worse in Brawl than Melee?

  • Bandreus

    You might want to research a bit on the why-s command inputs are designed the way they are, and the difference between good overlapping commands and bad overlapping commands.

    David Sirlin’s articles at his blog host a huge slew of information about that.

    Also, you need to consider how the input commands for a character’s move impacts on the gameplay implications of using that move.

    In your example, for instance, a qcb firebal or dragon punch would make those moves very powerful. Generally speaking (for games in which the back-command doubles for defence), string which include only neutral and backwards commands (like in the qcb and reverse dragon punch commands), tend to be very safe (if the opponent hits you during the short window needed to input the commands) you would automatically defend.

    Both the adouken and the shoryuken are powerful moves used to control space, aggressively (hadouken) and defensively (shoryuken), hence the qcf and forward dp commands. Furthermore, if the player links those in a combo which didn’t connect, his defense is wide open (because of the forward motions), offering possibilities for punishment.

    As a counter-example, if you take Q from 3s, all of his commands are of the qcb, hcb or hold back types, which makes sense: he is an extremely low (although mighty) character, hence a player using him is pretty much assured to be almost constantly under the opponent’s pressure. Q’s special moves can easily double as defensive/reaction type attacks. Also, Q’s combo-ing potential is very low (only a few attacks can be linked together), in order to keep the defensive advantage of his command-moves in check.

    Lastly, regarding 360° and 720° inputs specifically, you need to keep in mind an arcade stick is the preferred input-device for fighting games. While performing those commands on a game-pad is nearly impossible, it’s much easier (with practice) to get those right on a stick.

    360° and 720° have design reasons behind those as well. In fact they are exclusively used for powerful grab moves (normal, and special versions, respectively). Those kind of moves tend to be unblockable (nor tech-able) and deal a lot of damage (a 720° special can easily hit for half health bar), so those are balanced out by requiring more complex-to-perform commands.

    A double qcf Ultimate Atomic Buster Zangief would be an extremely broken Zangief.

  • keithburgun

    Yep. So this is obviously a little on the “too simple” side, but it’s interesting to notice that even with this 2 move setup, there is SOME emergent depth/strategy. Imagine increasing the number of moves to, say, 5. It’s not a linear increase in emergent complexity, it’s exponential.

    Now remember that Street Fighter 2 has freaking 30 moves, sort of arbitrarily.

  • keithburgun

    Honestly, and this is something I tend to keep to myself for the most part because I know it will start shit, but I think that melee and brawl are much more similar to each other than 64 and melee are. Melee is where the biggest changes happened.

    So, in short, you’re right. Brawl is actually only a little bit worse than Melee.

  • Kdansky

    Keith, did you delete my answer to this, or did I fuck up submitting?

    What I was trying to say in too many words:

    29 moves for kyo, not much more than SSBM.

  • keithburgun

    Sorry, anytime you include a link it asks me to approve it which I’ve just done!

  • Bandreus

    I didn’t expect being served ‘game design for dummies’. I found it funny and bookmarked it :)

  • Bandreus

    The animation for a jump attack might be the same, but there’s a meaningful difference when all jumps cover different amounts of space in different amounts of time. That’s especially important when you’re talking games in which space control is a major aspect, like fighting games are.

    (The move set for a single character in KoF might seem small, but keep in mind the player picks teams of 3s).

    Similarly, the 3 different variations for specials in the street fighter series are all but pointless (well, that’s clear to a moderately experienced player at the very least).

    But yeah, I don’t want to be nitpicking as well.

    We absolutely agree on complexity for complexity’s sake being a bad thing. Depth certainly is more important, but even more important is breadth.

    The less breadth (number of different, meaningful options) you have, the less important the guessing and space controlling games are.

    That being said, I’m still very interested in any kind of new approach and experimentation in this like in any other genre.

    I was just trying to point out you can’t really call something an ‘ambiguity sauce’ when it does serve a meaningful design goal.

