Fighting Games

Frame Data Patterns that ALL Game Designers Should Know

Celia Wagar2 Comments

Terms:
Startup, Active, Recovery, Hitstun
Frame Advantage, Frame Disadvantage, Combo, Unsafe
Frame Trap
Pushback

Frame: Also called “tick”. The smallest unit of discrete time in a video game.

Startup: The period of time in an attack animation up to the first hitbox (collider that hits other objects) appearing.
Active: The period of time in an attack animation across which a hitbox is present and capable of hitting other objects.
Recovery: The period of time in an attack animation after the hitbox has disappeared, before the character returns to a state where new actions can be input, or they return to an idle state.
Hitstun: An animation that plays when a character is damaged by a hitbox, during which the character is incapable of acting until the animation has completed.
Blockstun: Like hitstun, but for blocking. The stunned animation played by a blocking character when they are hit by an attack. Usually shorter than hitstun in duration.

Advantage/Disadvantage: The difference in time between when an attacker recovers from their move, and a defender recovers from hitstun or blockstun. Establishes who gets to act first, like a head-start in a race. When a player is at advantage, it is their “turn”.
Frame Advantage: When the attacker gets to act before the defender does, allowing them to perform a move with a head-start over the defender. Also called, “plus”.
Frame Disadvantage: When the defender gets to act before the attacker does. The attacker has handed over their “turn”. Also called “minus”.
Combo: When a move has so much frame advantage on hit, that another move can hit the defender before hitstun ends. A sequence of moves that each hit the opponent while they are still stuck in hitstun.
Block String: A sequence of moves that each leave the attacker at frame advantage (plus) on block. If there is no gap, it is a “true” block string.
Unsafe: When a move has so much frame disadvantage that the defender can hit the attacker before their move recovers, punishing them.

Frame Trap: When a move leaves an attacker at frame advantage, allowing them to launch another attack that will catch any action the defender makes, besides blocking. If the defender tries to attack when the attacker is at advantage, they will get frame trapped by the attacker’s next move.
Pushback: Each attack pushes the defender backwards. Attacks tend to obey a rule where longer ranged attacks have less frame advantage, and longer startup, so frame traps get looser as the defender is pushed further back.

This Dustloop page is a helpful companion to this article.

Every attack has 3 phases, startup, active, and recovery. Startup is time before the attack has a hitbox out. Active is the time that the hitbox is out. Recovery is the time after the hitbox is out before you return to idle.

Attacks inflict a stunned animation on those that they hit, either hitstun or blockstun. This stun animation will interrupt whatever animation is currently playing, even if that animation is stun. In most games, the hitstun and blockstun animations are different durations, with hitstun being longer than blockstun.

If the hitstun animation is longer than the recovery frames of the move, then the attacker will be able to act first, before the enemy who was stunned. Frame Advantage means it’s your turn, and the amount is how much of a head start you get.

Continue reading →

Emergent Gameplay in Fighting Games

Celia Wagar5 Comments

A bunch of different videos have popped up lately in fighting game circles, about “Emergent Gameplay”. I’ve watched a number of them and they’re grasping at concepts they can’t totally describe. They use a lot of vague terminology and almost say what they want to, but not quite. The gist is, old games had Emergent Gameplay, new games don’t, but why?

Continue reading →

How Fighting Games can Retain Players

Celia Wagar4 Comments

Fighting Games don’t attract a lot of new blood. The majority of people who buy fighting games will never attend a single tournament for that game, never post about it online, and never interact with the community in any way. This means that the success of a competitive scene is tangential to the success of the game overall.

Fighting games being unpopular compared to other genres of competitive game is a demographic problem. Demographic problems require systemic solutions, not individual solutions. It’s easy to say that the problem is players who don’t have the drive to improve themselves, but a lot of people just bounce off of fighting games, even though they really want to go on an improvement journey. Fighting games can do a lot structure themselves to get these new players on the right track to understanding the game.

