Tag: Combat Design

Final Fantasy 7 Remake: Active Time Battle Finally Comes Home

There was a lot that impressed me about the Final Fantasy VII Remake, released 2020. For me, it set the standard for the full revitalization of an old game. Or rather, it would be if it weren’t only part one, a game that covers a fraction of the story of the original. Repackaging the story of one game into several full-sized, full-priced games, re-imagining the combat systems completely, expanding and rewriting the story. These are all controversial decisions, though they are assuaged somewhat by the existence of Ever Crisis, another full remake of Final Fantasy VII that is a more traditional, more faithful remake using updated assets. Indeed, there are so many interesting questions surrounding this game one could discuss. For now, let’s talk about the changes to Remake‘s combat systems, what experience they’re trying to accomplish, and why they are the way they are.

A girl in a plush animal-shaped hood and a stoic man with a staff team up to beat the ever-loving-stuffing out of an ostrich-like carnivorous bird.
ATB: The gentleman’s method of abstracting physical altercations with wildlife

One feature of Remake‘s combat that I find most interesting is the presence of the ATB bar. For those unfamiliar, Final Fantasy has employed at various points since its fourth installment the Active Time Battle system, or ATB for short. For the folks who may have been riled by Square Enix’s recent efforts to bring Final Fantasy more into the realm of action games, and deployment of increasingly arcane systems combining action with turn-based combat, rest assured that this is not a new trend. They have been trying to do this since the 90s. In its traditional use, the ATB bar is a timer that counts down to when a player character can act in battle. The faster that character’s speed stat, the more often they can act. When action is not possible, characters stand idle. So essentially, it’s a traditional turn-based combat, but with the wrinkle of an active timer.

At the heart of the ATB system’s ambitions, successes, and failures, I think is this tension a lot of combat-oriented games experience between their action elements and strategic elements. There advantages in verisimilitude, theoretically, as a real-time element brings with it less abstraction, but also a loss of many of the advantages that turn-based represents.

Turn-based games are inherently strategic in their foundation. The benefit of having combat play out over an indefinite amount of time is giving the player the space to make informed and meaningful decisions to gain an advantage over the opposition, which is where the challenge in turn-based combat comes from. Enemies need to be powerful enough to require critical thinking and problem solving in order to be engaging. Problem solving and critical thinking requires time to think. On the other hand, action combat is often all about twitch reaction, muscle memory, and being present in the moment as developments unfold around you. This is why you often finding me describing real-time game systems as ‘action’ ones. Now, action games can and often do have strategic depth as well, but the strategy of a game like soccer exists on a different state of mind than a game like chess.

There is an inherent friction between these ideas that can crop up in any system that aims to combine the two experiences. If action is meant to be all about the pressure of events happening in real-time, when do you get the time to develop your strategy? If that space for critical thought and puzzle-solving is needed for a turn-based game to have its strategic depth, how do you provide that space without disrupting the flow and rhythm of an action game?

In the ATB system’s early uses, the main advantage I can see is its ability to offer variable turn frequency through the speed state. It’s an interest dimension to scale character power, not one that is super frequently employed in turn-based games. Usually enemies and players only get one turn each, in even ratio to one another. Allowing that ratio to change can create some very interesting scenarios. There are disadvantages though. Personally, I love taking my time to formulate the perfect strategy and course of action in any given situation in a turn-based game. It’s a big part of the draw for me. The ATB system, which allows enemies to continue acting in spite of the player, demands that one keep taking actions or quickly become overwhelmed. Older Final Fantasy games would often have a mode to address this, but the wait mode of of old Final Fantasy is effective the same as regular old turn-based, kind of defeating the purpose. I believe Final Fantasy VII Remake has an optional mode akin to this, though I would argue Remake’s combat is a fuller realization of ATB’s design goals regardless. The argument that the wait mode represents a lack of confidence in the ATB systems early on is stronger I think, when one examines Final Fantasy X, a game in which the speed stat still exists in the form of agility, but represents the frequency of turns for an actor within a non-real-time system. Static turns are simply queued up in an ordered list, with faster characters getting more turns. The real time element here, was unnecessary, and therefor not included.

A girl in a plush animal-shaped hood teams up with a stoic man wielding a staff to pummel a leopard-like monster. Time stops after a moment while a UI element is navigated, after which both characters unleash powerful flashy moves.
In FFVIIR, real-time actions fuel strategic resources, which fuel combat advantages

What are those design goals then? Well going by the nomenclature it seems the Active Time Battle systems was meant to make turn-based combat more actively engage the player with its time-sensitive elements. Final Fantasy, like Dragon Quest before it, utilize turn-based combat akin to the tabletop games that inspired them. Turn-based combat was developed as a way to create an abstract representation of combat for the safe environment of a game. There aren’t a lot of practical ways to really play out hitting each other with swords while gathered around a table, without this layer of abstraction. Nowadays, I think there’s a lot of appreciation for the unique experiences possible only through a sophisticated turn-based system. I get the sense that even in the 90s, all the way up to now, some at Square Enix felt as though the turn-based system imposes limitations on what the combat can be, and imposes a barrier between the player and being immersed in the game’s world.

I think that last point is debatable, but what isn’t, so it seems to me, is that the ATB system wants the player to feel more pressured, with a more present sense of danger and action playing out in real-time. The problem for me, is that the simplicity of the ATB system feels like somewhat of a half measure. It’s a little at odds with the strategic systems of Final Fantasy combat, but also doesn’t fully create a sense of urgency or action for me. It’s kind of just a turn-based combat systems where you jump through a few extra hoops. The first real exception to this for me… is… not Final Fantasy VII Remake, which would come a lot later. Chrono Trigger is a vary notable instance of a game taking the ATB system and pushing it in an interesting direction.

A red-haired swordsman is accosted by a pair of blue goblins. One kicks him. The swordsman then spins his blade around in a cyclone, hitting and dispatching both goblins at once.
Wait for the opportune time to use the right move, and this real-time element makes a lot of sense

The key for what makes the ATB system work for me in Chrono Trigger is the importance of player and enemy positioning. Some of the Final Fantasy games had shades of this, but it was very digital and binary. IE: If you’ve ambushed enemies you have an advantage, if they’ve ambushed you, they have the advantage. In Chrono Trigger though, the relative position of enemies is analogue, constantly changing, and an ever-present strategic element. This element is often just a small, simple detail. For example, when several enemies are close together, you may be able to hit all of them at once with a single special attack, but not if they’re spread apart. If an enemy is too close to the player character when attacked, they may immediately respond with a counterattack. That sort of thing.

This simple extra dimension was like a missing link for me. Real-time is only one element of what makes action games engaging. In a traditional turn-based game. Things such as multi-hit attacks and counters are often very static and abstract. A multi-hit attack might always hit three enemies, or always hit enemies adjacent to one another, though they never move. In Chrono Trigger, and games influence by it, enemy positions are organic and ever-shifting, a natural and intuitive extension of the world. This ingredient alone makes the ATB bar make sense, as the dynamics of battle change in real-time along with the turn order. As I’ve often said elsewhere, relative position and spacing management on the part of the player is also an essential element. Chrono Trigger had a great commitment to holistic nature of its world. There are no random encounters. There are no battle transitions – the combat takes place in the same physical space it is initiated in. And, of course, battles occur in real-time. The game works to minimize abstraction in the broad strokes to make the game as seamless and cohesive as possible. Chrono Trigger brings these elements together in the context of a turn-based game, and was the last real evolution of the ATB system, in my opinion… until…

A red-haired young swordsman teams up with a blonde young woman with a crossbow to fight a large red robot in a festival courtyard. The swordsman strikes, but is immediately countered by the robot, who unleashes a spring-loaded boxing glove fist from a compartment in his chest.
Disadvantaged Positioning
A red-haired young swordsman teams up with a blonde young woman with a crossbow to fight a large red robot in a festival courtyard. After a moment, the robot wanders a little ways away from its opponents. The swordsman runs up to strike, then jumps back to his starting position, and the game's UI reads "Too far away to counterattack." at the top of the display.
Advantaged Positioning
Chrono Trigger‘s sophisticated spacing mechanics seem to have been a big influence on FF7R

Well okay, there are a couple of other Square Enix games I want to talk about that finally lead us to where we are today with Final Fantasy VII Remake, but first I’ll finally talk about Remake. Remake‘s primary conceit in its combat is that you can run your player characters around a 3D space and directly command them to attack in real-time, taking after the likes of a Kingdom Hearts or Devil May Cry. While not as robust a system as those, Remake‘s action makes for an effective stopgap of realism to reinforce what is being represented by the more abstract strategic decision making that takes place between – the action forms an important bedrock of context. Kingdom Hearts is another series that initially attempted to marry turn-based combat’s strategic elements to an action system, but ultimately fell head-first into the symbolic dream-ocean of full-on action game by its first sequel. The wrinkle for Remake, though, is the player’s ability to pause the action at any time to assess the situation and execute specific commands for the party members, commands which are invariably a lot more powerful than the standard real-time attacks.

A lot of games have tried this ‘stop and go’ sort of combat. Most prominently I think of old CRPGs like Baldur’s Gate, Planescape: Torment, or Dragon Age for a more recent example. These games take direct inspiration from Dungeons and Dragons or are indeed, even licensed games tied to the property. Weirdly though they’d often not employ turn-based combat at all, unlike the actual tabletop game they’re based on, almost as if turn-based is this inherent limitation to be overcome. I wonder if the developers at Square Enix ever played these games, or were inspired by their indifference to turn-based combat, haha. Suffice to say I think turn-based combat has almost always been underappreciated for its merits in many spheres of video game development. What made combat for a game like Dragon Age a little dull for me, was in how the real-time segments and paused segments, the player characters’ capabilities are exactly the same. So ultimately, I just felt like I was disrupting whenever I needed to pause the game to line up a spell, and since all characters are only indirectly controlled through commands, I never felt fully engaged during real-time either.

A girl in a hood shaped like a plush animal summons mystic runes around her to launch magic attacks at a soldier. Her partner, a stoic man with a staff, dashes up alongside the soldier at the same time she does, to do a combination attack.
Satisfying action, satisfying strategy, but most importantly the two modes support one another

This is where Final Fantasy VII Remake really accels, I think. It’s built on the solid foundation of an action game, with a full suite of direct control for the player, from blocking to dodging to parrying. And you’ll need those options, as the enemies take full advantage of their freedom of movement while the game is un-paused. Engaging with this action system builds up your ATB gauge, no longer a mere function of time, it is a function of engagement with the core systems. At certain thresholds of ATB, you can command a party member to unleash a powerful attack, and are given the ability to pause the action to carefully consider how and where to employ these ATB abilities. This pausing never disrupted the flow of the game for me, because they’re always big moments I was building to – rewards for effectively playing the game. The barrier between what can be done during real-time, and what cannot be, also very effectively delineates the two modes, so I never felt as if one is intruding on the other. I have to pause the game to heal a party member, or to unleash a skull-cracking dive kick. These actions are not something I can just do on the fly. They require strategy and consideration, so their presence within a pause sub-mode of the game feels natural. The ability to pause is a godsend too, as the player characters each have very intricate combat mechanics unique to them that, when used in tandem with careful consideration, can have some devastatingly powerful results.

Cloud, Tifa, and Barret from Final Fantasy 7, beat the stuffing out of a soldier in a sci-fi wearhouse. The action pauses mid-fight, in dramatic slow-mo fashion, then the beatdown resumes.
It’s also all very pretty to look at.

Another key to this is the stagger system, which is also an ongoing iterative system featured in many Square Enix games. It creates a long-term goal in any given combat encounter that the player can build up to. By using the right moves at the right times, enemies can be staggered, opening them up for massive damage. A balance of micro-goals and long-term goals is essential to a compelling combat system. From not getting hit, to setting up an enormous combo. It’s these interlaced interest curves, build ups and releases, that keeps combat from getting stale.

This makes for a sophisticated system of setups and payoffs, where every party member is instrumental in stringing together combination maneuvers. Cloud can easily set up a stagger on an enemy with his powerful counter attacks in real time and big sword strikes from the command menu. Meanwhile, Tifa can empower her abilities by stringing together a series of strikes. At the same time, Barret, who can take a lot of punishment, is safe to set up some buffs like haste to increase the party’s speed. Once the target is staggered, Tifa can unleash her built up power to fully exploit in the enemy’s weakness, and Cloud can finish them off once they’re low on health. The fantasy that Remake really delivers on for me is twofold; One, it creates this delightful scenario of being able to pause and survey a complex battlefield, almost like stepping into a freeze-frame of an action movie, owed to how well the real-time segments of the game operate in isolation to create impressive spectacle and compelling scenarios. I can survey these wild combat engagements with a strategic, discerning eye without fear of reprisal. Secondly, the way all of the character’s complex mechanics mesh together creates for me the sense that they are a real, team, working together to accomplish incredible things they could not do alone.

Cloud, Barret, and Tifa team up against soldiers in a futuristic warehouse. Barret, a man with a gun for an arm, casts a spell empowering Tifa, a woman who then punches a soldier dozens of times in a second, before the action pauses again, then resumes as Tifa knocks him over with a dropkick.
Buff up, wear down the target, knock em down. Just as planned!

This schema didn’t appear fully formed in Remake from thin air or divine inspiration, though. Many might not realize Final Fantasy has been trying to achieve this awakening of the ATB system’s dream for years. I can think of three major stepping stones of this journey: Final Fantasy XIII, Final Fantasy XIII: Lightning Returns, and Final Fantasy XV. All three of these games have bits and pieces of the puzzle that would make up the combat system of VII Remake, each containing the germ of an idea that you can see grow into what we have today.

XIII established party-wide strategic command overlaid upon action that exists in a different mode from that strategy. There’s been a lot of criticism over the lack of player engagement for this in-between mode, but the basis of VII Remake started as far back as this game. It became an example how combing two modes of play without disrupting pacing could work, although the quality of the pacing of XIII‘s combat is debatable. The later sequel Lightning Returns did away party dynamics, but intimately explored the stagger system, and how indirectly controlling player characters in real-time could work.

XV was an interesting beast. The result of an overlong development time, branched off from a very different design concept that leaned more on action gameplay. XV shows signs of its troubled development, but I think none is more prominent than the combat system. It isn’t bad, per say, but it feels very underdeveloped. A lot of the ideas that shine in VII Remake exist here in a sort of primordial form. Optional wait modes exist alongside a limit-action systems that puts emphasis of enemies positioning, and party-wide combination attacks. Everything is just a little off. A lot of those micro-goals I mentioned don’t feel as though they are totally in-sync, and long-term goals within combat aren’t totally clear. Nothing quite meshes like it feels as though it should, but all the ideas are there. Whether you actually like these games or not, the long and short of this is that Final Fantasies XIII, XIII: Lightning Returns, and XV walked so VII Remake could run.

The old lessons from Chrono Trigger live at the heart of Final Fantasy VII Remake as well, with the relative positions of enemies and party members in 3D space determining how actions resolve. A big sword swing might hit a dozen enemies at once if you’re careful about positioning! This is taken further than Chrono Trigger even, with the ability to finely control exactly where each party member is standing, and a third dimension to play around with. It’s an intricate dance of set ups, payoffs, and strategic positioning wrapped in a solid-as-stone action game bedrock crafted by old hands when it comes to hack-and-slash ARPG gameplay.