    Other than that, it’s just a matter of tastes and being able to discern the well designed games from the bad designed ones. And that only is my humble opinion.

  • Kdansky

    720°: Considering the fact that Pro players can even execute complex moves in a single frame to begin with, I don’t see how that is a balance concern. This only works at low levels of play, but not at the very top. I mean, look at Morrigan’s Fly/Flycancel/Fireball/Fly shenanigans. That requires ridiculous input speed, and Chris G. totally dominates with it.

    Since we’re all interested in how the game plays at the very top, it makes no sense to have functionality purely designed to make it hard on beginners. Sirlin himself says this is the case (though I think I’ve seen a more fitting quote, but can’t find it):

    I totally agree that “forward” and “backward” are relevant parts of a move because they influence blocking. But considering that there are a multitude of characters with fireballs on qcb (such as KoF Andy), and characters with reversal-moves without a forward in them (such as EVERY Persona Arena character where you get a special button for it), and games where blocking is a button (Smash, MK, SC, Tekken) and other difficult moves that have no relevant input (1-frame Sakura links) this is *much* less relevant than you think. Yes, the *current* meta-game of *some* characters in *some* games is *somewhat* based around that. But if you remove this “feature”, the meta-game will quickly shift to something else that’s just as important. Case in point: There are many character which do not have this issue, and they work just fine. KoF-Andy has a forward-Reversal and a back-Fireball!

    And even if you disagree with all that: Instead of qcf you could just press forward + button for the exact same outcome: Inability to block during reversal.

    In fact, there are bad things happening right now due to the command’s difficulty: Players at tournaments pay attention to their enemy’s stick movement (mostly by sound) so they can tell when he’s buffering a super. I find that a really bad development.

  • Bandreus

    No need to be too nitpicky here.

    “Since we’re all interested in how the game plays at the very top”: actually… kinda.

    The game mechanics and commands need to be considered at all levels of play. As I believe we can agree on, the more complex the commands, the higher the learning curve.

    But that said, I believe putting too much emphasis on the absolute top pro players level of play would be pointless (to the end of this discussion). If all you cared about was the top level of play, then complexity of input commands would be irrelevant (since pros can pull it off anyways).

    When you look at less experienced players, I honestly think the commands in MOST (please read as: “not every single fighting game in history”) fighting games are pretty much okey.

    Yes, if you are a complete newb, even a qcf (or simpler commands, for the matter) could be a problem. But be honest here, you’re talking like learning to perform a qcf (or any other command) took friggin’ years of training to be pulled off, which isn’t the case.

    When you instead talk about the most complex commands (1 frame links, creazy combo tricks, kara-ing in 3rd strike, etc. etc.), those indeed mainly pertain to higher levels of play. I don’t expect a newbie to be able to perform those input commands after an afternoon of training, nor should you pretend it to be the case or blame it on the designers.

    Which leads us to the key point here. We have been talking mainly about fighting games which have been designed with Tournament-level play in mind.

    These games feature:
    1) input-commands which are simple enough for newer players to execute those with relative ease, while still not being trivial (otherwise Button Mashing would become a serious issue).
    2) More advanced and complex commands (these usually are mainly for the most advanced combos/techniques). This might seem arbitrary, but it’s really not: the designers DID WANT the pros to not be able to pull out the top-notch combos 100% of the times, trivially.

    I don’t understand the rumble about different game series using different command inputs for similar things. Like, what’s the matter with Andy having a reverse fireball command vs Ryu having a forward command vs some games having one-button fireballs? Different games -> Different designers -> Different design goals -> different everything.

    Again, pick a game and first discern if it’s a well designed one or not. Then (and only after that) feel free to discuss how the commands differ from other games in the series, and we could have a very interesting discussion about that. But using using those differences as an argument to support the idea that ALL fighting games should use simplified and dumbed-down inputs is pointless.