Magic the Gathering went through a similar issue early in their game’s lifespan. At first, they catered to pro players, and the game started to slowly die as MTG became more complex and more irrelevant to beginners. MTG ended up revitalizing themselves by decreasing their investment in competitive play, and instead focusing on creating appealing characters that people could connect to, and an easy introductory experience for beginners.

Wizards of the Coast called the non-competitive kitchen table players, “The Invisibles” because they don’t participate in the broader community, yet they make up the majority of the consumers. This is the case for every game or media product. The majority of a game’s fans will never ever participate in an observable way, but they’re the people who are the backbone of your sales, and they’re the people who filter into the competitive scene. Therefore, having an approachable game creates a more healthy and active scene for everyone.

Continue reading →

SFV Proposed fixes

Celia Wagar

SFV is kind of a mess, kind of controversial. I’ve done my fair share of defending it in the early days, but I eventually quit the game because I wasn’t really happy with it or the direction it was going in. I’d say that the core issue with the game is forced commitment. People had a lot of issues with SFIV being non-committal, things like invincible backdashes, crouch tech, safe sweeps, uppercut FADC, too many features like this tend to make the game about forcing the opponent to endure your shit and hoping they eventually mess up. On the other hand, having high commitment to everything makes the game more like rock paper scissors, as things more cleanly win or lose versus one another, and you don’t have ways to hedge your bets to get a draw.

I admit I’m not an expert player, and since I quit a while back, I don’t really know what’s up with the current round of patches or more character specific issues. If I were to do a balance patch, I’d exclusively buff lower tier characters. I don’t think the top tier have any degenerate tactics that need nerfing, they’re good all-around and don’t violate the spirit of the game, and it’s been this way for the majority of the game’s life. I’d probably revert each character to the point they were at their best, or slightly below that point if their best was truly broken (Mika, Abigail), but largely buff lower tier characters.

  • Increase pushback on sweeps (making sweeps unsafe was a good move from SFIV, but making them unsafe when spaced is too far)
  • Increase pushback and decrease startup/recovery of fireballs to be closer to SFIV standards (fireballs being unsafe on block as well as being very high commit is a large part of why they’re bad)
  • Revert command grab recoveries to before season 2.5 (having them vulnerable to neutral jump is reasonable, having them vulnerable to many jump back punishes is crazy)
  • Increase distance moved backwards on air reset, to prevent jab anti-air from being a dash-under mixup, or make a special exception for just this move or something.
  • Make all reversal uppercut moves invincible without spending meter, or at least invincible up to the first active frame.
  • Remove counterhit property from backdash, make it airborne from frame 1 (lets you use backdash to evade throws and minimize melee damage without avoiding it completely, giving you another wakeup option vs throws without letting you ignore meaty pressure like invincible backdashes) Alternatively, just remove the counterhit property, but make it throw invincible at the start for like 6 frames, so you can get a combo, but not like a crush combo.
  • Increase the number of normals with enough frame advantage to link into other normals (linking routes on most characters are rather boring due to this). Avoid doing this for light attacks, unless they cannot be self-chained or linked into from other light attacks.
  • Decrease the hurtbox size under jumping attacks so they’re closer in dimensions to the hitbox size (weakens anti-air jabs significantly, makes AA in general slightly less guaranteed)
  • Increase the duration that hurtboxes stay out after a move whiffs, animate to generally match the limb as it retracts, rather than jerkily appearing/disappearing (make whiff punishing more viable)
  • Increase the range of the hitbox on poking moves, in particular crouching medium kicks, so that they feel less stubby.
  • Allow any special that is forward jump only to be performed on neutral and back jumps, except divekicks, where only EX versions can be performed on back jumps.
  • Allow chip kills with EX moves (a decent compromise from only supers chip killing and allowing any special to chip kill)
  • Bump up character walk speeds slightly, like 6-10%.
  • Make overheads safer on block, or only punishable with light attacks (decent risk, low reward instead of  moderately high risk, low reward).
  • Make it so crush counter moves are never plus on block, and so poking ones are not plus on normal hit.
  • Add charge partitioning??? (this one is kind of off the wall, and I don’t really know how it would affect the game)
  • Make limb hurtboxes slimmer, so limbs that visually appear to go over/under each other do so with more reliability.
  • Reduce min heights on air specials.
  • change all hits of all attacks to be JP6 at minimum, make all EX attacks JP∞. (More Juggle Potential would open up the combo system a lot without much more effort, dunno if this would fuck anything up, but whatever, I’m not here to do a careful study)
  • change V-reversal to be more similar to GG’s dead angle attacks, rather than GG’s bursts. So 10-13 frames rather than 17-18 frames, early frames are completely invincible, later frames vulnerable to attacks.
  • Bring back advantage on throw, increase throw invincibility on wakeup to 4-5f, so you need to scare the person knocked down into throwing, like in every other game.