A girl in a plush animal-shaped hood, wielding a giant ninja star, runs up to a group of giant ratlike monsters. The action pauses for a moment, then resumes as she places a magic rune on the ground, which sucks in all the rat monsters, dispatching them all at once.
Real-time danger is the risk one can weight against the reward of advantageous positioning

So Final Fantasy VII Remake does not have turn-based combat, and that is an understandably big loss for some people. What it is not, however, is a turn-based combat system trying to masquerade as a hybrid of action and strategy. Rather, it is the realization of that decades long dream to create an engaging hybridization of strategy and action RPGs. It is at last, finally, the ATB system come home. It is the design goals set way back then finally pushed to their apex, at least so far. Fully realized, the ATB can shine in accomplishing what it’s struggled to do for years. I’m also pretty impressed, and thankful, that a second completely differently imagined remake of Final Fantasy VII is due in the form of Ever Crisis, which seems to retain the old combat system, as a companion piece to Remake. Remake‘s redone combat aims to capture its predecessor’s strengths in a heightened form. It’s a brilliant system to build off of in that regard, and has become one of the biggest reasons I remain very excited for the future installments of Final Fantasy VII Remake.

A blonde young man with a huge sword, an unarmed, long-haired woman, and a large man with a gun for an arm team up to battle soldiers in a futuristic warehouse. Time slows to a crawl as a menu of options is manipulated in a game UI. When time resumes, the party teams up to unleash devastating attacks on the soldiers.

Attack while its tail is up! It’s gonna counterattack with its laser…

Kingdom Hearts ‘Floatiness’ and Gamefeel

Alternatively: “What The Heck is ‘Floatiness’ and Is It Ruining Kingdom Hearts? I Settle it Forever“

‘Floatiness’ is the extremely scientific term that parts of the Kingdom Hearts fan community have adopted to refer to a sort of shift in how some of the more modern and spin-off Kingdom Hearts games feel to play, as opposed to their predecessors. I’ve seen this term come in and out of vogue when it comes to in-depth and armchair analysis of Kingdom Hearts‘s combat, but it’s always fascinated me. I’m going to try to define what this somewhat nebulous term is specifically referring to, and how it’s been supposedly creeping its way into one of my favorite action RPG franchises. What causes floatiness in particular? What is the greater context of design which is causing whatever this ‘floatiness’ is to happen? Perhaps this exercise can help us explore the often imprecise art of ascertaining gamefeel in general. I certainly hope so.

Sora, spiky-haired protagonist of Kingdom Hearts III spins his key-shaped sword around quickly, then slashes with an uppercut at the air in the dimly lit streets of San Fransokyo. He then jumps and does the same motion elevated about 15 feet in the air before falling back to the ground.
Pictured: Floatiness?

First off, we’ve got to define it, this floatiness. Informally, it is reported as the sensation that the Kingdom Hearts player character lacks weight, impact, and to my interpretation, immediacy, or some combination thereof. For those unaware, Kingdom Hearts is a series of action games with a particular emphasis on exaggerated, superhuman feats of acrobatic melee combat, favoring style, spectacle, and emotive action, akin to combat-heavy anime and manga. All that while immersed in a fantasy world full of Disney characters, Final Fantasy characters, and a unique dream-like fictional mythology of its own. Trying to define this term will be much of what this article is seeking to accomplish.

With combatants consisting primarily of magical swordsman and dark wizards, while borrowing aesthetically from classic Disney animation, Kingdom Hearts offered its players a combat system increasingly free of the bounds of gravity, allowing its player characters to bound meters into the air at a time, maintaining that airtime through continued swinging of their sword, with a cartoonish sort of physical logic.

So, all this in mind, I set out to compare the gamefeel of several Kingdom Hearts games in my own experience to try and get a better idea. To make this process somewhat more precise, I investigated what is by my estimation one of the key metrics in determining the gamefeel of an action combat system – the timing of the primary attacks. Every action in a game as visually rich as Kingdom Hearts and its sequels has a startup time and ending lag. The former is the real-time between the player’s button input and the resolution of the action they’ve input, the latter is the real-time between the resolution of the action, and when the player’s next input can be resolved. For example, if a player character is to swing a big hammer, they’ll push the swing hammer button. As their avatar lifts the hammer with a grunt and struggle to convey its weight, we have our startup time, and as he pulls the hammer back to his side after the big swing, we have our ending lag, with the avatar unable to run or jump while resetting the hammer, in this example.

Sora from Kingdom Hearts II swings his keyblade at a volleyball in elaborate spinning arcs, launching the ball further into the air with each strike, and Sora continues after it, ascending higher and higher before falling back down.
Trust me it… it does feel very responsive.

Now imagine the hammer is a comically oversized key and you have a pretty good idea of what we’ll be measuring. Floatiness became an almost absurdly hot-button topic among Kingdom Hearts diehards due to its relationship with the much anticipated Kingdom Hearts 3. As the successor to the beloved Kingdom Hearts 2, and first numbered entry in the series for a period of thirteen years (though by no means the only Kingdom Hearts game to release in that period), a lot of anticipation was placed upon what the game would feel like to play. Its gamefeel, so to speak. Kingdom Hearts 2 has stayed in many people’s hearts as an all-time favorite action game due to the smoothness of its combat system, as any fan would tell you. So comparisons were inevitable. The stakes were high.

Kingdom Hearts 3, despite being the best selling entry in the series, gained a healthy amount of criticism from long-time fans in regards to its gamefeel. Floatiness was oft invoked. The developers seemed to take feedback from hardcore fans seriously, as the game would be later patched in key ways to increase speed and fluidity of the gameplay. More on that later. So what’s the truth? Has floatiness ruined Kingdom Hearts forever? Are we doomed to forever circle the drain of game design discourse without knowing what on earth we’re actually talking about? I’ve spoken to friends who know how to use a computer and they allege that numbers are involved here, believe it or not. So I’m afraid I have no choice but to deploy the diagram.

Data World

A 2 axis chart. A list of parameter such as "1st ground attack startup time" make up the Y axis. The different Kingdom Hearts games being tested make up the X axis. KH2 seems to be the fastest game judging by data, with KH1 shortly behind it. KH3 is in the middle range, behind KH1. BBS and DDD are nearly tied for slowest game.
So beautiful and horrible all at once!

First let’s define some terms. All footage I used for this analysis is from the Final Mix, 60fps versions of the games, captured on a Playstation 5. All values are measured in animation frames. For the sake of consistency, in each game I tested the animations of main player character Sora or his closest equivalent (In Birth By Sleep the resident Sora-like character is named Ventus, but he’s close enough and for our purposes we’ll consider him a “Sora”). The games I will be looking at are as follows:

Kingdom Hearts

Kingdom Hearts 2

Kingdom Hearts 3 pre-“Remind” patch

Kingdom Hearts 3 post-“Remind” patch

Kingdom Hearts: Birth By Sleep

Kingdom Hearts: Dream Drop Distance

“Startup” here is defined as the time between Sora beginning an attack and that attack “resolving”, here defined as the moment his weapon crosses in front of his body. “Ending lag” is the time between an attack resolving and when Sora can move again, no longer committed to the attack. “Cancel Time” is the time it takes before Sora’s first attack can be interrupted to perform his second combo attack. “Gravity Hang Time” is the amount of time between an aerial attack beginning and when Sora reaches his maximum falling speed. More on that later. “Engage” here means the time between inputting an attack and when sora can strike a distant enemy from a neutral position.

So these results are pretty much in line with what I expected going into this project. KH2 is the fastest game, KH1 is second fastest. BBS and DDD are dead-last, KH3 is about in the middle, while KH3 with its post-launch patch just a little bit faster than that.

Some surprises:

Every KH game seems to have the exact same startup time for the first most basic attack of a combo. Seems like they’ve been pretty happy with this one from the word ‘go’ and consistently on from there.

KH1’s movement lag from a grounded attack is insanely long, as long as DDD’s. KH2’s is way faster, and nearly the same as KH3’s.

KH1, KH2, and KH3 all cancel their first attack into an attack combo incredibly fast at just four frames, another astonishing level of consistency, although BBS and DDD combos take twice as long or more to come out.

KH2’s biggest advantages in speed are its gravity hang time, and time to engage enemies. KH3’s gravity hang time actually got worse with its post-launch patch, though not noticeably, and this has to do with a new attack added in the patch that has a longer animation in general.

BBS and DDD don’t do too badly in the enemy engagement department, better than KH1 and KH3 pre-patch, even. KH2 is still the champ here, but this fast engage time might be a saving grace for BBS and DDD.

There is a lot to talk about, so I’ll drop the chart here again real quick, just because I know you all wanted to look at it again.

A 2 axis chart. A list of parameter such as "1st ground attack startup time" make up the Y axis. The different Kingdom Hearts games being tested make up the X axis. KH2 seems to be the fastest game judging by data, with KH1 shortly behind it. KH3 is in the middle range, behind KH1. BBS and DDD are nearly tied for slowest game.
Even more beautiful.. and horrible the second time!

Okay, So What Is It?

What’s what? Oh! Floatiness. What is ‘floatiness’? My hypothesis going into this was that it’s tied up with the aforementioned startup and ending lag statistics, which would have a huge impact on the speed of the game. Kingdom Hearts: Birth By Sleep, Kingdom Hearts: Dream Drop Distance, and to a lesser extent Kingdom Hearts 3 are somewhat known among hardcore fans as the ‘floatier’ games, the former two by far the most guilty. I had presumed that in gathering my data, it would bear out that those two games would have the higher numbers when recording the delays, lag, and hang time. Kingdom Hearts 2 would be the fastest game, only slightly faster than Kingdom Hearts, and Kingdom Hearts 3 would fall somewhere in the middle. Indeed this was the result. So. Case closed? Shorter periods of loss of character control equals better combat? Probably a little more nuanced than that.

Golden calf Kingdom Hearts 2 absolutely has the advantage over its successors and predecessor when it comes to attack commitment. Its startup times, ending lag, and combo speed are universally faster across the board. But… they’re not that different from KH1 and KH3, even pre-patch. Almost immediately when I started this experiment, replaying each of the games side-by-side I came to realize the nuances of how gravity works in each game.

The Gravity Situation

As the Kingdom Hearts series progressed, aerial combat became more prominent. I always imagined in my head that the first Kingdom Hearts game was “the more grounded one”. And it kind of is. In the first game, attacking launches Sora high in the air toward aerial targets, and repeated attacks can keep him airborn somewhat. Kingdom Hearts 2 pushed this much further, with far more automated and accurate tracking on enemies, pushing Sora high into the air with each attack. I was kind of shocked to find that in Kingdom Hearts 2, the game I associate most with air combat, gravity is much more powerful than it is in any other game. With a delay of only 16 frames in KH2, compared to KH1’s 20, and double that at 32 frames in BBS, KH2 by far is the most beholden to gravity when the player is not attacking. This makes gravity a function of the player’s inaction. Never are they airborne without making it a conscious decision. In BBS and DDD, being strung up in the air against your will is almost the norm. There is a huge delay between aerial attacking and being affected by gravity.

Sora from Kingdom Hearts 1 jumps into the air and slashes at the three headed Cerberus from Disney's Hercules, before quickly falling back to the ground and dodging out of the way of a shadowy breath attack.
Gravity
Ventus from Kingdom Hearts Birth by Sleep (who resembles Sora) swings at an airborn enemy shaped like a pot. Several attacks miss, and Ventus is left hanging in the air for a moment before falling, after each attack.
Gravity?
Notice how in the first image Sora descends to the ground *before* his attack even ends

I think this is a major piece of the floatiness puzzle too. Being on the ground in these games naturally gives you a lot more options. You have access to your dodge roll. You are able to more precisely place yourself for things like blocking, or jumping out of the way, or lining up the right attack. Simply, being in the air is advantageous in some ways, but also a lot more risky. In the less ‘floaty’ games it feels as though being airborne is always a conscious choice, whereas the others place you there in spite of the player, and leave Sora hanging in a risky position for far longer than a player might intend. The ability to get back to the ground quickly gives finer control over the player’s risk vs. reward, and the game forms a more solid and consistent relationship with that risk vs. reward.

Gamefeel Good

The title of this article mentions gamefeel, so I’m going to take a moment to discuss that. Like ‘floatiness’ this term is a little hard to pin down, so, uh, how about wikipedia. They seem to know everything.

“Game feel (sometimes referred to as “game juice“) is the intangible, tactile sensation experienced when interacting with video games. The term was popularized by the book Game Feel: A Game Designer’s Guide to Virtual Sensation[1] written by Steve Swink. The term has no formal definition”

Wikipedia Article on Game Feel

Well that was absolutely no help whatsoever. Wikipedia cites input, response, context, aesthetic, metaphor, and rules as the features a game can change to influence gamefeel, and this seemingly comes from Mr. Swink as well. It’s as good a place as any to start, so let’s see how this applies to our floatiness dilemma.

I don’t want to get too into the weeds with these, so I’ll limit my observations to response, metaphor, and rules. Suffice to say I think KH does pretty universally well in the input and aesthetics department. Context is a bit more nuanced, but again, trying to limit our scope here.

Let’s be a little contrarian and try to make at least a very small part of this gamefeel equation tangible. We’ve already broken down that there seems to be a correlation between the delay between action and reaction in Kingdom Hearts, and the popular perception of how well the games feel to play. KH1 and KH2 are beloved with their snappy reaction times, KH3 is well liked but garnered a lot of fan feedback. BBS and DDD are liked but not necessarily for the fluidity of their combat, and I rarely see gamefeel cited among their strengths. The data easily illustrates the importance of response time as a variable.

The metaphor at play here is how the timing of attacks translates to what the game is trying to convey. The main thing the mechanics of KH is trying to convey is melee combat. All these varying measures of time and attack resolution and ending lag etc. is all metaphor for the physicality of swinging a sword at a guy. A real person cannot instantaneously move their arms, the sword has startup time and ending lag in real life. That’s why these time gaps are there in the first place. It wouldn’t be a very fun game if you pressed the attack button and just instantly won. So as in a real, actual battle, you have to consider how your various actions leave you open to counterattack. On the other hand, one can go overboard with this, as in real life you’re also trying to minimize your vulnerabilities, and a swordsman is not going to be flashy at the cost of leaving himself open to a hail of enemy attacks. These are things people intuitively understand, and so players bring with them preconceived notions of how entities should behave in these situations.

Sora from Kingdom Hearts 1 blocks an attack from a dark ball with a mouth. Immediately, he counters with a series of fast strikes in the air. He loses altitude with each strikes, and ends his combo standing on the ground.
Pictured: Gamefeel??

Drill even further down and you get to the rules of the game itself. Enemies are attacking you, while you’re trying to attack them. If you are locked into an animation you are unable to deploy any defense to stop them, and so we have a game. Risk, and reward. However, if the player’s control feels disconnected from how these rules operate gamefeel is thrown off. In Birth By Sleep and Dream Drop Distance, attacking creates great gaps of no-player time, where inputs on your controller mean nothing. You are vulnerable to attack, and a passive observer in the world. This disconnect throws off both the player’s relationship to response and to the rules of the game, as overly long and committed attacks can become more of a liability than an asset. In KH1 and KH2 throwing out attacks is almost always fun and satisfying, and usually rewarding. It’s often a lot more burdensome to do so BBS and DDD. BBS and DDD have other methods of dealing damage besides the basic attack, but since the basic attack is so pivotal in the main series, a lot of players are thrown off by how minimized it is in these spin off titles, which primarily utilize the command deck instead.

The Command Deck

So why is Birth by Sleep and Dream Drop Distance so damn slow compared to their peers? Well these two games share two very important elements in common. First, they were each developed for a handheld family of systems. Second, they share the game design system known as the command deck. These two points of similarity are very much entwined.

Kingdom Hearts and its two numbered sequels utilized the ‘command menu’, a little persistent menu in the corner of the screen that can be navigated in real time to select whether you want Sora to attack, use an item, or cast a magic spell, essentially. 90% of the time you will be spamming the attack button. The command menu was meant to evoke the strategic decision making of a turn-based Final Fantasy game, but re-tuned for an action combat context. Ultimately contextual combo modifiers became a much greater source of player expression in gameplay, and the strategic applications of the command menu are limited. However, it remains an elegant way to give the player a lot of options without bogging the game down with too many buttons.