    Remember that (I’ll paraphrase Keith’s and many other recognized designers’ words here) a well designed game needs to be as much complex as needed, but no less. I believe that that applies to the control interface as well.

    I will reiterate my opinion again: You can certainly design good and fun fighting games which feature simpler commands, that’s great. But that doesn’t at all mean some kinds of fighting games would benefit from having simpler commands.

    Also, forward + button for a fireball would be so terrible. The qcf (or qcb, or whatever) also serves the purpose of making the player throw a fireble by mistake less more likely.

    With forward+button, you would execute a fireball instead of a normal punch every other time (requiring the player to have the stick in neutral position in order to perform a normal attack, in a 2d fighters, would be very bad design).

    Please take note that some fighting games (can’t remember which ones exactly) do use the forward+punch/kick as the input-command for Strong punch/kick. Now, I personally don’t like that much, but that’s far less of a problem than using that for a fireball attack.

    Take, again, SF series. Forward + Fierce/Roundhouse (when close) was the command for throws, but they opted to change the input command into Fierce+Roundhouse (when close) instead.

    Having input commands which make you more likely to pull a simple move instead of another one by mistake hopefully is a thing of the past.

  • keithburgun

    >Also, forward + button for a fireball would be so terrible.

    No no, this is crazy talk. It would be fine, just make specials all use their own button that are only used for specials, just the way Smash Brothers does it, then it’s not a problem at all.

  • Bandreus

    “It would be fine, just make specials all use their own button that are only used for specials”.

    Yes but, I see problems with that for many different variations on the fighting game genre. It works fine with Smash because the whole game is entirely designed with that kind of controls in mind.

    But for different games? First, it would bring the total number of buttons to 5 (KoF) or 7 (SF).

    Your typical fighting game character usually comes with a minimum of 3 special moves (often times more than that). You could ofc allow for different combinations using the Specials button.

    Let’s take Ryu as an example. You could have each move command as Stick Direction + Special Button. Something like:
    Forward + Special = Hadoken
    Neutral + Special = Shoryu
    Backward + Special = Tatsu

    Now, each move comes in 3 variants, how would you make that work? You can ofc have 3 times 3 special moves with simple input commands in such a system (8 stick directions + neutral, although diagonal directions are very hard to comfortably get right when outside a qcf or similar stick motion).

    This is ruling out characters with more than 3 special moves and cutting out super/ultra moves and the special bar entirely.

    The game would be more simple and maybe still fun, but far less entertaining (the more powerful super/ultra moves allow for more strategizing and exciting come-back scenarios).

    Also, some ports of SF on consoles do feature a simplified input option (you press a direction on the pad’s 2nd cross-stick and a special magically comes out): any ideas about why that mode is rarely used, if ever? (hint: is because it doesn’t allow to perform the whole gamut of moves).

  • Kdansky

    It’s really not necessary to have three variants of every move. Sure, extra buttons are a problem when you want 6 basic attacks, and then again three powered up versions of every attack. But we don’t need 50+ attacks to begin with.

    Example with 4 buttons: Three basic attacks, and a special button would do fine. You can then press Special + X for a graduated effect: Forward/Special/Weak gives you a weak fireball, and so on. Forward/Special/two buttons gives you a Super. That gives us 4 specials (or variants) per direction. With just four cardinal directions (+neutral), you have 20 special moves while not in the air (and another 20 during jumps). Add to that the 3 basic attacks for each low/mid/jump, and you’re already above what KoF13 offers. And of course, you can use 9 directions instead of 5, add two-button basic attacks and command normals. Do you really think you need to have more than 100 attacks per character? Didn’t we establish that 20 is more than enough?

    There are “easy modes”, or rather “shitty modes” in some games, which are shitty on purpose. That’s a huge strawman! Of course nobody wants to use a mode where you get gimped by design!

  • Bandreus

    “It’s really not necessary to have three variants of every move. ”

    Why? That’s an arbitrary statement. You can obviously say “but game x only has 2 variations per move” or “but game y doesn’t have any variation on moves whatsoever”.