I don’t really have a solid solution for “robbery V-triggers”, V-triggers that can result in extremely high damage combos for late-game comebacks. Nerfing them is the obvious solution, but that would seriously reduce the number of skillful combos in the game and change the feel of many characters. I think a more healthy move for the game would probably be to integrate a lot of the alternate V-Trigger movesets as basic moves, and removing V-trigger entirely, or toning it down significantly, but there’s no clear solution.

Similarly, I don’t have a solid solution for the way that nearly all combos are too easy. I think average combo length is good currently, and making average combos harder would require either reducing hitstop (like 3rd strike), or reducing the buffer (which honestly wouldn’t be a horrible idea, a 1f buffer instead of 2f. The smallest link window being 2f wouldn’t be terrible). However I feel like these changes would also hurt the accessibility of the game. Over time Capcom has implemented V-Triggers that allow for harder combos, thanks to negative edge, quick stance cancels, quirky button combinations or the like, but working more skillful combos into the game without also increasing combo length and the difficulty of basic combos is an extremely tricky problem. Having more meaty-only combos, or combos based on delays or moving hitboxes could increase combo difficulty and situationality.

It’s relatively easy to mod SFV, if someone wants to make a mod with all these changes, that would be pretty cool.

What’s the point of combos in fighting games?

Celia Wagar6 Comments

Many beginners get into fighting games and see these COMBOS and feel like, “bullshit, it’s not fair that they can deal a billion damage. I hit them twice as much and they win off one hit.” While that can be frustrating, combos add a lot to the game that you can’t get any other way.

The deal with combos is they make certain hits under certain circumstances more damaging than just any random hit. Games with longer combos allow players to find different combo routes that lead to different types of advantages, like more damage, better screen positioning, knockdown, meter gain, easier confirms, and safety on block. Continue reading →

Don’t Diss a Genre You Don’t/Refuse to Understand

Celia Wagar28 Comments

I said I wouldn’t review any more videos of other critics, but I couldn’t stand to watch this one and say nothing. I’m reposting here, mostly because I think it goes a ways to explain the differences between traditional fighters and smash. If you’re a smash player, please play other fighting games too. Please stop sticking to one insular franchise.

Continue reading →

Learning a New Combo

Celia WagarLeave a comment

So lately I’ve been playing Under Night In-Birth EXE: Late [st]. I decided for this game I wanted to play a more challenging character than I usually pick, and I heard Vatista was on the more challenging side, being a charge character that lets you charge in ALL FOUR DIRECTIONS and with button holds on 3 different buttons. Oh yeah, this article will include a lot of fighting game notation, so be sure to brush up: http://wiki.shoryuken.com/Notation

Vatista’s toolset is pretty standard for UNIST, she launches by sweeping with 2C, then hitting them with 5C to launch, then does an air chain into a divekick that carries her opponent down, then hits them off the ground into a finisher. The trouble is fitting her charges into the combo, especially the button charges. When Vatista holds a button for 1 second, she can release it, canceling any normal move, producing a crystal that will explode for a lot of damage when struck. So Vatista’s optimal combos involve getting a launcher, doing an air chain, then in the process of OTGing them, creating crystals and detonating them as the ender. Since the crystals require buttons to be held, you need to hold the buttons used in earlier attacks to release crystals later.