When Kingdom Hearts went handheld it rethought its combat. Suddenly, we’re dealing with a much smaller screen, and the camera had to be pulled in on the player character to compensate. As a result, enemies can much more easily hit you from off-screen or sneak up behind you. Consideration was clearly given to a system that better fit playing on the go, where deep concentration on twitch-reaction may not be as viable or palatable. The command deck was the answer, a cooldown-based system where actions are selected from a rotating list and can be used at any time. Now the flow of combat is dictated by these ability cooldowns, and the regular attack combo was made to be more slow-paced, flashy, less freeform, and yes, more floaty so as act more like a stopgap to using the command deck, rather than the main form of attack itself. The command deck has a lot of advantages such as the ability to customize your attack loadout in a much more granular way, but it also has its own host of problems, which I won’t get into here.

Ventus from KH BBS slashes at green impish enemies on the ground. In the bottom left corner, the command deck can be seen.
Sora from Kingdom Hearts Dream Drop Distance slashes at the air while standing on the ground in a mailroom, then does in the same after jumping. The command deck can be seen in the bottom-left.
In each game the command deck is present in the bottom-left

The point is, Birth by Sleep and Dream Drop Distance are deliberately slower games made so because they were designed to be played on the go, and utilize entirely new systems that weren’t designed with KH1 or KH2’s speed and flow in mind. Because KH3, BBS, and DDD share a lot of development talent though, it seems like a lot of the design philosophy of these handheld games bled over into KH3’s headliner console game, which was jarring for a lot of people.

Okay So… Is This Bad?

In my opinion, one of Kingdom Hearts‘ greatest strengths is the immediacy and responsiveness of its gameplay. A fine degree of control gives the player a much wider spectrum of possibilities in any given scenario, and this is what leads to a creative space within gameplay for players to express themselves through the gameplay. This fluidity also allows disparate gameplay mechanics such as movement and sword attacks to blend. Kingdom Hearts 2 always feels as though its many, many mechanics are in a harmonious conversation with one another. Engaging a ground enemy can be converted into an aerial assault, which can flow into a strategic repositioning, etc.

So yes, I think it weakens Kingdom Hearts‘ identity somewhat when some of the more experimental entries in the series have dabbled with slowing down the combat and making its pieces more discrete. The particulars are a bit beyond this already enormous article, but I did want you to understand my motivations for trying to understand how and why Kingdom Hearts 2 feels so different to play than a game like Kingdom Hearts: Birth By Sleep.

Sora from Kingdom Hearts 2 slashes at a volleyball, sending it bouncing into the air.
Up… and Down

Kingdom Hearts is a series about ridiculous anime battles between magical wizards wielding baseball bat-sized keys as weapons. People fling themselves through the air to attack one another. It’s not realistic, but it exists within the heightened reality realm of animation. Its primary inspirations are, after all, some of the animation greats, including Disney of all things. Animation works to heighten reality because it has an understanding of its underpinnings. Even if people can jump twenty feet in the air, there still has to be a since of presence and weight. It has to give the impression that some sort of underlying laws of physics are at play even if they aren’t one-to-one to our own laws of physics.

Despite appearances, it wasn’t a magic genie spell that made KH2 so satisfying to play. It was hard work, clever planning, likely some very precise number tuning, and likely a whole lot of playtesting. I want to know the nitty gritty of what made Kingdom Hearts one of the smoothest action games in existence, what distanced it from that, and what brought it closer to that esteem once again. Because you see, there are still those Kingdom Hearts 3: Remind post-launch changes to talk about.

Kingdom Hearts III And The Combo Modifier

In Kingdom Hearts, Sora has a very basic attack combo, with little by way of alternate options for the player, at first blush. You press the attack button, and sorry does the next hit of the combo. Rinse, repeat. The wrinkle though, is that Sora can gain abilities which cause his attack combo to behave differently depending on where Sora is standing relative to his target.

Kingdom Hearts 2 pushed this contextual combo modifier idea much further, with hosts of new options that allowed the player to flex mastery over the system through mastery of manipulating these contextual attacks. They did things like make engaging enemies faster, extending the combo, increasing the power and range of combo finishers, among other things.

Kingdom Hearts 3 launched with its share of combo modifiers, but they lacked the breadth and applicability of KH2’s abilities, and they did little to speed up combat overall. That is, until it was patched in anticipation of the Kingdom Hearts 3: Remind DLC. Numerous combo modifiers were added which increased the overall speed of Sora’s basic combo attack and made the game feel overall more responsive. I don’t think the Remind DLC gets enough credit for just how much it improved, and it’s kind of refreshing how in tune it is with popular feedback. Real people playing your game can tell you a lot about that mystifying, all-important yet elusive gamefeel.

Sora from Kingdom Hearts 3 showcases an attack combo with a flashy series of slashes accentuated with powerful light energy, these attacks come out rather rapidly.
Combo Modifiers
Sora from Kingdom Hearts 3 showcases an attack combo with a series of slashes. These attacks come out noticeably slower than the other image in this gallery.
No Combo Modifiers

Airstep is Literally a Game-Changer

Kingdom Hearts as a series has done a decent job with teaching players about its systems and how they work. Some of the more advanced mechanics that deepen higher-level play are a bit more obscure, but usually reaching a baseline level of competency at core systems is pretty straightforward. Kingdom Hearts 3, however, I found stumbled at instilling the importance of oce of its most important new mechanics, the airstep. The airstep allows the player to manually aim the camera toward a distant foe, and fling Sora at them at high speed for a relatively quick engage across a huge battlefield. This was obviously done to compensate for the much larger environments featured in KH3 as compared to older games, but utilizing this move frequently and often, even in close quarters, really does change the way the game is played.

Airstep can cancel out of nearly any other action, you see. In terms of making the game more fluid, or moreover in terms of giving the player fine control over their actions, the airstep affords an enormous amount of power and precision to determine the pace of fights in Kingdom Hearts 3. This along with the ability to cancel combo finisher attacks, which was added in the Remind DLC update, makes the whole combat a lot more cohesive and much more fluid than you might initially expect based on the raw frame data. Speaking of which, the frame data was improved in that update as well! The fact is that this latest iteration evolved the combat in a lot of interesting ways that I don’t want to see abandoned in favor of just making the series going forward exactly like everything that came before.

Sora from Kingdom Hearts 3 blocks an attack, before the camera zooms in on his enemy, Sora then zooms to their location in a flash of light.
Here Sora is sent at the boss straight out of a guard, exploiting an opening in an exciting way

Kingdom Hearts Is Doing Fine, Actually (But It Can Do Better)

So yeah, when people say Kingdom Hearts 3 feels ‘floaty’ as compared to Kingdom Hearts and Kingdom Hearts 2 the evidence is there. Attacks are less responsive and more of a commitment, generally, and gravity is turned off more liberally. You will literally find Sora floating more without direct player intervention. However, the frame data differences, discounting the gravity aspect, are minor at best. It’s really that gravity delay that accentuates the small differences to give a pretty stark impression. Birth by Sleep and Dream Drop Distance are so far afield of the gamefeel of the main series games that it’s clear there was no intention of recreating those mobile games for KH3’s release. The developers have further made it clear their intention to deliver an experience their longtime fans can enjoy with the changes made to the Remind DLC.

In fairness, it had been 13 years since a main Kingdom Hearts game had been made when KH3 was released in 2019. With just a few tweaks I feel they were able to bring it up to parity with KH and KH2 in terms of fluidity. It’s not the tightly wound, exceptionally blisteringly fast experience of KH2 precisely, but do remember that KH2 was developed only three years after the original, so the developers were coming right off the back of designing an already exceptional combat system. The experience was fresh. From where I stand I don’t think the future of KH is in any danger. Three years on from KH3 as of the time of writing this in 2022, and KH4 seems to be around the corner, poised to improve in ways reminiscent of how KH2 improved.

The Kingdom Hearts 4 reveal trailer doesn’t actually show any basic attacking in favor of mostly displaying the new movement options, and any gameplay it does showcase is subject to a lot of change to begin with, but what it does show looks pretty snappy and responsive to me. I’ll let you be the judge. The KH team learns quickly, and their intentions for this series’ gameplay have been made clear vis a vis the Remind DLC update. They understand the data I’ve so painstakingly gone to lengths to describe and how it affects gameplay, clearly. The KH3 update was so precision built for improvement it’s kind of astonishing. And, with KH4’s refocus on ‘reality’ as an aesthetic, I would not be surprised if a little more gravity came back to the party to polish the combat system even further into something really special.

Sora from Kingdom Hearts stands on a floating platform with a shield-wielding enemy, and rapidly dashes at and through them, holding out his keyblade.

Whatever you’re talking about, I don’t care…

Sonic vs. Shadow Boss Breakdown: A Case Study

It’s April 2022, I just came back from the movies, and it’s got me in a Sonic mood. Sonic Adventure 2 was Sonic’s second major 3D outing and personal favorite of mine, warts and all. The big gimmick of this game was ‘hero versus dark’, or in other words two concurrent stories centering on our familiar heroic blue hedgehog Sonic, and the new dastardly red and black hedgehog Shadow, who has all the same powers that Sonic has!

The Sonic and Shadow showdown that’s built up across the game’s story was always a highlight for me as a child. It’s aesthetically and narratively well-realized, a great moment of tension and catharsis for one invested in the wacky world of Sonic. So it works on a number of levels. Taking an older and more analytical eye to it though, it is far from an ideal example of boss design in motion. Like much of Sonic Adventure 2, it has its share of flaws. So for this boss breakdown, I’m going to approach things a little differently, and go through an alternating list of the boss design’s pros and cons, trying to glean where this boss succeeds and where it falls short.

Sonic the Hedgehog stands in a long starship-like hallway. He is surprised when Shadow the Hedgehog walks up from behind him and begins to speak "You never cease to surprise me blue hedgehog. I thought that the capsule you were in exploded in space." Sonic replies, "You know, what can I say... I die hard!"
The motion-captured cut scene animations are still pretty comically bad, though.

First, a quick overview. The player takes the role of Sonic or Shadow, depending on which story path they chose. Their opponent is whatever hedgehog they’re opposed to. In either case, the fight is identical, as both characters share identical movesets as players and as AI bosses. The two hedgehogs run along an infinitely long looping pathway suspended in outer space. Pieces of it fall out from under them if they run too slowly. If the enemy boss hedgehog falls off the path, they teleport back to the stage ahead of the player’s position. If the player does the same, they fall to their death. The enemy will occasionally use a homing attack on the player, which can be deflected by jumping, which produces a protective spin-shield on the hedgehog. The enemy can do the same, however, meaning most homing attacks will be deflected unless the player can make an opening in the enemy’s defenses.

These openings, as far as I can tell, take three forms. First, if the player lags behind the enemy by too great a distance, the enemy will come to a halt, and launch a screen-wide super attack, which can be avoided with a well timed jump. During this moment, the enemy is vulnerable to attack. Second, if the player runs too far ahead, or the enemy falls off of the arena, they teleport back onto the road ahead of the player, after which they are vulnerable for a brief moment, although this window of vulnerable seems to narrow as the boss’s health gets lower. Finally, timing an attack such that it hits the boss in the short window of time they are on the ground after landing from a jump, they are vulnerable.

With that out of the way, let’s get to some of the particulars.

Pro: We Framed The Camera

So Sonic Adventure 2 (and a lot of 3D Sonic games of this era for that matter) has a bit of a reputation with its camera. So this one might come off as somewhat of a backhanded compliment, haha. To be fair, I think the direness of the state of SA2‘s camera is often vastly overstated, when compared to other 3D cameras of the time, or even other Sonic games. Point is, sharp turns and elaborate level geometry could be let’s say… challenging for the hard working early 2000’s 3D camera, so Sonic Team played to their tech’s limitations and its strengths and placed this boss fight on a seamless, infinitely long straightaway.

With no turns, nor obstructions, the player always has a clear view of the action and the goal- defeating that enemy hedgehog. This has the added bonus of creating a series of dynamic set pieces visually between Sonic and Shadow’s fight, with the camera being centered on the highway path.

Sonic the Hedgehog outruns Shadow. A moment later, Shadow teleports in front of Sonic in a flash of green light.
Nothing but you, me, and the cinematic cameraman following behind us

Con: Rings, Rings Everywhere

Someone once did a write up on Sonic The Hedgehog’s traditional health system that utilizes its collectible coins. This system has a lot of flaws, in my eyes. The simple version is, that because having even one ring on Sonic has nearly as much advantage as having a billion rings, while having zero rings leaves Sonic in a drastically more vulnerable state, it is very difficult to build a consistent level of challenge with this health system. The battle between Sonic and Shadow takes place on an infinitely long looping pathway. It would be very possible, even likely, that most players upon taking a hit from Shadow or Sonic simply kept running, missed picking up their rings, and were left with a crushingly difficult unreasonable encounter.

This is obviously not desirable in a game marketed to children, and so to counter this eventuality, it was decided that the infinite path contain infinite rings. This is also a problem. So long as you keep collecting these readily available, plentiful rings, it is essentially impossible to lose this boss fight that is an alleged death battle between super powered titans. It doesn’t matter how many times Shadow strikes Sonic with a chaos spear, or how often Sonic homing attacks Shadow. If the player can collect more rings, and they can, they’re in no danger. It kind of takes the wind out of the tension-sails when you realize this.

Pro: The Game Does Not Tell You What To Do

There’s a certain appeal to to the idea of just being placed in an arena with a hostile opponent and… going at it. No rules, no holds barred, no ifs ands or buts. Of course, there are rules, such boss designs simply give the impression that the player can control the pacing of the fight, and can achieve progress against the opponent with mechanisms entirely driven by their own ability, not by arbitrary timers or enemy behavior tables. Someone once did a write up on the subject using The Legend of Zelda as an example.

When fighting Sonic or Shadow it is not initially entirely clear how one is meant to deal damage to them. They have a health bar indicating damaging them is the goal, but not much else. There’s very little indication of the boss’s weakness. The player’s standard methods of attack up to this point, homing attacks or the spin dash, have no effect in most situations. For a final confrontation, this makes sense. Shadow and/or Sonic stand as an imposing ultimate challenge with no clear exploitable point. It’s up to the player to decide how best to approach the enemy supersonic hedgehog, as their opponent has all manner of ways to deflect incoming attacks, such as rolling into a ball, homing-attack the player in retaliation, or using one of their screen-wide super attacks. This bare-knuckle brawl kind of boss design works well in establishing the tone of two co-equal rivals finally duking it out and giving it their all. If Shadow or Sonic had a big glowing ‘attack this to kill me’ button somewhere it would certainly diminish the effect.

It’s also worth mentioning, this is the second such encounter between Sonic and Shadow. The first, in an enclosed square arena, is a lot less exciting to play and talk about. I bring it up because it does begin to engender some of the concepts of how to fight the enemy hedgehog that you see here, but that fight also suffers from not being entirely clear on its mechanics, so the player will likely not have learned a definitive approach to fighting Sonic or Shadow by this point.

Sonic the Hedgehog repeatedly homes in on Shadow the hedgehog, but Shadow rolls up in midair to protect himself, deflecting the attacks.
Standard methods of attack don’t work. Figure out a way past his defenses!

Con: The Game Does Not Tell You What The Heck To Do

Yeah okay, so there are advantages to unstructured boss fights, but Shadow and Sonic’s weakness does exist, and it is largely arbitrary. The game gives little to no indication at all how best to damage the boss, and as a result, even for players who’ve defeated this boss before, it may not be entirely clear how they did it, and reproducing those results may require a lot of fiddling around with the boss’s behaviors. There is a method to consistently damaging Shadow or Sonic in this encounter, but to get there you really have to do some meandering experimentation. The investigative nature of this isn’t without its fun but it does feel kind of weird and incongruous, like a puzzle whose pieces fit together, but do not form a picture of anything.