    It’s fine if a game is specifically designed around that, but there’s good in having multiple variations per move (if you design around that!) SF, again, is the perfect example. Every decent SF player would tell you all variations of a move have some meaningful use, that’s true for at least 70% of the game’s overall moveset.

    You say “we don’t need that many moves” just for the sake of invoking simplicity, but you’ve not been backing this by any kind of game-mechanics point of vie. You acknowledged You just say “Simpler games would be better because they would be simpler” which I find funny.

    I mean, if your intent is to design a simple, easily approachable fighting game, well that’s great, it’s awesome. Really! Go along and design that game, have 2 buttons for attacks (punck/kick) and a button for specials, that’s fine with me.

    But you would cut the play-space by many orders of magnitude. What about SF?

    Let’s suppose, just for discussion’s sake, SF has the right amount of normal attacks, specials, moves variations etc.

    Actually, nope. Let’s assume you don’t need any variations on moves whatsoever for the sake of simplicity.

    Also, mechanics-wise and assuming (assuming!) you could have every move in the game as easily to executable as tapping a button, would you be okay with. Let’s make a SSF4 example, shall we? I’m talking SSF4 just because it’s broadly accepted as the most played competitive 2d fighter atm.

    (from eventhub’s SSF4AE guides)

    Zangief’s Piledriver:
    – “The ultimate command grab. If you’re close to the other player, this is a huge threat as the range is terrific and the damage is awesome. This move is one of the biggest reasons Zangief is such a great character in this game.”
    – command: press specials button
    – range: close (still noticeably greater than normal throw range)
    – damage: 230

    Ryu’s Hadoken:
    “Ryu’s fireball is his iconic move. It has excellent recovery while allowing Ryu to chip opponents and build meter safely. They are also a safe combo ender on hit or block.

    How well you can play Ryu comes entirely down to how you space your fireballs and how well you know your opponent’s anti-fireball tendencies.”
    – command: press special button
    – range: max (multiple screens)
    – damage: 70

    Do you feel having these two moves executable with equal ease, both in terms of simplicity and time needed to input the command is fair? (ie, a button press/direction+button both are = 1 frame, leaving input lag aside).

    Also, since the Piledriver basically is a more powerful, longer range throw (which your opponent can’t counter!), why would a Zangief player ever use a normal throw when both commands are as easily executable? Don’t you think you might as well just replace Zangief’s normal throw with the Piledriver? Would you find that inbalanced? (keep in mind throws are paragon to SF gameplay).

    Also, going to the more nitty-gritty, less obvious but still very important stuff. One of the effects of slightly longer input strings as command inputs is that you basically “show” your intentions (even if for a brief moment) while performing those. This is less important when linking into combos, but alas.

    For instance, a player can “tell” (or, more precisely, try to guess) his opponent is going to perform an hadoken when he goes from neutral stance to crouch stance to neutral stance again. Take note that the shoryuken (the other threatening ryu’s move) has a different tell! (forward stance, followed by crouch stance, but no following neutral stance).

    The important things are:
    – The opponent has a few frames to react.
    – This system creates very important chances for the guessing game to develop itself.

    The latter is the most important, imho:
    – A Ryu player might “bait” me with a fake shoryuken motion.
    – I see the threat and block high (e sees me stand).
    – He performs a low attack, I’m hit.

    You might think “well, by blocking low I defend both the shoryuken AND low attacks”
    – Ryu baits me with fake shoryu motion.
    – I block low (he sees me crouch).
    – Ryu performs Collarbone Breaker (insta overhead attack with a brief delay), I’m hit.

    Now, SF is a fast paced game, you need to take decisions in the blink of an eye. But the very brief time needed to input command moves has a whole gamut of effects on how gameplay develops. Take note that even the most skilled SF player can’t perform a qcf in a 60th of a second, while the less skilled player can go pretty darn close when pressing a single button (or direction + button) cause the motion is nearly instant.