The other parts of Vatista’s moveset are a lot like Guile’s. [4]6 is a fireball, [2]8 is a flash kick. [8]2 is a divekick, which involves holding up for slightly longer than a regular jump.

This is made harder because I play on pad, so if I were to hold a button, I’d need to then fat-finger my thumb onto other buttons to press them to continue the combo. Alternatives are holding the pad with my left hand so my right hand can be turned so I use my index finger to manipulate the buttons, but that isn’t comfortable for me. Luckily, I own a Hori Fighting Commander 4. This contoller has R1 and R2 both on the face, and the shoulder. Fighting games don’t normally let you double bind buttons, but this controller gave me an exception. By binding B and C to R1 and R2, I now had them on the face and shoulder at the same time. This means I can hold a shoulder button while pressing the face buttons freely, and if I ever wanted to hold A, I can use the shoulder buttons to attack instead.

So I started the game out by learning an extremely basic combo, 2C 5C j.B j.C [8]2A 2C [2]8B. In this combo, 2C 5C launches, j.B and j.C buy time for me in the air so I can charge [8] for the divekick, which otherwise would take longer to charge than a whole jump. The divekick knocks them down, then I can OTG with 2C, and cancel that into a flashkick to end the combo.

Not so tough, I picked it up in a few training sessions, then eventually moved up to 2C 5C j.A j.B j.C land j.A j.B [8]2A 2C 2B [2]8B, which is basically the same combo, except you use the leftover hitstun from j.C to give you enough time to land and jump again to connect another j.A.

Then I moved up to using crystals, which meant finally using the shoulder buttons I had double bound. So the combo now looks closer to 2C 5C (delay) j.[B] j.C land j.A [8]2A 2C ]B[. This doesn’t have a real ender, but it sets up a crystal at the end. My first attempt at this, I decided to press j.B on the face, then immediately transfer my hold to the shoulder button, which worked, but eventually I realized it would be easier just to press the shoulder button directly. Another pain point was that I’d do 2C at the end, and release B, but no crystal would come out. I was releasing too quickly. I had to delay my release slightly to actually get the crystal out. That wasn’t so hard though. Easily enough, I had a crystal setup. It dealt less damage than my normal combo, but it was the first step, now I needed to learn the real combo.

2C 5C (delay) j.[B] j.C land j.A [8]2A 2[C] 2]B[ 2A 2B 5]c[
This is the combo I’m still working on. First thing to note is that I can skip the delay at the start if I’m willing to do j.A after launching (scales the combo, making it deal less damage), but that’s not really important. If I don’t delay properly, or use j.A, then the land into j.A later on won’t connect, they’ll be able to tech out after my j.C. The delay makes it so j.C happens later in the jump arc, as I’m landing. The next problem is the ender. It’s hard. So hard I decided to practice the ender by itself before integrating it into the combo. First I’d do this by holding down B for a second, then doing 2C 2]B[.

The first problem I had was doing 2A after the crystal release. I’d miss the link, then do a 2B that wouldn’t combo. Or I’d mash and get 2A 2A which wouldn’t combo. Eventually I fixed this by watching vatista’s crystal release animation for when she would retract her hand, instead of watching the opponent’s character (which I normally do during the juggle).

the next issue I had was releasing C to get the second crystal out. The second crystal needs to be released before the first one is done exploding. This means it needs to be released on the first hit of B and no later. My technique at the time for getting the C hold was pressing the C face button, then releasing the B shoulder button, then holding the C shoulder with the same finger, and then doing A and B with my thumb on the face buttons. But this technique had a problem, I was releasing C when I was transfering my hold from B to C on the shoulder buttons. So I’d do the pickup correctly from the first crystal release, and oddly find I could not release the second crystal off the first hit of 2B. I needed to wait a little, and waiting at all meant the second crystal would not be released soon enough to detonate.