Why is it, for example, that Shadow will only use his super move ‘chaos spear’ when Sonic has let him run far ahead down the path? Why does he stop dead in his tracks when chaos spear is used, leaving him vulnerable? This might have made a lot more sense if Shadow had a special animation to indicate he had to slow down to concentrate when using chaos spear, which would also explain his vulnerability and why he’d wait until he got some distance from Sonic before using it, but the player is left to essentially fill in these blanks themselves. Given the loose, fiddly nature of the boss’s behaviors, it’s also very possible to simply mash buttons and attacks against them until the encounter is won. Combined with the ring systems, there’s really no way to enforce a failure state to dissuade this, which in my eyes is an issue. The player should feel like they accomplished something even it’s an easy, so the possibility of a largely accidental victory is undesirable.

Sonic the Hedgehog and Shadow the Hedgehog repeatedly roll up into balls and bounce against each other, until Sonic lands on shadow's head and he dies
I can barely tell what happened here, and this is my gameplay footage

Pro: Forward Momentum

Sonic’s the name and, you may have heard, speed’s his game. If that’s the case it’s kind of wild that a ton of Sonic boss battles have nothing to do with the aforementioned. The premise of Shadow the hedgehog as a character is that he’s a dark mirror to Sonic the hedgehog, capable of all the same feats of speed and agility. Since that’s the big selling point here, we want our boss fight to illustrate that, thus the long straightaway.

Infinitely looping paths along which to run whilst fighting a boss would eventually become a halmark of the Sonic the Hedgehog franchise. It was actually kind of a rarity at the time of Sonic Adventure 2, though. The advantages are obvious. With an unobstructed pathway Sonic is free to run at full blast, the most fun thing to do in a Sonic game, while Shadow matches the speed, and the spectacle of a high-speed superpowered battled is nearly realized right there. Bits of the looping pathway collapse, leisurely, as they are cycled off-screen, gently encouraging the player to not stop running, keeping the pace of the fight breezy. It all works very well to construct the scenario they were going for.

Sonic Adventure 2 also iterated on the lightspeed dash, in which Sonic can quickly move across a path of rings with a well-timed button press, which gives an excellent and rewarding avenue for expressing player skill. Stringing together dashes along the many paths of rings is a blast.

Sonic zips across a series of paths made of gold rings at incredible speeds.
God I love this game

Con: The Enemy Is Not Proactive

I did praise this boss for being rather player-driven, allowing the player to decide the pace of the fight. However, the fact that the boss will do very little to actually attack you is a major problem for me. Shadow and Sonic are in a dire situation involving the fate of the earth and an orbital laser beam capable of blowing up half the moon. Things need to be tense, and the danger needs to run high. The hedgehog being controlled by the AI, though, just doesn’t feel especially desperate to kill the player. They will happily run alongside you, but never really do much to harm you unless you position yourself in a specifically vulnerable spot. The enemy homing attack is especially easy to deflect. If you run too far behind they’ll do that super attack I mentioned earlier. I feel as though a lot more could’ve been done to establish that the boss is very motivated to defeat you, and I would’ve included some more regular attacks and obstacles that the player has to maneuver to keep up. A lot of what makes running fast in Sonic, after all, is the satisfaction of skillful execution in avoiding hazards.

Pro: Timing and Execution

As unclear as the ‘rules’ of the boss’s behavior are, they do exist, and executing on them can be a lot of fun once you understand them. One of my favorite strategies for damaging the boss is to hang back, causing them to use their screen-wide super, then rush ahead and use the ring-dash to dash straight through the vulnerable enemy. These interactions are actually very compelling, requiring a reasonable bit of skill and control to execute on. It’s actually so fun to land hits on the enemy hedgehog once you understand what is actually going on in their behavior table, that it makes it even more frustrating that a lot of this is not clearly communicated.

Shadow the hedgehog stops running for a moment to shoot orange energy projectiles at Sonic, but Sonic dodges them and homes in on Shadow to damage him.
Baiting the enemy into an attack and retaliating does feel pretty good.

Pro Again: Aesthetics

Alright this article has been pretty critical of my blood brother Sonic Adventure 2, so I’ll end with one last pro that’s less to do with the gameplay design, but still an essential feature. If you didn’t know, aesthetics deal with all the principles of beauty, not just visuals. It’s an overall feel. A style, if you will. And man, does this game have style coming out of its comically large cartoon hedgehog ears. Shadow and Sonic are under-lit with an eerie green light that accentuates the alien environment and makes Shadow look positively menacing. The techno-electronic music blares with strange bell samples with some of the edgiest lyrics you’ve ever heard in your life. You can see the earth frames against the black sky. Meteorites and pieces of the space colony zip by as you run across this collapsing sci-fi highway. The fight has a lot to be desired in terms of gameplay depth, once you really deconstruct it, but it is an absolute juggernaut of presentation, for the year it debuted. Sonic has always done spectacle well, and the promise made by this game’s marketing: an epic showdown between an unstoppable force and an immovable object in the form of two equal superpowers of opposing worldviews, is fully realized through the aesthetics. Aesthetics can elevate an experience, and Sonic Adventure 2 leans into this hard.

Sonic the Hedgehog runs alongside the menacing Shadow the Hedgehog as "Shadow" is spelled across his image. The two speed down a scifi highway amidst meteorites and space colony towers against the black of space.
This is one of the coolest things ever produced and I will entertain no dissent on the subject

It’s been a lot of fun reliving this fight, one of the longest-held in my memories, with the benefit of many years of design knowledge. The ‘rival’ fights in Sonic Adventure 2, in which the boss is a player character from the opposite story path, were heavily limited in what they could do, as they exclusively utilize the player character assets, animations, movement systems, etc. That said, it’s clear that this climactic battle was trying really hard to do a lot with a little. Not everything about it is stellar, but I think it’s still commendable they managed to make a pretty fun boss encounter that at least heavily delivers on the narrative promises of the game, if nothing else. I still find so much value in looking back at design of games like Sonic Adventure 2 that shaped my childhood, even with all of their flaws.

Sonic The Hedgehog celebrates over the unconscious defeated body of Shadow the hedgehog, standing on a sci-fi skybridge in the black of space, surrounded by inverted skyscraper-like structures.

That blue hedgehog again, of all places…

Difficulty as Narrative Design – The Emotions That Might Happen When You Fight Sans

I’ve had some stuff on the brain lately, in regards to difficulty’s place in design, which is what tends to happen when you play Elden Ring for so many hours straight. I’ve also been replaying Toby Fox’s Deltarune with a friend, another game that uses difficulty in interesting ways. I’ve had this thought for a while, to do a write up about how difficulty can be, and is, deployed in design to affect the greater experience. This article contains major spoilers for Undertale and mild spoilers for FromSoftware’s Elden Ring.

To be unambiguous here – difficulty is a very nuanced and at times personal subject in design that touches on a host of other things such as game balance, technical depth, general play enjoyment, and of course accessibility. These are very complex subjects that deserve their own discussions. What I’m specifically focusing on in this article is how difficulty can be deployed with purpose, and often has more relevance to the overall design than is often attributed to it, as a simple measure of player competence for the purposes of challenge. I wanted to look at an example of a game where difficulty is an intimate part of its narrative design, where the reactions it illicit is very much a product of how difficulty is utilized.

The idea that difficulty in gameplay can be a narrative tool should be fairly straightforward to grasp when looking at a couple of examples. In Elden Ring, all of your primary boss characters are demigods, children of gods, who once fought over the shards of the titular ring. The demigod Radahn fought his half-sister Malenia to a standstill. Radahn is oft touted as the strongest of all demigods – he holds the stars in stasis by his own power – he takes an entire platoon of elite soldiers in gameplay just to take down! This assertion that Radahn is the strongest remains more or less unchallenged for some time. There are harder bosses, but none that require so much backup to defeat, nor any nearly as hobbled with injury as poor Radahn.

There is a secret and hidden boss, however, another demigod called Malenia, who is still alive. When Radahn is found, Malenia’s power, the same power that has scarred the landscape around Radahn has left him ‘divested of his wits’, and fighting like a wild animal. Malenia, however, is more or less totally lucid, angrily awaiting the return of her missing twin brother and liege lord, Miquella. Malenia has never been in better form – there was nothing stopping her from taking Radahn’s shard of the Elden Ring and yet she did not, so clearly she has no interest in ruling. Indeed her dialogue reinforces the notion that she fought only for loyalty to her brother’s ambitions.

Two warrior women face each other in a lush cavern filled with white flowers. One has red hair and is in golden Valkyrie garb, with a sword. One is in a blue hood, with a spear. The Valkyrie ascends into the air and swings her sword with such ferocity it creates white-hot slashes of air in a blurring flurry around her. The blood-hooded woman runs and rolls around the attacks.
I can practically feel the hairs being shaved off the back of my neck.

Any who’ve fought Malenia will tell you, the idea that Radahn could stand a chance against Malenia in combat, is laughable. They could tell you entirely because of how demanding of a boss she is, how difficult she is tells you the entire story. There’s no possible way she left her encounter with Radahn in defeat, or even in a draw. Her swordsmanship is deadly and near insurmountable, and she hides an even greater power beyond that. She defeated him, and he was left without his senses. She must have left because her brother, the real aspirant to the Elden Ring, went missing. The player will know this intuitively, through experience. They lived it. They will feel it in their bones. Radahn could not have defeated Malenia, and the rest of the story follows. Without Miquella, there would be no reason to collect Radahn’s shard. If you’ve explored the world of Elden Ring thoroughly, this line of thinking is vindicated, as you’ll know Miquella underwent a sudden and shocking disappearance, followed by an extended and secretive absence.

A woman in a blue hood runs her spear through the chest of a taller woman with red hair and golden Valkyrie garb, the stabbed Valkyrie falls onto her back in a pool of blood as the spear is removed.
Difficulty is a marker of power in games, and examining power is essential in stories of conflict

If you’ll indulge me to invoke the first of two quotes from Bennett Foddy, designer of Getting Over It With Bennett Foddy, a notoriously difficult game.

“The act of climbing, in the digital world or in real life, has certain essential properties that give the game its flavor. No amount of forward progress is guaranteed; some cliffs are too sheer or too slippery. And the player is constantly, unremittingly in danger of falling and losing everything.” – Bennett Foddy

All that said, difficulty is not just a mechanical gameplay consideration. Like all aspects of a game, it is an essential part of the cumulative experience. I am of the opinion that if an obstacle within your narrative is meant to be threatening, formidable, out to kill our dear player character, then the player should get the sense that this force is threatening and formidable. To trivialize it, or deny the sensation that there is an opposing force trying to halt the player’s forward motion, is to render the narrative dishonest, and rob it of its power. If conflict is about power, than difficulty is one of the most genuine ways power can be communicated in an interactive system. This isn’t to say that every game needs extreme challenge, or even that every game with conflict is necessarily trying to create the aforementioned sense of opposing force. This is but one type of experience you might seek to create, a goal your art may aspire to. In fact, this is just one way to deploy difficulty as a mode of narrative design. That brings us to Undertale.

The skeleton Sans stands in a black void, above a battle UI overlay. He says "ready?" then suddenly unleashes a barrage of bones and laser beams to attack the player, represented by a red heart-shaped cursor.
“No.”

In Undertale, the story is persistent – and any runs of the game, even when reset, are remembered and color the experience of playing Undertale going forward in little ways. Death and resetting is diegetic, meaning the player character is literally dying and coming back to life at a previous point in time, within the game’s fiction. In this way failure is kind of inherently tied to the narrative. Undertale comes packaged with a few predefined paths to play that present themselves based on how the player tackles obstacles. Killing monsters casually as they come to confront you will result in one of several ‘neutral’ endings, in which the player’s human character escapes the world of monsters, which is left in varying states of disarray as a result. The ‘pacifist’ run will see the player avoiding lethal violence, and reaching out the hand of friendship to major characters to achieve the best world for everyone. The ‘no mercy’ run is the third and most obscure path, in which not only is lethal force deployed against all obstacles, lethal force is deployed against every potential obstacle, wiping out all monsters in the underground.

To do this, the player has to spend an inordinate amount of time trawling around for enemies to fight. Every single one needs to be killed for the No Mercy ending to hold true. This process is long, repetitive, somewhat dull, and even grueling at times. And yet, it remains an immensely popular way to play this already immensely popular game. There is a purpose to all this consternation, though. I think it pretty noncontroversial to say Undertale‘s ultimate message is one of nonviolence – that the best way to solve conflict is through open communication and a curious, empathic heart. The No Mercy run exists as a counterpoint to this message, to prove its efficacy. Killing everything in Undertale is a pain, frankly. It takes a lot of effort but not necessarily the kind of effort a player seeking challenge might be after. More of that exists along the less violent story routes. No, Undertale is instilling through the avenue of frustration that ‘the easy way out’ isn’t always easy, and while ‘the high road’ isn’t always easy either, it’s a heck of a lot more fun than willful cruelty, which is a continuous and conscious effort on the part of the abuser.

And yet, most playing through will persist. They have buy in, and as Undertale expects, most will be curious enough to want to know what happens next, not in spite of the frustration, but perhaps even because of it. One of Undertale‘s most infamous features is the normally comedic, friendly, and jovial character Sans, who is a bit of an internet meme. There’s a lot of reasons for that, but I think one of them has to be his sudden transformation into the game’s greatest and most stubborn challenge. The boss battle against Sans, with one other exception, is the only real challenge in the No Mercy run, with all other opposition crumpling like paper before the player. The player has not had a ramp up in difficulty in this point, and Sans comes out of the gate swinging with one of the most demanding gameplay experiences in modern popular interactive media. No punches pulled here, Sans is meant to be a brick wall of a boss, one that will have to be worn down with patience if it’s to be cleared at all.

Sans the skeleton stares you down from a black void with a battle UI overlay, as the player's heart-cursor, now blue, jumps across platforms littered with bones.
I have SO MUCH patience right now, you wouldn’t believe. Oh god the patience I have.

Fans of Undertale have such a personal relationship with it, and given its immense popularity that is quite impressive. The player at this point, is acting as an agent within the narrative, separate and apart from their controlled character, Frisk, who ambiguously is either mind-controlled by the player, or influenced by the player subtly to act or fight. Sans tries everything he can to appeal to the player to start over, to do anything but follow through on the path they’re on. He pleads, he appeals to humanity, he threatens, and he even cheats. After each failure, Sans comes up with some new unique dialogue with which to taunt and belittle you for trying. The player can come back as many times as they want to try again, so words and his ability to act as an immovable object are Sans’s only real forms of power over you. The ironclad stubbornness of this encounter, the unerring, unflinching confidence in its unreasonableness makes it feel real, like Sans is a thinking actor specifically trying to get under your skin, and make your goal unreachable, and that is what makes it feel personal.

Sans the skeleton says, "sounds strange, but before all this i was secretly hoping we could be friends. i always thought the anomaly was doing this cause they were unhappy. and when they got what they wanted, they would stop." He then fires a bevy of skull-shaped laser cannons at the player's red cursor.
Oh. Kind of makes me feel bad I’m trying so hard to kil- OH GOD LASERS

Sans isn’t trying to kill you – he knows that is beyond his power. He’s trying to wear you down, to frustrate you, to bore you, whatever it takes to make you give up on your killing spree, and maybe start over, or even give up. The story was very carefully set up to make this a legitimate way to cap off the narrative. In Undertale, the story is persistent – and any runs of the game, even when reset, are remembered and color the experience of playing Undertale going forward in little ways. In Undertale, giving up and starting over is a legitimate and designed-for chapter of the narrative.