    If you implement simple inputs (all specials with a button press or direction + button press), you remove all of that. There’s no more, or greatly reduced chance for baiting. Because all moves come out instantly:
    – you wouldn’t see Ryu crouching before a hado
    – you wouldn’t see Ryu advancing before a shoryu
    – The opponent much less or no time to react what’s coming next
    – No faking, you can’t “begin the motion” without completing it, either you perform the move or not.
    – No baiting (all moves come out instantly, you can’t fake input commands), much less interesting guessing game.

    You can still analyze the patterns and try to guess the opponent’s next move, but w/o faking and baiting, you can’t make educated guesses, it’s just a matter of “I was lucky or I reacted quickly enough and blocked, or didn’t” but not “I saw the move coming, but I realized it was a fake, I managed to defend his bait when I saw the overhead coming”.

    It might be subtle, it might not seem that big of a deal, but SF is built around that. Only the most causal players don’t realize that. Performing a qcf is a piece of cake (you learn the motion in under 5 minutes), but the consequences of having to input that motion are far greater than the actual motion itself. The gameplay is more varied, the developments of the fight far more interesting.

    SF: Simplified Edition guessing game would be dangerously close to that of Rock Paper Scissors. I don’t remember the last time I saw a RPS tournament.

  • keithburgun

    The problem that’s being had here is that you – bandreus – are thinking of the Street Fighter system. This system itself may indeed REQUIRE 30 or whatever unique moves, because the system itself isn’t that inherently interesting. What I am suggesting is that you have a more inherently interesting system with more emergent complexity, that would need less inherent complexity.

  • Bandreus

    If wishes were horses, then beggars would ride.

    Man, did you even read the discussion between me Kdansky? It was specifically about SF-like fighting games.

    In your article, you suggest we should throw out of the window 25+ years of game design history, but yet you have very few suggestions about how such an amazing system should be built.

    You wrote: “Imagine that we have two characters who can walk left and right, attack, block, and throw, on a continuous space. Attacks beat throws, throws beat blocks, and blocks beat attacks.”. Sounds about what SF is based upon to me. Luckily enough you also mention such a simple game wouldn’t be very interesting.

    You also wrote: “But now imagine how much emergent complexity you get from adding maybe just one or two more moves (maybe a jump, a longer range slower attack, and a fireball). […] Assuming these moves have interlocking relationships with the previous three, that’s a huge amount of emergent complexity.” Mmmmh, sounds terribly close to a game I named once or twice…

    Now imagine how much more emergent complexity you could have by adding 30 moves…

    Right, 30 moves are clearly too much, right? A fighting game should never ever have that many moves, otherwise it would mean it REQUIRES that many moves.

    Sure, you also wrote: “But there’s an even better approach […] What if you just had one attack, but it was totally directional? […] So that’s just one move, but by using continuous time and space smartly, you get a whole range of effects (if you designed it right, it could be 360 different degrees of effects).”

    Keith, is THAT your definition of emergent complexity?

    So, let’s summarize. The fighting game of your dreams would need to have:
    – Simple, directional attacks.
    – Blocking.
    – Throwing.
    – Jumping.
    – Maybe a fireball.
    – Ability to attack and move in continuous space.

    Keith, the game of your dreams already exists, although it works slightly differently in some interesting ways. It’s called Soul Calibur.

    In some ports of the SC series, you can use the analog stick to move freely, in 360° around the arena.

    While attacking in 360° in SC is technically possible, the game still auto-targets the enemy for you. Do you know why? Because, in a fighting game, the only direction you want to strike your blows is the one where your enemy is.

    Making you able to strike in 360°, while “inherently interesting” (is it really?), also makes it increasingly more difficult to just tap the right direction on the analog stick.