Eventually I realized that I needed to continuously hold C from when it was first pressed, but this has another problem, I can’t hold down the C face button because that means I can’t press A to do the OTG. And I can’t just press the C shoulder button, because that would mean releasing B shoulder, so I couldn’t release the first crystal (I could use 2 fingers, but I don’t want to). So my solution was to hold face C a bit longer while I transferred my hold from shoulder B to shoulder C, before I needed to press 2A for the OTG. With a bit of experimentation, I found this worked and I could now do the entire ender! Now that I could do that consistently, I just had to put it together, but that’s where I ran into another problem.

I’d do the entire air chain successfully into the knockdown, then 2C and the crystal release, but couldn’t do the 2A to follow up. It wouldn’t combo! This seemed similar to my issue of timing from earlier, but that wasn’t it. I could do the link fine when practiced by itself, but I can’t do it in the middle of the combo, because I was canceling 2C into crystal release too late. Why’s that? because when I practice just the ender, 2C is my first input, and I only press it once. When I was doing it in the middle of the combo, I was mashing 2C to make sure it came out, but this meant I didn’t know when it was actually coming out, and I couldn’t cancel into crystal release at the correct timing on reaction. So I needed to reduce my mashing to one single press at the right moment in order to do the whole ender.

And that’s about where I am. I can do every part of the combo individually, and I have landed the whole thing in a match once before, but I still need to practice to actually get consistent at it. The hard part is, there’s so many little parts that make this difficult, it’s hard to keep consistent at all of them simultaneously. Also obviously playing on Pad makes this a lot harder than it normally should be, but I’m having fun with it, and nothing is impossible to pull off, even with my ridiculous limitations, which fits the bill for trying out a higher execution character. Plus Vatista is insanely fun, she’s the charge queen with a crazy amount of footsies play too.

Overall, I think there’s something really special in coming up against hurdles, experimenting to figure out the nature of the system, including the actual game, my controller of choice, and my own physical and mental processes of execution.

DBFZ Impressions

Celia Wagar6 Comments

What’s your opinion on FighterZ so far?

Alright, I checked out the DBFZ beta while it was up. I played some ranked, and honestly got my ass scraped. Had like, a 50% win ratio exactly. Feel like I gotta count my losses too.

Felt like I had no idea what the hell I was doing or how anything fit together, but then I watched a vid that explained the universal moves and the basic combo mechanics and it made a lot more sense. Once I actually got the game, I started winning a lot more, and overall the pieces fit into place a lot more, it can just be a bit rough starting out.

I’m gonna use anime numpad notation, so look that up if you don’t know it. 2M and 6M are a universal low, and a universal overhead respectively. 6M is a lot like the 3rd strike UOH, but it can’t be hit meaty, and doesn’t chain or cancel into anything, so unless you call an assist before using it, you can’t combo off it. 2M is a sweep that usually moves forward a bit.

Dragon rush is this game’s equivalent of throw, but it has a startup period before rushing forward to grab. It has low priority, so anything with range will knock you out of it. It sometimes loses to normals, always loses to beams, it’s a pretty crappy throw. It also serves as your snapback, letting you force the opponent’s character out if you land it. Since it has a startup, tick throwing in this game is basically absent. You’re only gonna get tick dragon rush off maybe an assist block pressure, or a reset really. Otherwise, you need to make them scared enough to continue to hold block long enough for the dragonrush to start up, and that’s probably reactable. This means that if you want to fuck up someone who is blocking, you need to mix them up, high and low.