The skeleton Sans sends an onslaught of bones and laser beams at the player's heart-shaped cursor, turned blue now. After a moment he sends the blue heart careening into a deadly maze of bones as it flies against its will to the right side of the screen.
What a reasonable amount of garbage that can instantly kill me, on the screen, all at once

Giving up can mean a new beginning, a world where the player is not a force for destruction and misery, but a force for change and friendship. Whenever I play Undertale, I love to play the part of the sinister player destroying the world and its inhabitants for callous entertainment (and in a way, I truly am that), but then our protagonist, Frisk, overtaken by sorrow after killing Sans, is able to wrestle back control and ease me into a more peaceful, and ultimately more fulfilling world. I’ll play a No Mercy run just up until I’ve killed Sans, and no further. I’ll then roleplay the regretful monster, the powerful demon whose lost everything, and has no more mountains to conquer. From there I return, back to the beginning of the game, anew with a desire to learn and try again. Undertale makes failure an avenue for learning and improving at the game yes, but also a potential narrative moment of fulfillment.

I love this scenario. It creates a full arc for me, as the will and intention of the player character Frisk, to go through. It’s a rich narrative that unfolds entirely through gameplay that I get to be a part of. That’s the real magic of difficulty in games for me, it’s something entirely unique to the medium, a level of interactivity other forms of art simply cannot achieve. Sans is blisteringly difficult, to the point that he may even feel antagonistic to the human behind the screen. But the game isn’t trying to punish you, nor look down on you, it’s trying to play with you. It is a game, after all. It is interactive theater, a stage show where you are the star. And maybe just maybe you’ll get something valuable out of the experience.

“Death and rebirth, trying and overcoming—we want that cycle to be enjoyable. In life, death is a horrible thing. In play, it can be something else.“-Hidetaka Miyazaki

You are meant to be along for this emotional ride through joy, through sorrow, through fear, through love, through distress, and yes, through frustration. It’s a frustrating thing to be denied passage, to face an opposing force that’ll do everything in it’s power to stop you. If the art is to be evocative, it may be necessary to instill that sense of frustration. I will deploy the second of two Bennett Foddy quotes, as I admire the way he puts it;

“What’s the feeling like? Are you stressed? I guess you don’t hate it if you got this far, feeling frustrated. It’s underrated. An orange, a sweet juicy fruit locked inside a bitter peel. That’s not how I feel about a challenge. I only want the bitterness. It’s coffee, it’s grapefruit, it’s licorice.” – Bennett Foddy

Sans the skeleton sleeps, standing up, in the center of a screen with a battle UI. A red heart-shaped cursor moves over to the UI button labeled "FIGHT". A slashing effect moves toward Sans, but he slides out of the way and begins to speak, but is cut off by a second attack, which leaves a violent gash across his chest.
Frustration and loss isn’t just a roadblock to joy and catharsis, it’s an essential part of the whole.

Frustration is not the opposite of fun. I think the runaway success of games like Dark Souls, Elden Ring, and Undertale, games that very much use frustration as feature of their storytelling, are strong evidence of this. There are hosts of games that follow similar patterns. When you play and watch people play difficult games as much as I do, you begin to notice that not only is frustration not a deterrent to the fun for most, it often accompanies the highest highs of player’s positive emotional reactions. Art is not a vehicle for merely delivering joy and nothing else. Life is a rich tapestry of a variety of emotions, and if art is to speak truth, then I think it’s worth considering how best to accurately reflect that. I’ve been talking a lot about feelings and emotional reaction, and I can’t overstate how subjective such things can be. You’re walking a fine line when utilizing traditionally negative emotions such as frustration to tell a story. As I said before, difficulty is a very nuanced and complex topic and this is just one aspect of it, one feature of difficulty to consider when configuring the shape of the experience you want to create. Difficulty can be used to tell and legitimize interactive narrative in a very profound way. That said, not all games need to, and by no means should they, take the same shape. Knowing how best to achieve the goals of your design starts with understanding your goals, and understanding the tools at your disposal.

Sans the skeleton sleeps soundly, standing up, in the center of a screen with a battle UI overlay.

You have something called ‘determination.’ So as long as you hold on… so as long as you do what’s in your heart… I believe you can do the right thing…

The Evolving Boss Design of Elden Ring: Godrick The Grafted

Yes, yet another FromSoft post. In my defense, Elden Ring, 2022’s action RPG release, and FromSoftware’s foremost foray into the open world game format is impossibly large, and has proven just as impossible to keep out of my head. Boss battles, with their big climactic set pieces and heightened sense of challenge have always been FromSoft’s bread and butter when it comes to their contemporary era of action games. It seems these games are constantly pushing the envelope of how grotesquely deadly these big baddies can be. The opening boss battles of Elden Ring are a strong showing, coming in with all the bombast, shock, and awe that I’ve come to expect, but I did notice something – a subtle shift in the design for the likes of Margit The Fell Omen and the first major ‘chapter boss’, so to speak, Godrick The Grafted. This new design direction seems to carry forward with the rest of the bosses of the game, so I really wanted to talk about it.

In a castle courtyard, a huge man made of many limbs grotesquely frankenstein'd together swings his axe above his head, summoning gusts of wind upon which he rings up into the air, then slams his axe down on a thief, who dodges out of the way.

For those unfamiliar, combat in FromSoft RPGs generally can be broken down like this; The boss has a pool of attacks and moves from which they will select their action, reactive to the player character’s position relative to themselves. A dragon may swing its tail if the player is behind them, or breath fire if the player is standing at a distance, and such. In response, the player has a number of options including blocking with their shield, moving out of the way of the attack, or avoiding damage with the dodge roll maneuver, which gives the player a brief window of invincibility. The dodge roll is very very powerful, and potentially frees up a hand that would otherwise be occupied with a shield, so the player can two-hand a more powerful weapon, or offhand a second weapon. As such, the dodge roll is traditionally the player’s greatest tool of avoiding harm while they look for punishes, brief windows in which a boss cannot counterattack. These punishes happen when the end of a boss’s action animation precludes them from taking another action for a brief moment. The recoil of swinging a huge axe, for example.

Traditionally, FromSoft RPGs like Dark Souls and the like have a very powerful strategy when it comes to fighting bosses – getting behind them. Though even as far back as Demon’s Souls, many enemies were equipped with back-facing attacks meant to dissuade players from becoming too comfortable back there, the caveat was generally that these back-facing attacks were a lot more manageable than many of the front-facing ones, so circle strafing enemies remained a pretty dominant strategy in most situations, all the way up to Dark Souls 3, with some exceptions, but we’ll get to that.

When I first fought the major boss Godrick The Grafted I found myself failing and dying. A lot. To be expected of the infamously difficult FromSoft RPGs, right? Well sure, but Godrick is a very early game boss, so I had to rethink my approach. Surely the first major boss couldn’t be that difficult. Godrick’s fight is split into two phases, though both operate generally the same. He has a large pool of attack options, but I’ll mention some of the most common ones. First he has what I’ll call his basic attack combo, which he prefers to use on players that are standing medium-close to him, in front of him. It starts with a very slow axe swing to teach players to rely on their eyes for dodge cues, not just prediction, as aggressive predictions tend to result in early dodge rolls that are punished by such slow attacks. He’ll then follow up this axe swing with a series of swipes of his arms and axe, using a combo of variable length. If the player disengages, the combo stops at two attacks, but continues for up to four attacks if the player stays close.

In a castle courtyard, a huge man made of many limbs grotesquely frankenstein'd together brings his axe down on a thief, who ducks and dodges toward the arc of the axe swing, avoiding harm, but getting hit by the grotesque man's swinging arm, just as the thief is standing up.
The hitbox of this move is designed very specifically to catch players getting too close.

This attack presented a bit of a problem for my muscle memory. The dominant strategy of old in Dark Souls was to ‘dodge through’ enemy attacks. If you got the timing right, the invincibility of the dodge roll could carry you through the hitbox of an enemy sword swing, and the best way to do this would be to dodge into the arc of the swing, toward the direction it’s coming from. This minimizes the time your player character and the enemy hitbox intersect, making a successful dodge more likely. This results in a lot of getting behind enemies and proverbially hugging them up close. Try this on many of Godrick’s moves, however, and he’ll have a hasty response.

The better maneuver is to out-range Godrick, and dodge away from him as he attacks, dodging or moving toward him only to punish his openings, although these can be difficult to judge at first because he has so many followups. Remember how I said his basic attack has a variable length? Many many of Elden Ring‘s baddies have followups like that, contingencies to account for the different places the player might be standing to avoid harm, so sometimes what you think is a punishable opening on a boss is simply not.

In a castle courtyard, a huge man made of many limbs grotesquely frankenstein'd together wildly swings his axe at a thief, as the thief sprints away, out of his reach.
Dodging this attack with rolls is theoretically possible, but so high-risk as to be ill-advised

Godrick’s ‘big swing’ attack, as I like to call it, happens when he rears up his axe with both hands slowly, giving the player ample time to reposition themselves in response. This attack has five very large swings, some of which hit in nearly 360 degrees. They’re also very deadly, making standing up close to him equally as deadly. This attack is very punishable, but only at the end of its run. Godrick has a lot of attacks like this, such has his whirlwind, which not only nudges players away in a circle around him with a bit of damage, but is immediately followed by one of two attacks that very difficult to distinguish without the full picture, and dodging the wrong one could mean taking a big hit. A lot of Godrick’s and many Elden Ring enemies’ moves are like this. They are of variable length, and potentially can sweep the player up in a very long combo. It’s not that getting behind a boss is impossible, it’s just not very safe.

Another feature worth mentioning, is the stagger system, somewhat borrowed from Sekiro. All enemies have an invisible value that is diminished when they take a hit, but constantly regenerates over time. If this value hits a certain minimum, the enemy will stagger, and become open to a critical attack. Due to its nature, and the many variables that go into it, this value can be somewhat unpredictable, especially as it is not displayed anywhere, unlike in Sekiro. Previous FromSoft RPGs had systems like this, though they were less powerful, only allowing an extra hit or two when enemies stagger, rather than a chance for big damage. Since the exact moment a boss will stagger is so unpredictable, the best thing to do is to just keep up the pressure, to increase your likelihood of getting a stagger, which naturally makes Elden Ring combat more aggressive and high-tension.

In a castle courtyard, a huge man made of many limbs grotesquely frankenstein'd together leaps into the air riding on a whirlwind and slams his axe down, but his target, a thief, dodges out of the way, then stabs his exposed body several times, causing the hulking man to gush blood, and fall onto his knees.
Whenever a stagger does occur, it creates a rewarding moment of catharsis

Frequently, if you do manage to get behind Godrick he’ll roll out of the way, and quickly prepare another attack. This often results in taking a hit, and disrupts your ability to consistently damage Godrick, meaning you’re less likely to stagger him with this route.

In a castle courtyard, a huge man made of many limbs grotesquely frankenstein'd together rolls away from a thief as the thief positions himself behind the hulking man. As he finishes rolling, the man swings his arms at the thief.
Godrick is quick to respond to unfavorable positioning.

So what’s going on here? Why do the bosses in Elden Ring feel so different? I think I know. It seems to me that a major design goal for the bosses of Elden Ring ties into that verisimilitude I talked about when discussing Sekiro. FromSoft wanted fights in this game to feel more genuine, to reduce the friction between the gameplay and narrative of a life-and-death fight. In a real fight, you’d expect there to be a lot less dancing around opponents and a lot more squaring up, staring them down face-to-face, standing opposite one another as you trade blows. As such, everything about Godrick’s boss design makes him equipped to enforce that scenario. Getting behind him is dangerous, standing too close is dangerous, all defined by Godrick’s particularly designed behavior.

In a castle courtyard, a huge man made of many limbs grotesquely frankenstein'd together takes his own battle axe and slams it into his other arm, screaming all the while, before succeeding in cutting it off, the end of his arm landing bloodily in the foreground.
Extremely normal behavior.

Between a much broader repertoire of tools at his disposal for hitting players at any angle, the ability to quickly reposition himself, and a much more unpredictable, variable moveset that adapts in length and reach to player positions in real-time, fighting Godrick the old way is much less viable. When players have to rethink their approach, they’ll likely come to a similar conclusion I did; the safest place to fight Godrick is in front of him, at a decent distance, where they can see his attacks clearly, and where dodging is less likely to be a death sentence. Compare Godrick to an early boss from the first Dark Souls game.

A rotund demon in a crumbling ruin swings a mighty hammer at a thief, he dodges the swing and rolls behind the demon with ease. The demons stands there for a moment, then swings his hammer feebly in front of him, unable to reach the thief.
Godrick would have killed me three times by now.

Notice how completely unable the asylum demon is to respond to my parking behind his rump. The Asylum demon does have an attack that can reach behind him, but it is exceptionally easy to dodge, with a huge slow windup. Also, it’s only just the one. Outside of trying to butt-stomp the player, the asylum demon really has no options. He can’t reposition himself either. Godrick, on the other hand, can… slam the ground to create a large earthquake in a circle around him, quickly whip up a whirlwing that knocks nearby players way, jump into the air to reposition himself and attack simultaneously, roll away to a more advantageous position, initiate his 5-hit super combo, etc.

Sped-up footage of a thief continuously circling around a large demon, standing behind him as the demon fumbles to try and attack the thief, in an old dimly lit ruin.
Yep, it pretty much just goes like this.

The old way of fighting bosses in Dark Souls was, and still is, fun. Some of the most fun I’ve had in a video game, but FromSoft isn’t a developer to rest on their laurels. The evolution of their boss design seems to be working more and more to create a believable experience with a strong sense of weight and presence. Verisimilitude is a major priority, but not only that, there are a lot of important advantages to designing bosses this way. For one, the player will be able to admire and appreciate the art of the boss they are fighting a lot more from a nice, framed, middle-distance shot, than from standing between their legs (depending on the boss, and the player, I guess..). The clearer framing also makes reading enemy telegraphs a lot clearer and easier. If a player is rightfully standing behind a boss because it’s the safest approach, camera zoomed up way close, they might understandably become frustrated by attacks they cannot see, and won’t determine the reason, succeeding by their strategy in the end, but having less fun in the process. Since so many of a boss’s attacks are triggered by the player position, having the player spend so much time in an awkward position may result in them not even seeing the boss’s coolest most impressive attacks, diminishing the entire set piece.

It seems FromSoft has been working toward this goal of the ‘squared off, face-to-face’ boss battle as a default for a long time. As far back as Dark Souls 2, we saw evidence of this. It is very popular to complain about the enemies’ ability in that game to track the player’s position with their attacks, meaning the enemy’s entire model actively rotates to align itself with player position, making circle strafing a weaker strategy. This worked to an extent, but also was pretty transparently artificial. People really just don’t move that way, spinning on their heel to align the swinging of their arms in a way that defied physics, so much so that my friends and I had taken to calling the process the ‘ballerina twirl’. Bloodborne reduced the ballerina twirling quite a bit, making up for it by giving the player a tool which made squaring up with bosses a lot safer – an extremely powerful frontal parry. Since the player needed their enemies to attack them from the front to succeed in a parry, it became a much more desirable spot to stand. The backstab of Bloodborne, being so slow, also meant circling around was suitable for stealth, but less suitable for active combat. Dark Souls 3 introduced a lot of the standard for how attacks are shaped in these games; big sweeping things that sometimes make as much as 360 degrees around the boss a danger zone. That game also introduced a lot of the adaptability and variability of enemy move-sets that made them more equipped to react to players constantly trying to get behind them.