    Have you ever played Power Stone? Man, trying to hit a very close enemy can be so frustrating, when you are in the midst of a crazy fight. The controls get so frustrating. I dare say picking the right direction to hit the opponent with 360° freedom is as hard as pulling out a qcf, unless you make hitboxes disproportionately huge.

    Obviously Soul Calibur (just like SF or any other of the fighting games you so much dislike) is ALL about positioning and guessing. You also have ring out in there, something you seem to like.

    But yes, here we go: the number of moves and complexity in execution.

    So obviously Soul Calibur REQUIRES lots of moves… because it requires lots of moves I guess.

    No Keith, it has lots of moves because otherwise the game would be too simple, less interesting.

    Keith, I just think, by growing up, you started to dislike certain kind of games. It’s fine really. But please, stop dismissing people who are trying to have interesting, meaningful discussions, just because they don’t share your same opinion.

  • keithburgun

    >>So, let’s summarize. The fighting game of your dreams would need to have:

    You’re missing my larger point which is that the basic kernel of gameplay in Street Fighter, and Soul Calibur and almost any other fighting game is inherently UNINTERESTING. It’s inherently un-dynamic, it’s DIRECT. It’s “grabbing victory points out of the box” gameplay. Hit points are at the core of most fighting games, and that’s what we need to get past.

    I’m advocating for something fundamentally new. My little explanation about “take basic RPS then add a few moves” was only to illustrate the complexity issues of Street Fighter, NOT to suggest that reducing the number of moves would be a solution.

  • Bandreus

    Okay, sure, fair enough. So, since in your words “I’m missing the point”, what would a different system means?

    It’s never said in your article, so either you don’t have a clue yourself or you enjoy teasing us I suppose.

    You write at large about what needs not to be in there (basically everything which made fighting games popular in the last 25 year).

    Since you enjoyed Smash so much, should Keith’s Fighting Game only be about ring out?

    Keep in mind, in the Soul Calibur example above, you could set infinite health and basically the only way to win a match would be via ring out. In which case, the fact remains Keith’s Fighting Game happens to be incredibly similar to Soul Calibur (let aside it has that many moves). Would it be fine in your opinion? Would it be inherently interesting enough for your standards?

    Or maybe, you are talking about something completely different and extreme. Did you check the link I posted about BaraBariBall? Are you thinking to something about that? Where the goal of the game is something else than knocking out the opponent, while “fighting” still is a tool for the player to use?

    Keep in mind, I’m not trying to troll you in any way. It’s just, your article is so vague, if you strip out all the useless wording, the bashing on the genre and collateral information, it could be summarized without fear of losing information to:

    “A fighting game in which the goal is not to deplete the opponent’s HP, with emphasis on positioning, free-form movement and attacks in a 3d arena and which featured RPS-like relations between player’s actions would be inherently interesting and would allow for more emergent complexity.”

    Which, I mean, as an opinion is completely acceptable. A positive discussion could sprang and everyone would happily share their opinions about that.

    But nope, you hide your points within lots of noise. It’s like you’re enjoy throwing shit on what in your opinion is “bad” design more than discussing what in your opinion would be “good, if different” design. Keep in mind, people find it irritating when someone throws shit on stuff they enjoy, especially when mostly conjectures and personal taste is being used as the basis of your opinion, so criticism and counter-arguments should be expected! [/psychology 101]

    Now, if you don’t start nitpicking my whole post to point out the “wrong” bits while dismissing the whole of the rest, my PERSONAL opinion (which is not aimed destroy yours) is that the current fighting games paradigm has nothing inherently wrong with it. I see complexity as something positive, within the overall context (easy to learn, hard to master). I think games having a moderately steep learning curve is not a bad thing 100% of the times. I see lots of emergent complexity being a possibility even within such systems.

    I would also like to point out that, in my opinion, the vast majority of fighting being released have brought little to no advancement in the genre, let alone of the interesting kind. If you ask me, the absolute pinnacle of the genre would be SFIII: 3rd Strike, and that’s mainly because of the Parry System (a mechanic which made SFIII even more about great reflexes and crazy execution skills, sure, but it also made the system so friggin better and allowed for VASTLY more emergent gameplay).