Also, you can cancel ground pressure into superdash, vanish, or S (Ki Blast) to extend it. Ki blast can be canceled into projectile specials, then super, which sometimes can serve as a weak confirm, however your real damage comes from either confirming into H, or getting a combo off 2M.

So that’s the next deal, there’s an interesting progression in the combo system that has all these parts that are really easy to tack on to add up to bigger and bigger combos. First you have the L and M autocombos, which do like 2 normal hits, then a unique attack, then for the L combo, it leads into a launcher, then hard knockdown, and for the M combo, it does the 3 hits, then cancels into special, then super. There doesn’t appear to be any scaling on these, so they get decent damage, like a quarter to third of someone’s health. If you hold back, you can get the unique 3rd hit, without getting the rest of the autocombo. So you can confirm off random hits in neutral, then autocombo your way to victory as an introduction, then you can ramp up to doing L > M > H, which the tutorial shows you how to do. 5H and 2H are your launchers. 5H will launch horizontally and wallbounce, and is usually a big poking move, 2H launches vertically, and is invincible to air attacks, which is important. Then you can do an air combo like L > M > H for a hard knockdown, and then super when you land.

Once you’ve gotten those basics down, you can get a bit more mileage in a couple different ways. First, you can do 2M to sweep, then 5M to launch from the sweep and jump cancel the 5M to follow into the air. Then in the air, instead of doing L > M > H, you can do L > M and jump cancel into another L > M, then cancel to special, then to super. Or, instead of the special into super, you could do 2H, which launches them even higher, and lets you follow up into L > M > special > super. You could also vanish instead of super, letting you combo the wallbounce from vanish into dragon rush for a snapback. You can integrate these combo extensions at pretty much any point for a little extra damage, and far as I know, they’re universal across the cast. Even if you don’t do 2M > 5M, you can still integrate jumping L > M double jump L > M from the H launchers.

Oh, and the chain system is weird, you can chain any button into any other version of that button, but you can’t have the same move in a chain twice, so you can chain 2M > 5M or 5M > 2M, but not 5M > 2M > 5M, which is what would lead to a launch, if possible. This also means you can chain 2M > 6M, the low into the overhead, but 6M cannot chain, so the overhead only combos with an assist.

Beyond that, there’s advanced character-specific combos, which involve comboing off specials or using assists or other business to extend, but still, there’s a basic template that lets you get a good combo, which you can steadily ramp up through and improve at without adding too much complexity at any given point. It also means you can pick up a new character and figure out how to do basic-bitch combos with them fairly quickly by following this template.

The superdash is a big deal, it’s kind of like a street fighter jump, in that it lets you get in, it’s safe on block, and combos on hit, but if they’re paying enough attention to AA you, there’s not a damn thing you can do, and they’ll get pretty decent damage off it to boot. 2H is completely air invincible, so if properly timed, it will beat superdashes clean. Superdashes can come out quickly however, and are tough to react to at close range. They deal a hit as they come in, which can be combo’d off. If blocked, then the attacker is not punishable, and they can get out an attack before hitting the ground, making it kind of a mixup scenario on block, which I’d guess is weighted slightly in the defender’s favor. The attacker can also double jump on block to make it almost completely safe. Unlike normal aerials, superdash can be blocked low or high, and you can cancel into superdash from L, M, and H, extending pressure. This move is gonna be the noob killer, because of how difficult it is to react to and shut down and how high priority it is. It goes through small ki blasts, but can’t go through beams and larger projectiles, so you can use those as a less guaranteed and lower reward anti-air, albeit with less harsh timing requirements than 2H. Superdash also has reasonably high priority vs normals.

I’m a bit disappointed by the tutorials and combo trials. The tutorials explain the bare basics, but don’t explain how to play the standard game, or that 2H is invincible. And it’s lacking Guilty Gear Revelator’s more advanced mission modes that fill in that type of information. The combo trials are also fairly simple, and don’t really go through the ramp up in complexity that I mentioned, at least, not nearly as far as Guilty Gear Revelator’s. So a lot of people are gonna have to go online to learn more information about the game, which is disappointing, and might detract from the image of the game, as people get stymied by simple tactics.