This new design direction seems to have culminated in Sekiro, which was developed alongside Elden Ring. In Sekiro, parrying is everything, so enemies had to constantly be positioning themselves and the player into a squared off position. A lot of strategies employed to accomplish this seem to have carried over to Elden Ring, adapted to account for the lack of a strong parry, of course. The behavior of Sekiro‘s largest bosses is rather reminiscent of Elden Ring bosses. I can’t say I dislike this new direction FromSoft is taking its boss design. Godrick is a fun and exciting challenge, and one I actually got to see visually in its full glory! Just while recording footage for this post, I was struck by how consistently beautiful and exciting the action looked in motion, so by my estimation the efforts to make fights in Elden Ring feel more genuine, worked.

A hunched, cloaked figure gently caresses the face of a skewered dragon carcass in a castle courtyard. A close up of the figure reveals a six fingered hand.

…And one day, we’ll return together …to our home, bathed in rays of gold…

Torrent of Elden Ring: Gaming’s Most Powerful Horse

My god, so that Elden Ring huh? Certainly has given me a lot to chew on. There are any number of topics I want to write about on that game eventually, from the new approach to boss design, to the integration of more summoned NPCs, the nuances of the tweaked melee combat, the expanded magic toolset, to how the open world has changed the way enemy encounters are designed. So I had to narrow it down, and I want to talk about one of the game’s most prominent new features and a major marketing point of the game – the presence of a horse and mounted combat.

I’ve wanted to, for a while now, do a write up on game design’s strange and somewhat hilarious history with mounted riding animals in a broader sense, but Elden Ring‘s local speedy boy Torrent has given me cause to talk about the ways Fromsoft has distinguished mounted combat and movement specifically. There are a lot of really cool design choices I’ve noticed that went into making Torrent a beast on the battlefield, with combat that feels as good as any of Fromsoft’s previous unmounted offerings. Here are just a few.

Don’t Forget, We’re Playing a Video Game

When I say ‘we’re playing a video game’ I don’t mean to say that player immersion and realism needs to go out the window. Kind of the opposite, actually. I feel like the tendency in some games is to put a lot of physics and terrain based restrictions on how player characters move. Maybe in the name of realism, maybe in the name of feeling cinematic, but often this approach leads to characters that control like tanks, ultimately creating more friction between the player and the experience, not less. My preference is to make the character controlling process as smooth as possible.

Torrent is a horse, and the design makes considerations to make his control feel more like a horse. He doesn’t stop on a dime. He’s got some acceleration and deceleration to him. He can’t turn on a dime, he needs space to swivel his rear around when you about face. Thing is, he doesn’t need that much space, and he doesn’t take that long to reach full speed or come to a stop. Torrent has movement limitations, but only barely enough to convince you he’s moving a like a horse. Fromsoft put the utmost priority into making sure he just moves smoothly, minimizing friction. Even given the very minor limitations placed on mounted movement to give the impression of riding a horse, enough to reinforce verisimilitude, these are somewhat circumvented by the presence of a double jump!

A warrior astride a horned horse leaps over brush and rocks as they ride across a field. To jump over a tree, the horse gains additional height with a spectral magic circle in midair.
Imagine animals actually being able to navigate their environment. Incredible.

Yes, the horse can double-jump, completely redirecting his momentum in midair, enhancing the length and height of his jump in the process. Torrent doesn’t turn as hard as the player can on foot, but he can totally reorient himself by jumping. This mechanic is in place for several very important reasons. First, Torrent is an exploration tool and Elden Ring is filled with very precarious drops and complex terrain geometry. Something a lot of other games featuring horses have stumbled on is how damn incompatible the mounts are with their own game world. Lots of video game horses can move fast, but can’t go anywhere. A glut of limitations on mounted movement can make it too specialized and niche, not versatile enough to be used often.

Torrent doesn’t feel like a second entity your player is fighting to control, but rather an extension of themselves. In that way Torrent is treated more like a vehicle. It’s a compromise, in that this perhaps distances the design from the realism of riding a horse, but for the design goals they were trying to meet, it makes a lot of sense. Those design goals being, that riding Torrent needs to serve as a distinct combat style that feels as seamless and satisfying as grounded combat, an extension of Elden Ring‘s primary gameplay mode.

A Horse Is Much Faster Than You

This is a weird one to see not fulfilled so many times. Part of what makes Torrent so successful as a mount is that he is just a lot more maneuverable than you, the human player. He’s a horse; he can move at like double your top speed. I’ve played an astounding number of games with mounts that don’t feel all that much faster than just walking. The double jump plays into this too. Torrent has horns as well, kind of giving him the impression of a mountain goat, which communicates his exquisite climbing ability. Mountains, cliffs, and ravines are common obstacles in Elden Ring‘s world of the Lands Between. Distinguishing Torrent with his much more robust set of movements for dealing with such obstacles proves his worth to the player, and was very important in cementing his place in the game.

Torrent Isn’t Made of Paper

For a number of reasons, video game horses are often not equipped for combat. Perhaps the game simply does not have a design for mounted combat. Perhaps the design does not account for a constant companion that would throw off the game’s design. Maybe they just thought it’d be awkward if you constantly had a horse following you, biting and kicking goblins. Whatever the reason, horses in games often wind up very weak. Ride them off a two-foot-high outcropping and they crumple like they’re made of paper mache. Get jumped by a dragon, and you’re never seeing that horse again.

First of all, Torrent can fall very very far without issue. He needs to navigate complex environments with a lot of versatility and if he was constantly dying from underneath the player, it just would not be workable. Torrent has knees of steel, and can bound off of cliffs with ease.

A warrior astride a horned horse jump off of a cliff from grassland into a shallow lake. The horse lands gracefully and harmlessly.
Oh no no NO NOT OFF THE CLIFF oh- oh never mind, we’re fine.

What’s more, Torrent is meant to be used in combat, and thus can take a hit or two. In fact, he’s quite a bit more sturdy than even the player. Riding Torrent is a very safe place to be, gameplay wise. He can even take hits for you, depending on the angle of attack. To counteract this big advantage, riding Torrent is given a specific risk. You can take a lot of hits, but if you are knocked off of Torrent, either by losing your balance, or if Torrent dies, you are sprawled onto the ground and left very vulnerable. Re-summoning Torrent will require taking a moment and possibly sacrificing some healing resources.

A warrior astride a horned horse pass in front of a gray dragon, which angrily bites at the duo, drawing blood. The horse stumbles, but quickly composes itself and rides on, warrior still in tow.
Good God this horse is built different.

We’re Just Not Bothering With The Idle Problem

This one is funny to me, but I respect it. If you’ve played a video game with a horse you’ve seen it. The world’s jankiest implementation of entity spawning known to man. I’ve legitimately played a AAA game where I’ve seen a horse pop into existence upon use of the ‘horse summon’ button. The problem is, what do you do with mounted animals when they’re idling, not being ridden? Do they just run off somewhere? What’s the visual of that? How is the horse summoned and where does it come from? Does the horse exist in real-time, meaning it stays where you leave it until you come to pick it up? Torrent kind of just… double-jumps over this problem entirely. When summoned, he appears underneath your player in an instant, and disappears just as instantly when dismissed. Sometimes, the realism of a thing isn’t worth the headache. Torrent works better if his presence is never in question, so it just isn’t.

The side-benefit of this is that Elden Ring can switch between its two combat modes, mounted and on-foot on the fly and seamlessly. Torrent is always available outdoors if you need him. You can even ride Torrent into battle and jump off for a cool dismounting attack.

In a shallow lake dotted with dead trees, a lone warrior whistles on her fingers, summoning a horse that phases out of spectral energy beneath her as she mounts up, and rides toward a large gray dragon in the distance.
Man, I love not even having to think about what stupid stuff Torrent’s Horse AI would get up to.

As Above So Below

This is a straightforward idea, but one that implies a lot of extra work in creating game assets. While mounted, the player is allowed to use basically any attack they can use while unmounted. Making the mounted combat feel seamless and parallel to unmounted combat was very important for Elden Ring, so a lot of extra animation and frame data was created to ensure the player’s preferred weapon and spells were available to them while mounted on Torrent as well.

Combat As a Spatial Problem

As I’ve said before, action combat is mostly all about relative spatial relationships between player and hazard. Where a player is standing when an enemy attacks determines if damage is dealt, etc. A friend of mine pointed out that video games are and have always been, very very good at mapping spatial problems, and with this in mind, real-time combat can often be boiled down to very elaborate spatial problem. Elden Ring‘s mounted combat leans hard into this concept. Normally, combat in Elden Ring and other action RPGs from Fromsoft centers majorly around the dodge roll mechanic, in which well timed button pressed can impart a moment of invincibility that can be used to circumvent damage regardless of positioning, although positioning remains important due to the invincibility window’s briefness.

Torrent does not have a dodge with invincibility frames, this seems to have been the main thrust of distinguishing mounted combat in Eldren Ring from its on-foot counterpart. There are other key differences, of course, but they all seem to revolve around this one major change. Having no iframes on Torrent changes the way one approaches combat immensely. ‘Rolling into’ attacks, so to speak, such that when a dodge roll ends, the enemy’s attack has already moved past the player’s position, is a fundamental strategy normally. But now, keeping your distance is a player’s best method for not getting hit. Since Torrent can’t dodge in the way the player can unmounted, dodging attacks becomes entirely a matter of positioning. In place of a dodge, Torrent has a dash which gives him a quick burst of speed, useful for getting player-seeking projectiles off your tail or outmaneuvering other mounted combatants. So you’re encouraged to always disengage after riding up to an enemy to hit them.

Dead serious, fighting this dragon is an all time great achievement of experiential design

The result of this simple change is that combat much more resembles a series of ride-bys where the player rides up to their foe, hits them quickly, then circles behind or gets out of their reach in some way. There’s a lot of two horses riding past each other in a sort of joust… kind of like how mounted combat actually works, or at least how one might imagine it works! There’s that verisimilitude again! Dang. By removing this seemingly essential tool, Fromsoft has reinforced a method of gameplay that resembles what they are trying to simulate, brilliant! When you can’t dodge with invincibility, you have to outmaneuver danger, and that means riding around and past enemies like you’re an actual mounted cavalry, you can’t just stand squared up to your foe and hit them without expecting to get hit back, and thus you are encourage to be constantly on the move, kind of like you’re on a riding animal. Dang. It just keeps fitting together, doesn’t it?

A warrior astride a horned horse gallop in the foreground as brush and trees rush by, a dragon in the background flies across the sky raining down fire narrowly outpaced by the mounted warrior and his companion.
No amount of rigid unresponsive tank controls would have made this moment feel more cinematic

So this is all well and good, but Elden Ring is an extremely three-dimensional game with a lot of danger happening, left, right, center, up, and down all the time. Torrent needed another tool to avoid damage. Dang, he already has one we’ve talked about, doesn’t he? The damn horse can double-jump. Torrent’s extreme level of vertical maneuverability makes awareness of enemy attacks even more essential, and some can be completely circumvented with well placed and timed jumps. Very little in this, or any video game feels as instantly exhilarating as jumping over a dragon’s flaming gullet, the camera pulled way out to capture the action. Forget harshly limited controls that make characters feel more ‘realistic’ or ‘cinematic’. No, good gameplay can do that just as well.

A warrior astride a horned horse runs up on a large gray dragon in a shallow lake as it begins to breath fire, the horse runs up a large rock and bounds into the air, over the dragon's head. The warrior skewers it with a spear, cutting off the fire breathe and felling the beast. The two ride out, then come to a stop as the words "GREAT ENEMY FELLED" appear.
WOW! Someone get this horse an apple and some sugar cubes.

So yeah. Elden Ring impresses on a number of fronts, but it’s easily got the best mounted combat I’ve ever played. Not that I’ve played a huge number of mounted combat games, but riding Torrent is just as engaging as fighting enemies on foot, which is kind of impressive. The riding in and out to swipe enemies in the side as I pass, jumping over dragons, and covering great distances as I bound over cliffs, it’s all very exciting. Rarely have I had just a smooth experience with mounted animals in games. Fighting that dragon? Almost indescribable how elated I was, to fight a dragon that felt like it had an appropriate sense of danger, scale, and gravitas. Couldn’t have done it without you Torrent, here’s to many more adventures in the Lands Between.

Torrent has chosen you. Treat him with respect…

Boss Breakdown: Bloodborne’s Blood-Starved Beast

You step into a large chapel overgrown with sickly vines and eerily empty, save for one hunch-over figure at the back, nearest the chapel shrine. It stalks toward you like an animal as its boss music kicks in, and you’re introduced to its name: Blood-Starved Beast. The Blood-Starved beast is a mid-game boss in FromSoft’s action RPG Bloodborne. Though technically optional to complete the game, Blood-Starved is centrally located, and gatekeeps one of the game’s major features; the chalice dungeons. It is thus likely to be a boss that most players encounter toward the start of Bloodborne‘s mid-game. This malnourished and emaciated figure fights with the ferocity of a starved predator, and its design backs up this idea while also serving appropriate functions within the overall experience of Bloodborne.

A hunter in a long black coat and tricorn hat pushes through a wall of fog into a run down stone chapel, lined with columns and torches. A lone figure, on all fours, slowly trudges toward the hunter from the distant shrine of the chapel.
FromSoft really knows how to set the scene

My philosophy when it comes to design, and especially in regards to big set-piece combat encounters like this is to reinforce the overall feel and experience of the game. It’s one of the reasons I so admire FromSoft’s design ethos overal.. Bloodborne, at this point in its story, is a visceral gothic horror about hunting horrible, bloody beasts. It’s gritty, it’s guttural, and it’s dangerous. The blood-starved beast begins to reinforce Bloodborne‘s overall aesthetics and feel from the moment you see its visual design. It’s a hunched-over, somewhat skeletal, feral humanoid figure with large portions of its skin bloodily peeled from its back, and draped over its head like a shawl. It’s a gruesome sight that reflects the environments and tone of Bloodborne. It fights with a hunter’s aggression, homing in on the player with a ravenous intent, keeping the fight high-intensity. The Blood-Starved beast is a quintessential representation of Bloodborne‘s hunt. It’s bloody, it’s animalistic, brutal, and imminently deadly.

A hunter in a long black coat and tricorn hat runs at a semi-humanoid skeletal beast, who charges at him on all fours, before wildly swiping at the air with large claws.
This thing’s animations and attacking pacing create a frantic sense of danger

Blood-Starved’s sense of danger is essential to making its fight come together. It needs to feel as though this thing could rip out your throat at any moment, a violent, unstable, rabid animal. This is most embodied in its grab attack, which does a huge amount of damage and can potentially kill a player outright. Highly lethal attacks like this are a favorite of FromSoft’s design to establish the threat of their enemies, and keep them oppressive. The ever-present looming promise of an attack that can potentially outright end the fight in a game over keeps the player on their guard, and shores up the tension. That said, such attacks can feel cheap and unfair if not handled carefully. If an attack is extremely deadly, best practice is keep it predictable and telegraphed, so if the player does fail to avoid it, they don’t feel as though they were blindsided, and the mechanic remains one of skill in the player’s mind, not a random vector of bad luck. The Blood-Starved Beast assumes a very particular posture, it’s normally gyrating and animated movements become still, and focused, it’s arms almost exactly shoulder-width apart in an even stance. This stance really stands out once you know to look for it, rewarding close observation of the boss, which is a standard for FromSoft enemy encounter design, something they are very much always looking to reinforce.

A hunter in a long black coat and tricorn hat strikes a vaguely humanoid skeletal beast with an overhead swing of a large battleaxe. The beast reels, but then assumes a steady, wide stance, before leaping at the hunter, who barely dodges out of the way.
After the first axe strike, you can see the beast assume this very still, square stance, a clear telegraph for its deadly grab attack, seen here missing the player.

On the more usual and rote side of things, the Blood-Starved Beasts’s primary attacks almost exclusively are aimed in front of it, and they have very little player tracking, meaning just a bit of movement will move the player out of danger. If the player moves too far away, however, the beats will initiate a Dashing attack that covers a lot of ground, to reset the neutral positions of the fight. These patterns, while predictable and simplistic, create the very strong spacial dynamics of the fight.