    That’s my opinion only! Don’t feel the urge to point out how I’m missing the point entirely, nor to dismiss my opinion with a two-liner and little/no explaining to dispute it.

    That being said, I’m still genuinely curious about your own opinions and design philosophy. If at all possible, I would like you to post more focused articles about it. Please, try and touch on the mechanical aspects of the systems you’re thinking about, try to expose your game play ideas in a clearer, less vague fashion. In a few words, give us something useful.

    Your truly.

  • keithburgun

    >>>Okay, sure, fair enough. So, since in your words “I’m missing the point”, what would a different system means?

    I mean you’re basically saying “design a new game for me in this comment.” There are a couple of things I think people don’t realize.

    – The wealth of possibilities for game design is FAR, FAR wider than people expect. They’ve played SF2 variants for so long that it’s difficult for them to imagine something breaking that mold and doing something totally different. Also, the focus on fantasy simulation for so long has had a similar effect. When you’re allowed to be a little more abstract, the possibilities are huge.

    – Designing a game is actually really, really hard to do. It’s so hard, in fact, that few developers even TRY. It’s not this thing where you can just sit down and “come up with a game design” right away. It’s more like writing a screenplay. You might start working on it, and then it just completely fails and falls apart in your hands. It might take years.

    So, with these in mind, asking me to come up with “what a different system would be” isn’t fair.

    >>>>Since you enjoyed Smash so much, should Keith’s Fighting Game only be about ring out?

    Again, you’re thinking too narrowly. Ring out is superior, I think, to health bars because it’s more tied to the gameplay. But by no means do I think all fighting games should be about ring-outs. I think there are tons of ideas that are possible.

    >>>>Keith, I just think, by growing up, you started to dislike certain kind of games. It’s fine really.

    Not true at all. Still LOVE Super Smash Brothers 64, and I play SFA3 despite the two games flaws.

    Mass complexity is life-support for an inherently flat system.

  • Kdansky

    This column is so narrow, it gets hard to read. Can’t you change the CSS so the right lines are closer together, Keith?

    That said, I will just answer a few things you brought up, and not write another giant essay.

    >”Too many moves for easy commands”
    Yes, if you want sixty moves per character, you need complex moves. That’s exactly the problem when your depth relies on complexity. I’ve shown that you can distill KoF down to a simpler control scheme, and be extension, also GGXX, BlazBlu, Marvel and nearly every other fighter that is not 6-button like SF4. I can live without 6-button fighters.

    > Every decent SF player would tell you all variations of a move have some meaningful use, that’s true for at least 70% of the game’s overall moveset.
    Great, that’s 30% of the moves we can cut. Amusingly, that is nearly exactly two out of six buttons (33%) that are superfluous, so there is really no problem any more. “Perfection is achieved when nothing can be taken away.”
    Example: Kula’s crouching A in KoF. There is literally no reason to ever use it, because it’s not only really bad compared to every other crouching A or B, but also her B is one of the best crouching attacks in the whole game. I guarantee we can remove 33% of all SF4 moves without problems. A slow and a fast fireball are great, you don’t need a medium fireball. If anything, that only dilutes your choice.

    >Do you feel having these two moves [Piledriver vs Hadouken] executable with equal ease, both in terms of simplicity and time needed to input the command is fair?
    Yes, I think Piledriver should be the normal throw. This is commonly called “asymmetry” and leads to great games! In your great analysis of PD vs Hadouken you forgot the small difference of Hadouken having a giant range. Would you rather play a Ryu without any Hadouken variants, or a Zangief without any command throws? Ryu is nothing without his fireballs!

  • Keith Burgun

    I updated the comments system. Might have ruined this thread but let me know if it’s better in general.

  • coglet