The game has a ton of hidden features, both character-wise and system-wise, so it’s shaping up to be an excellent game overall. My notes file on the game is huge. I like the game a lot so far, but I haven’t dedicated much time to improving in it.

How to Cater to Fighting Game Beginners

Celia WagarLeave a comment

How do you feel SFV and MvCI’s approach to catering to newcomers compares to Sakurai’s approach with Brawl and Smash 4?

It’s not nearly as bad, but it’s very similar in its heavy-handedness.

Instead of trying to lower the skill floor of the game, they did that and lowered the ceiling as well. Honestly, I don’t know as much about MVCI, it sounded like that game did fairly well except for hitstun decay being tied to damage, and the discoveries people (infinites, hitstun decay glitch, the new unblockable) making progressively making the game less and less interesting to play, but whatever.

Characters in SFV don’t have as much potential for combo creativity as characters in SF4. They made combos easier and more understandable, but they also reduced the number of moves on each character that link and limited the juggle properties of characters even more severely than the juggle points system in SF4 (unless I’m mistaken. SFV’s juggle system is still a bit weird and hard to understand). SFV was intended to correct SF4’s mistakes, but they overcorrected and ended up making new problems.

The issue is, they removed a lot of the potential for advanced play from the game in the name of making the game easier to pick up, and those two goals aren’t mutually inclusive. Having a lower skill ceiling doesn’t attract beginners, and honestly, neither does making the game easier. SF4 was one of the best selling SF games in the franchise, despite being arguably so difficult. Casual fans are more attracted by having a large roster of characters that they like than anything specific about the game systems. Casual fans are more attracted by single player content like Mortal Kombat’s. Brawl and Smash 4 at least delivered on those fronts, there’s a lot you can do without needing friends to play with you, and the rosters are large.

Dragon Ball FighterZ has a really neat auto-combo system that helps bridge the gap even better, without compromising high level combos in the process. Basically, you have 2 autocombos, LLL and MMM, and these will do simple combos. The light autocombo does 3 unique light attacks in a row, the 3rd one launching into a low air combo, where you can again press LLL to do light, medium, and heavy in a row. The Medium autocombo does standing medium, crouching medium, a special, then super if you have meter. What’s cool about these is, lights chain into mediums, and mediums chain into heavies. So instead of doing the next hit of the auto combo, you can at any time do a better attack instead.

What’s also cool is, later on in the combo, you can default back to doing the auto combo. So beginners can easily do LLL or MMM, but they can substitute for better attacks, then default back to the easy stuff where they don’t know the way yet. So you can do MM, but instead of continuing, you could press H to launch the opponent, then go back to doing LLL in the air to finish up the combo. If you do crouching H, then you will even get a hard knockdown off this, which you can combo into super.

In this way, beginners can slowly ramp up their combos into more optimal sequences, LMH > LLL, or LMH > LM > special > super, or 2M 5M > jc > LM 2H > LM dj LM > special > vanish > special > super. The combo trials hint at a lot of these common combo properties across characters, though I still wish they were more extensive, and the tutorial was better.

And the best part is, advanced players still have difficult and advanced juggle combos they can perform for optimal damage, or using assists and so on, so unlike MVCI and SFV, there wasn’t anything compromised in the process of building this system. The air autocombo even has special height gaining properties that make it useful in combos.

I still think they dropped the ball a bit on the tutorial end of the game however. It can be difficult for players to realize all this, especially because they probably don’t understand what’s happening in the combo trials, they just see a list of moves and don’t understand all the different ways it can go. It also doesn’t help that they didn’t include the basic universal B&B anywhere in the Combo Trials. Oh well.

Fighting Game Auto-Combos

Celia Wagar1 Comment

What do you think of auto combos in fighting games?