Blood-Starved’s quick attacks and short response time make it dangerous to approach, and risky to engage in close combat. As an early-to-mid-game boss, Blood-Starved functions to help really cement the player’s skillset and prepare them for the steeper challenges that are to come, and it does so by emphasizing Bloodborne‘s parry mechanic, in which players shoot an oncoming attacker with a firearm just before being hit, stunning the enemy and preparing them for a visceral counter-attack. It’s a powerful option that can carve through enemy health bars quickly, and will remain useful throughout the rest of the game, so Blood-Starved really pushes the player to master this. Its attacks are fast but reactable, wide-reaching but mostly short-ranged, perfect for being parried by the player’s firearm. If the player utilizes the parry and visceral attack, the difficulty of the Blood-Starved Beast can be curbed to a great degree, rewarding mastery of the skill.

A hunter in a long black coat and tricorn hat is attacked by a vaguely humanoid skeletal beast who swings with its immense claws horizontally, striking the hunter just as he blasts the beast with a flintlock pistol. The beast reals from the gunshot, so the hunter grabs it by the neck, and violently rips out a gush of fluid, sending the beast flying onto its back.
Even though the player takes damage here, he is able to rally all of his health back thanks to a well-timed, aggressive parry, while grievously harming the boss to boot.

Later on in the fight, The Blood-Starved Beast will start to string a long series of attacks together. With a maneuver like this, the beast is more likely to get some licks in, to give players more opportunity to leverage the health-restoring rally system, which rewards reprising attackers. It also punishes a lack of attention paid. Once the attack begins to hit a player, it is not difficult to disjoint with a dodge so the remaining hits miss, but being caught unaware could mean quickly and drastically losing health. There is one more feature to this attack though, which is that, since it’s a rapid series of strikes it makes parrying it very easy. A lot of times FromSoft will employ rapid attacks like this that seem very intimidating, but realizing how parrying works – that it requires player input to intersect with a certain part of an enemy’s attack animation, means knowing that if you try to parry one of these rapid attacks you are very likely to succeed, as the enemy’s ‘vulnerable’ animation is flashing past over and over again, and it only needs to be snagged by your parry attempt once. As I supposed this boss is meant to greatly reinforce the use of Bloodborne‘s parry, I suspect this is one of the primary reasons for the presence of this attack, a reliable parry opportunity. It’s a powerful attack, but it can be turned against the beast, with an equal and opposite counter.

At the same time, Blood-Starved maintains the standard formula of FromSoft’s boss design through a moveset that reinforces spacing and timing. If the player is unable to master the parry mechanic, Blood-Starved is surmountable through diligent use of spacing. Fast yet telegraphed melee attacks make engaging with it directly from the front infeasible without parrying, so circling around behind can be effective. The boss’s preferred response to this is to create distance between itself and the player, which effectively sets up for it’s leaping grab attack or dashing slash attack, both of which require precise timing to avoid. This makes the boss feel proactive and responsive to the player’s actions, while ensuring player’s are unlikely to escape the battle without seeing their enemy’s most deadly attacks. Circle strafing it is an effective tactic, but not one that carries no risk. In a game that relies so much on its atmosphere, this illusion of a thinking and responsive agent behind enemy AI is essential.

A hunter in a long black coat and tricorn hat steps to the side as a vaguely humanoid skeletal beast tries to bite him. He responds with a strong vertical swing of his battle axe.
The best mainly will attack forward and in front of itself, making circle-strafing very effective

Once The Blood-Starved beast has taken significant enough damage, it will begin the first of two phase transitions, marked by a screeching roar it lets loose. Starting in phase 2, its attacks will now apply poison to the player. The addition of poison to the fight accomplishes two things. One, it ramps up the tension and danger of the encounter. If you’re hit by the beast too many times, even if you’ve got enough healing to recover from the damage, you’ll be poisoned and a constant ticking clock will hang over you. This punishment effectively sets the standard of how often the player should expect to successfully dodge boss attacks without incurring major disadvantages in the future, establishing their expectations accordingly. Secondly, the building of a slow poison encourages a more aggressive playstyle, which is one of Bloodborne‘s chief design goals and hallmarks, which sets it apart from its more slow and calculating predecessors, Dark Souls and Demon’s Souls. Players in high-intensity situations will often try to ‘turtle-up’, so to speak, avoiding direct encounters so-as to preserve their resources and maximize safety. This is somewhat the opposite of Bloodborne‘s strengths as an action game, so FromSoft went to great lengths to reward proactive behavior.

A hunter in a long black coat and tricorn hat fires his pistol repeatedly at a vaguely humanoid skeletal beast as it mills about on the floor. After a moment, it screams and rears up, a cloud of toxic mist emanating from its body and its wounds.
The poison is mechanically functional in gameplay, but also very stylish and intimidating

The presence of the poison’s ticking clock communicates that your resources will be depleted if you don’t return a good amount of strikes that the Blood-Starved Beast can dish out to you, complemented by Bloodborne‘s rally system, which restores health if player damage is followed quickly by successful attacks against enemies. Blood-Starved serves as an excellent tool for engendering a more aggressive playstyle that matches the frantic and gritty nature of Bloodborne’s encounters. Phase three sees this pushed further, with clouds of poison now spewing from the beast’s wounds in all directions as it attacks. Spacial awareness becomes even more pivotal, and reliance on just one strategy becomes dicey. Players who can utilize a variety of approaches based on the situation, and do so proactively and aggressively, will find the final phase of the Blood-Starved Beast much smoother.

If a player is too risk-averse or simply not yet up to playing very aggressively, there are alternative solutions to besting the beast for the more strategically-minded. ‘Antidotes’ are plentiful from the carrion crow enemies that dot the level leading up to Blood-Starved Beast, and keeping them handy can put off the pressure from poison, if you find a moment to take one in-between dodging attacks. The pungent blood cocktail, a common item meant to distract the bloodthirsty minor beasts players can encounter, appropriately, works on the Blood-Starved Beast. It will become distracted if one is thrown, and clamor after the small traces of blood within. It’s really refreshing to see that kind of flavorful ingenuity that relies on knowledge of the game’s fiction be rewarded that way. Purely from a gameplay perspective, it might not be obvious that an item like that would work, but from a narrative standpoint it makes perfect sense. Anyway, I always appreciate these little alternate routes to victory in difficult games. Thinking around a problem is as impressive a show of skill as tackling it head on, and players should be encouraged to do so. Reward experimentation.

Blood-Starved is an extremely effective mid-level fight, I think. It’s rather straightforward with its only real gimmick coming in the form of the poison, which as I explained, fits rather well in what the beast’s design utility is. That is, the beast is at home in Bloodborne‘s oppressive atmosphere, with its oppressive combat style, and helps to reinforce a lot of Bloodborne‘s combat fundamentals leading up to the more advanced challenges that await afterward. Players who find themselves able to fight aggressively, and reliably perform parries, counters, and sidesteps, will find themselves well equipped for what comes after. You might say the Blood-Starved Beast is a ‘skill check’ in that way. Designs that focus on both teaching the player abstract technical concepts while engrossing them in the atmosphere of the world is what FromSoft does best in its boss design, and I think the Blood-Starved Beast is pretty emblematic of that.

A hunter in a long black coat and tricorn hat is struck multiple times by a vaguely humanoid skeletal beast, then backs away. He injects himself in the outer thigh with something, then dodges out of the way as the beast tries to bite him.

Hunters are killers, nothing less…

Rethinking ‘Health’ as Game Mechanic: Sekiro’s Verisimilitude

Talking about Sekiro again. Sekiro is a ninja action game by FromSoftware, that involves a lot of sneaking around and backstabbing, but nearly as much sword-to-sword clashing and front-stabbing as well!

A ninja in a red coat and scarf attacks a spearman in an old-japanese-style interior. The ninja is stabbed, but blocks the followup attack, breaking his opponent's guard before stabbing him through.
That beautiful yellow bar there, that’s the posture bar. Love that thing. We’ll get to that.

Verisimilitude! A big word, and one of my favorite ‘game-designerisms’. What exactly does it mean? So if realism is invocation of the real, of reality upon your fiction, verisimilitude in the context of game design is the invocation of what seems real, or rather feels right to give the impression of reality. That distinction might seem fuzzy, but it’s very important. What makes a game realistic is an adherence to the facts of the subject you are simulating, from an objective standpoint. What gives a game strong verisimilitude is a respect for the experience of what you’re trying to depict. In the latter case, it is more important that the gameplay feels right than it is for the game to be objectively close to reality. A horse’s hooves are expected to make a certain sound in television and movies, but this sound is often expected to be the sound of clapping coconut shells, instilled in audiences after many years of the sound effect’s prominence in television and movies. As you can see, creating a fulfilling sense of verisimilitude involves a complex balance of player expectations and affordances. This becomes even more complex, as these things tend to be, when you involve the dimension of interactivity. In Sekiro, I want to talk specifically about the verisimilitude of the game’s sword fighting.

I think one of the things I respect most about Sekiro: Shadows Die Twice is the design’s willingness to reevaluate even the most fundamental assumptions of action games. The health bar, for one, is ubiquitous in such games. But, if you think about it, ‘health’ as a gameplay mechanic is a little weird, right? There is an arbitrary number representing your opponent’s closeness to death. Maybe the enemy will pull out some stronger ‘desperation moves’ when their health gets low, but so long as it isn’t zero, they can fight as well as they ever have. When that last HP or ‘hit point’ is depleted though, POOF! Their body’s ability to sustain its own weight goes up in smoke and they crumple like their bones have suddenly turned to sawdust. The HP mechanic exists for a reason, it works well. The suspension of disbelief needed to accept this abstraction of how fighting capability works is easy enough to achieve such that most players will not question it, in return for the fun gameplay that HP pools provide. Sekiro wants to go a step further though, and bring the abstraction of their combat closer to a cinematic realization of an idealized sword fight.

A ninja slashes a flurry of blows at a spearman in a old-japanese-style interior. The ninja steps on an oncoming spear attack, knocking it out of the way, then breaks the spearman's guard with a final blow, before stabbing him to death.
The enemy’s posture indicated by a yellow bar above them (the red bar is HP). As the posture bar fills, the enemy is closer to loses their stance.

In reality, a sword fight does not involve two guys hitting each other back and forth until one guy runs out of hit points and collapses. Really, it’s closer to the first guy to land a clean hit… has probably just killed his opponent. So Sekiro has created a system that abstracts this reality – Sekiro is not generally a game about cutting opponents over and over again until you’ve punched out all of their blood, but rather, you strike at opponents as they deflect your blows until you can land a clean hit, killing them instantly. This makes a lot of sense. In a real fight, wearing down your opponents’ defenses, finding ways around their guard, is paramount. Most opponents are not going to just stand there and let you stab them with a sword. The idea Sekiro proposes is that ‘hit points’ are secondary to ‘posture’, or the strength of one’s defenses.

Two ninja duel in a Japanese manor courtyard, one garbed in grey and a mask, one garbed in a red coat and scarf. The masked ninja strikes at the other several times, each time deflected in a flurry of sparks, after the fourth strike, the masked ninja loses his stance, and the red-coat ninja quickly stabs him to death.
clang, clang clang, CLANG. And its over. The nature of the posture system creates a ‘correct’ sense of lethality and back-and-forth in combat.

Posture in Sekiro can be reduced by striking an opponent’s guard to rattle their stance, deflecting their attacks with a well-timed parry, or participating in a number of other contextual maneuvers such as the mikiri counter, a special counter against thrusting attacks, or jumping over enemy sweeping attacks. Posture naturally regenerates over time, and the player or enemy combatants can quickly restore it by holding their position and guard. When posture is completely gone, enemies become vulnerable to a deathblow, which will kill them instantly, while the player becomes stunned if their posture is gone, forcing them to dodge or take a hit. The idea of becoming too tired or psyched-out to defend myself in a tense situation due to relentless pressure from my opponent is much more relatable to me than suddenly exploding into red mist and collectible coins because I was punched one too many times. I am able to, through this system, put a bit of myself in the battle. Feeling my defenses wear down is something I have really experienced, in real life, so that feeling of the real draws me further into the game through Sekiro‘s clever abstraction of the concept.

Deathblows becomes the real hinging point of combat in Sekiro. As I said, whenever an enemy’s posture is depleted, they become vulnerable to the instant-killing deathblow. I say instant-killing, but some enemies can endure two or even three deathblows before really dying. Sekiro takes places in a fictionalized and fantasy-tinged realization of 16th century japan, with inhuman warriors, monsters, and demons, so the idea that a giant ogre-man can survive two ‘clean hits’ before he goes down for good does nothing to hurt the verisimilitude in my estimation. It’s not realistic, but it still feels right. The thought that even the strongest demigod opponents you face can be brought down by just two or three true strikes really grounds the world in a sense of grit and lethality. The illusion is given that life in Sekiro is nearly as fragile as it is in reality, even with divine powers. These ‘deathblow counters’ work brilliantly with Sekiro‘s stealth mechanics as well, as one needs only sneak up on an opponent to get a free deathblow, no risky one-on-one fighting required. The two systems work so well together that I feel I could write an entire article just about that, so I’ll move on for now.

A giant white, headless gorilla charges a ninja in a red coat and scarf in a knee-high spring in a valley. The gorilla, wielding a giant sword and its own severed head, swipes at the ninja. The ninja manages to block, being pushed back by meters with each p;arry.
The verisimilitude of the combat mechanics highlights and heightens the impact of the supernatural elements. So it still works when a beast keeps fighting after decapitation.

Now hit points still exist in Sekiro, but their role has been shifted somewhat. Enemies will still die if their HP reaches zero, though this is much less likely to occur than their posture reaching zero. That’s not to say wearing down your opponent’s HP is pointless. The lower a combatant’s HP is, the slower their posture can regenerate. Now if HP is still meant to represent ‘health’, then this makes a lot of sense. A clean hit is likely to just kill someone in a sword fight, but a series of glancing blows are very likely to make someone less able to defend themselves. Cuts, bruises, a broken finger or two, such things would definitely add up to a poorer and poorer defense, and so this adds to the overall sense of internal consistency in Sekiro‘s combat. No matter the true reality, fighting in Sekiro creates an experience that meets the player’s imagination of how being a sword-fighting ninja would actually work. This level of gradient to the effects of losing HP, where the more injured a combatant, the worse they can defend themselves, addresses much of the inherent weirdness of HP I mentioned earlier, and it does it in such a simple, smart fashion.

Verisimilitude is all well and good, but I wanted to quickly go over some other practical advantages of this system. Since posture is an ever-shifting and renewable resource, it creates a very dynamic tension that can shift back and forth freely as the situation demands. Empowering reversals of fortune on the part of the player are common, as posture greatly rewards consistent performance, making a win when you’re on the ropes a lot more feasible than you might expect. Posture also allows many little tweaks and knobs for the designers to create a great variety of enemy fighting styles based on their posture stats alone. Maybe one boss restores their posture absurdly quickly, but doesn’t have huge stores of it, meaning this boss must only be parried a few times, but they cannot be given the downtime to recover. Maybe another boss has huge stores of posture, but recovers it very slowly. Some enemies might recover posture quickly, but lose this advantage if they’re even a little injured. Etc. There are a huge number of possible variations of these, and Sekiro implements pretty much all of them.

A twelve-foot-tall woman in a skull mask and large monk robes swings a naginata polearm in wide arcs at a ninja in a red coat and scarf, atop a bridge in the mountains, covered in snow and autumn leaves. The ninja is able to deflect several strikes, then steps on the naginata to block it, The ninja cuts the monk twice, then blocks two more strikes, before leaping over a sweeping attack and stomping her shin, breaking her stance, and stabbing her.
The versatility of the posture system can create endurance battles of attrition, or deadly quick-draw duels, depending on context.