Short answer: Letting people mash a single button to get an okay-ish high-commit combo that does not mix up on block in a game that is heavily based on landing difficult high-damage combos is a good idea.

Marvel style games and anime fighters are frequently about landing big, high-execution combos for tons of damage. I’ve learned a few combos in skullgirls and marvel 3, and they get pretty long and tough, even if most of the individual stages are fairly lenient.

Individual hits in these games tend to be low damage, and exist more to define the entry-point of the combo, limiting your ability to confirm and what route you take through the combo. Damage scaling and proration also affect the total damage of the combo based on the opener. It might be easier to confirm off a light (because they self-cancel and come out fast), but you might be shaving off tons of damage in the process.

The perspective to take for these types of games is to think of the combo as one extremely high-damage hit, imagine that in in-game time, you’re skipping directly from the beginning of the combo to the end. Having the combo have all these moves in-between allows combos to vary more widely in their effects and damage, so in “one hit” you can have a much wider variety of outcomes, pieced together by all the moves the attacker chooses to assemble in the combo. From that they can get pure damage, resets, corner optimized combos, corner carry, hard knockdown, character switching, etc.

It also means that different players can demonstrate their skill and creativity with the combo system, practice to improve their damage output, produce different results from the combo system.

However getting to that level is hard. It requires understanding a lot about the game, and having skill at actually executing combos.

So auto-combos give people an easy in. They can play the neutral game using all the different moves they’re given, then confirm into an auto-combo off their openers. They let people deal respectable amounts of damage in a game about dealing a lot of damage, but ultimately to get the real damage, they’re gonna have to learn. Autocombos don’t really affect competitive play in any way, they don’t imbalance the game, they let beginners get an idea how to use combos, how to confirm, etc. So in this respect, pros don’t care about autocombos, so it’s fine to have them.

At a low to intermediate level however, autocombos feel weird. I did a twitter thread on this: https://twitter.com/aGrimVale/status/953268702056763392

As a beginner who was good at neutral, bad at combos, I felt like autocombo systems were a violation of the idea of effort vs reward, and they are. I disliked combos in general at the time, so I didn’t really like games based on long combos. Combos can feel unfair, even if they aren’t actually unfair. It’s like, “why are we even playing each other if only one guy is doing anything?” As you get better at combos and dealing with setups, you get more patient with other people’s combos too, and you know to look out for when they drop or reset. Unless it’s Hokuto no Ken, then you just leave and grab a drink.

Beginners don’t have a good sense of the fairness of fighting games. A common perception is that mashing is just as good as deliberate play, and autocombos add to that, because low level players aren’t familiar with blocking and punishing. It’s also feels like handholding, like simplification of the game, and you’ll see the same combo a lot and you know they’re never ever gonna drop it, so that kinda sucks. So in a way, it simplifies the game at a lower level, because low level players shoot to just do the same thing over and over again, because it’s so much more rewarding than anything else you can do, and it’s frustrating to try to play without that until you can do real combos. You feel like you’re trying to engage with the actual game, but you’re being punished for it, because using multiple moves in neutral and trying to make do without combos is so much less effective than just using the damn auto-combo.

The trouble with auto-combos isn’t that they’re broken, it’s that they give people an easy crutch, but they don’t understand how to build from there. We don’t have the best on-roading structures for leading people to develop better fighting game knowledge and skills. A lot of it requires research outside the game, or having a friend who knows better, and even then, it’s difficult to really understand the basics. The tutorials in most games are insufficient, and they’re external to the process of actually playing. Most games don’t have a medium between autocombo and manual combo, and it doesn’t seem like the kind of thing you can really build (Though DBFZ tried its damnedest).

I don’t have a good solution to this, but hey, autocombos are a neat stop-gap that hurt the game less than people think. If we can figure out a way to lead people to naturally learn about combo systems without having to divert them to another mode, that would be great, but I just don’t see how we can do that.