Obviously glowing praise is my default position for much of Sekiro‘s systems, but it’s worth mentioning some drawbacks. Sekiro‘s posture system works specifically because of how intense a game it is. It demands constant, focused attention of players in a way that may not be appropriate for every game experience. It’s very much ‘Sekiro‘ but every action game may not benefit from such high levels of gameplay intensity so consistently, and so the posture system isn’t a one-size-fits-all solution that can be injected into every action game without adaptation considerations or consequences. I would like to see more systems like the posture system see popularity, but it is a very delicately balanced system that would demand a lot of care in the implementation. Just plain old HP is much simpler, and it’s been the de facto way to represent combat in so many games for so long for a reason. As I’ve said before, the level of abstraction it creates can be an acceptable tradeoff if it’s supported by a strong combat system, and the simplicity of it fits your game. Another aspect of Sekiro‘s posture system is not only the gameplay intensity, but the intensity on the player’s hand in controlling the thing. The system demands a lot of very rapid and often repetitive inputs that may not be suited for all players. It’s an obvious drawback, perhaps not inherent to a posture system, but correlated to how it’s implemented, and so the system would likely need some fundamental changes and re-tuning to map to a more inclusive suite of control options.

I think the posture system is a great look at how we can rethink the traditionally established rules of how games need to be. Creating new and exciting interactive experiences means being willing to accept that no one game mechanic can be sacred, even if it is ubiquitous. HP bars are so universally standard that any deviation from that mold feels almost alien. The posture system picks apart what HP is meant to represent and repackages it in a way that is extremely conducive to the kind of ninja-action experience Sekiro aims to create, in a way that is evocative of the real, even if it is still very abstract. Sekiro‘s posture system isn’t purely realistic, it’s not what sword combat actually looks like, but it feels very real to experience. It feels like pitched life-or-death battle with high stakes, and real tension.

A ninja man in a red coat and scarf clashes swords with an old woman ninja dual wielding kunai. She jumps from a hidden wire connected to the walls of a crypt-like cellar. Her attacks are deflected repeatedly by the red-coat ninja before losing her stance and being stabbed.

My Lord, I Have Come For You. This… Will Only Take A Moment…

Dark Souls 3’s Brigand Twindaggers or How I Stopped Worrying and Learned To Love Status Effects

It is February 2022, my dudes, and you know what that means. Elden Ring is around the corner and there was absolutely no way I wasn’t going to just talk about Fromsoft games all month. I want to share an interesting experience I found while playing Dark Souls 3 as it pertains to a specific weapon found in the game, the brigand twindaggers. This is less a breakdown of this weapon’s moveset or particular attributes and more a little anecdote about how the daggers won me over once Dark Souls 3 had sold me on using status effects on my weapons, and the subsequent analysis that followed.

In RPGs with character customization I love to bring in my own stock of characters to populate the world, and there are some old standbys I revisit frequently, such as a thief who wields a pair of daggers, whom is often my player character in Fromsoft games. If at all possible, I will deck out my characters in their appropriate gear. I kind of like the RP side of RPG that way.

In Fromsoft’s Bloodborne, an action RPG that preceded Dark Souls 3, there is a weapon called the blades of mercy, a sword that can transform into a pair of daggers. I often refer to this weapon as a lawnmower – it absolutely shreds enemies to pieces with a massive amount of damage output. It scales incredibly well with Bloodborne‘s equivalent of the dexterity attribute, and its moveset is a flowing, seamless series of rapid strikes. Its the most satisfying to use pair of daggers in any of Fromsoft’s games to that point, so I had high hopes and expectations when it was revealed that Dark Souls 3 would have, as a new feature, paired weapons – weapons that come in a set, made specifically for dual wielding. I went over the game with a fine-toothed comb when I got it, clamoring to find a pair of daggers. Surely if paired weapons were a thing, I’d find dual wield knives.

And so I found them, and it quickly became difficult to contain my disappointment. Compared to Fromsoft’s previous outing with dual wielded daggers, these brigand twindaggers had a slower moveset with a much longer startup, and tremendously pathetic damage by comparison. They didn’t even scale that well with dexterity. I tried to like them, and tried to use them throughout my first play-through, I really did. In the end, there were just so many better speedy weapons, whose damage scaled so much better, that I could really not justify using the daggers any longer. I ended up embracing my inner edgelord and used a paired katana and wakizashi, two japanese style swords, as their damage output was insane compared to the daggers.

I had a blast on my first playthrough of Dark Souls 3, but I always regret not making a player build I was satisfied with involving the daggers. It didn’t feel true to my player character to not have him using knives, and I wished the knives were better. They didn’t really need to be though, after some experimentation, I would discover I just needed to change my approach. It started when I saw a player versus player showcase of the brigand twindaggers. The very skilled video author was destroying human opponents, seemingly with ease, utilizing the weapon I had condemned as largely useless. His secret? The daggers were enhanced with a bleeding effect. The Dark Souls series has always had weapon status effects, special attributes that can be applied to weapons to make them debilitate enemies in specific ways, applied if enemies are hit enough times rapidly. Primarily, this takes the form of bleed weapons and poison weapons.

On two pedestals, side-by-side, sits a jagged stone covered in a shiny oozing red liquid, dribbling onto one pedestal. On the other, a pair of curved knives crossed over each other.
Like peanut butter and jelly.

Enhancing a weapon in Dark Souls always modifies its base damage and damage scaling in some way. Status effect weapons on the whole tend to deal a lot less base damage as a tradeoff, and so I’d often shy away from them not just in Fromsoft games but in RPGs in general. It just felt like it was an unnecessary extra step, compared to simply dealing more damage directly. But I wanted those daggers to work, so I gave it a try. In Dark Souls bleed is a status effect that builds up by hitting your target repeatedly, and when it’s built up completely, the victim loses a large chunk of their health to a hemorrhage, all at once. It’s a rather cool mechanic that gives the player a smaller micro-goal to achieve while fighting enemies, that is, quickly building up bleed, in addition to just fighting. It makes for an interesting playstyle and when I tried it out, I found I was having a ton more fun than before. Even when modified, the daggers still deal similar damage to their sharpened variant, and yet now acted as a powerful poison or open wound delivery system.

An undead wrapped in tattered garb thrashes two daggers at a frostbitten undead ghoul, in a snowy medieval city. The ghoul gushes blood as they are struck. After several hits, the ghoul's health indicator suddenly takes a large amount of damage.
See how quickly enemies vulnerable to bleed pop like balloons? It’s a great time, all round.

Status effects in Dark Souls 3 just work. Astounding. But why is this such a pain point for me in so many other RPGs? What is it about Dark Souls 3 in particular that makes it work? I think I’ve identified a few factors that majorly contributed to my enjoyment of using bleed and poison variants of the brigand twindaggers. First off…

It Works

Yeah okay so this one is a little self explanatory. Players aren’t likely to use a game mechanic that doesn’t work, obviously. It goes deeper than that though, players aren’t likely to use a game mechanic that isn’t effective. Every enemy in the game could be vulnerable to bleed, but if it only did a piddly pathetic amount of damage nobody would care enough to go that route. Thankfully bleed is very effective, and can often kill enemies even faster than raw damage. It was also seen fit to make nearly every enemy in the game vulnerable to bleed, a very wise decision. Some are resistant to it, some weak to it, but only a handful are completely invulnerable to bleed. Was this point even worth mentioning? Yeah I think so. Each of these points is something I’ve seen failed in many many games before. There are tons of games where status effects are simply unreliable to the point of near-uselessness. What good is a poison effect if it takes a dozen tries before it actually sticks? What’s more, what is the point if the poison is super hard to apply, but it barely does anything as a result? By then I could have just beaten by opponent to death with a stick. Floundering around with weak status effects feels terrible, and they need to be at least as viable as the more straightforward option.

It Works On Bosses

I cannot stress this one enough. Nothing will make me drop a combat mechanic which requires a time investment more definitively than seeing it is ineffective against boss encounters. Often in combat centric games bosses are the height of the combat system, pushing it to its limit where the most fun to be had is, or even the central axis about which the rest of the gameplay turns. If a combat mechanic breaks down in a boss fight, as a player I often feel as though it’s not worth my time. Status effects work on bosses in Dark Souls 3, generally, or at least frequently enough that I never find myself despairing at the futility of using them.

It is so strange to me that so many RPGs see fit to make bosses immune to status effects. On the one hand I can see the perspective – status effects tend to be very powerful in certain contexts, especially when they are not direct damage dealers, like disables or other utility effects, and one does not want to trivialize combat encounters. And yet. If one has been relying on a certain game mechanic, they begin to take ownership of it as a playstyle. They feel clever or powerful for utilizing it. Taking it away at the most crucial encounter feels awful. There are ways to design around the brute force method of just making bosses immune. Perhaps bosses are merely resistant, and incur a diminished form of status effects applied to them. Perhaps bosses have the ability to remove their own status effects under the right circumstances, or perhaps they last less time. Perhaps status effects are balanced as such to simply be generally useful, but not overly powerful against bosses. I think it’s rarely ever wrong to let players think around a problem, and removing a strategic tool such as status effects from their arsenal, when they can be employed elsewhere feels like player punishment.

In Dark Souls 3, applying a bleed effect deals a flat chunk of damage to enemies, usually enough to kill lesser foes. On bosses, it’s merely a nice step toward their defeat, but not an utter showstopper by any stretch. For bosses that may be felled too quickly if they are bled out repeatedly, it was decided they would be resistant to bleed effects. You can still get that extra damage, and it’s not that hard to do, it just takes a little longer, and the balance is kept that way. There are some enemies and bosses which are immune to bleeding, but not nearly enough to make me question the status effect’s efficacy, and what’s more it contributes to the overall fiction of the game, which has to do with my next point.

Strong Feedback

Dark Souls 3 has very strong audio-visual feedback for when you’re wailing on an enemy. Blood shoots out in exaggerated sprays along with a crunchy *squelch*ing sound with each strike of your weapon. Against armor, you can hear the rattling clang of steel on steel. This is obviously good design from a gamefeel standpoint, but it also provides the very useful advantage of illustrating what can and cannot be inflicted with a status effect. Hitting stuff that can be bled tends to use that exaggerated blood graphic I mentioned, but things that are resistant or immune will show less blood when struck, or none at all.

A well-armored women swings a wrist-mounted blade at a cage full of reanimated corpses. Blood shoots out when it is struck, and after several hits, an explosion of blood gushes from it.
A cage full of reanimated corpses? Can the cage bleed? Kind of ambiguous, except that it shoots blood off when hit. Okay so it can be inflicted with bleed!

I’ve mentioned that some enemies are immune to status effects and how that enhances the fiction of Dark Souls 3. What I mean by that is, it is effectively intuitive what enemies can and cannot bleed. Bulbous fleshy beasts, dripping and shambling undead, living creatures. Things that obviously have blood, are all vulnerable to the bleed effect. Things like enchanted empty suits of armor, a giant tree, skeletons. These things obviously do not have blood, and thus do not bleed. It seems like a simple trick not to miss, but yet again I’ve seen this very concept done poorly too often. Consistency is key. The player shouldn’t have to guess, or at least not guess blindly whether or not their combat tools will even work. Obviously you can’t bleed a skeleton, but obviously you can bleed a giant rat. Design the game so players can trust their own eyes and ears, and the play experience will feel much more seamless. Immunities and resistances should have logical reasoning grounded in the rules of the real world, even if the game takes place in a fantastic one, so your player has a hint of familiarity with which they can decipher the rules of your game.

An undead wrapped in tattered garb thrashes two daggers at a giant rat with exposed wounds. The rat gushes blood as it is hit, and quickly keels over dead.
An undead wrapped in tattered garb thrashes two daggers at a cloaked skeleton in an dank catacomb. The skeleton takes damage. but the hit effects against it are dull and understated.
Giant fleshy rat with exposed red meat? Blood pinata. Spooky skeleton? Not so much. Notice the understated hit effect on the skeleton vs. the huge amount of blood on the rat.
We’ll have to come up with some other clever solution to deal with the skeletons

Conclusion

So I guess my takeaways from this experience are twofold: certain weapons can be satisfying to use in how they fulfill certain gameplay niches. The brigand twindaggers are an excellent status effect tool in how they apply effects quickly through rapid hits. My other takeaway is that a lot of games could do status effects in a much more satisfying way that makes them feel powerful and useful, something a lot of the designs I’ve seen are often too bashful about. They can be a viable alternative gameplay style all their own, you just need to put in the legwork to make sure this gameplay style feels strong and effective. Locking it out of boss fights makes it feel like a lesser, illegitimate gameplay style, an afterthought. Players should be able to discern the applicability of status effects with audio and visuals alone, without having to consult a wiki. Overall, I think status effects can be underappreciated in games mostly because they so often could be implemented better. When games get it right, I think it’s worth giving a closer look to see exactly what went right. Status effects in Dark Souls 3 were fun enough to use, and strong enough to completely reverse my opinion of an entire weapon’s implementation.

An undead wrapped in tattered garb thrashes two daggers at a giant armored mage wielding a flaming staff. in a snowy medieval city. The mage gushes blood as they are struck. After several hits, the mage's health indicator suddenly takes a large amount of damage.

Such weapons inflict lacerating damage. Most effective with sharp or spiked weapons…

Boss Breakdown: The Ender Dragon

I had one of those weeks. You know the type. An update for worldwide phenomenon Minecraft, survival-based multiplayer sandbox game in which all the world is cubes that can be deconstructed and rebuilt as the player desires with enough energy and time, came out recently. So some friends decided to boot up a new multiplayer server, and there goes all of my free time for the week. We just accomplished Minecraft‘s nominal win condition for its survival mode, the condition which triggers the game’s strange credits sequence; killing a creature known as the Ender Dragon.

Minecraft is a game that, by its very nature does not lend itself to a win condition at all. Even in survival mode, at its core Minecraft opens itself to players defining their own goals and aspirations. Perhaps you wish to build a tall, imposing tower, or a glass city under the sea. Maybe you just want to explore and find beautiful landscapes, or dig into an enemy stronghold and steal their treasure. Maybe you want to raise animals, or start a zoo, or automate the production of gunpowder. All of these things are possible in the game, and none of them really have anything to do with, nor contribute to, the process of killing a dragon, at least not directly. Minecraft isn’t really a combat game, you see. Sure, it has combat in it, and there’s even some nuance to how it works, but it isn’t exactly a challenging set of deep mechanics one is naturally predisposed toward mastering, especially when Minecraft is so rewarding towards its other forms of play, such as digging, landscaping, crafting, and building. All that aside, it was decided Minecraft would have an overriding game-defined goal.

So met with the quandary are we, how does one design a boss for a game in which, A: combat is barely involved, and B: is primarily a game of non-combat mechanics like traversal, building, and uh, mining and crafting. The Ender Dragon is an interesting example of this idea in practice. There are a lot of cool ideas the design of the Ender Dragon utilizes to make it interesting within the context of Minecraft’s unique suite of sandbox game mechanics. If I were to make an appraisal of its overall success in this regard, I’d say the Ender Dragon is… okay. It is at once somewhat convoluted yet also simplistic, and more than a little janky. I’ll get into a little more detail, but overall the fight is fun, and though I don’t think it is the utmost pinnacle of this idea, it is one of the most popular – Minecraft is played by millions, and is in a unique position to examine this idea of boss fights or boss-fight like gameplay engagements for games that are not primarily about combat.

From a first person perspective, a person stands atop a high obsidian pillar, while shooting arrows at a black dragon made of cubes, who flies and circles in the distance. Another minecraft guy stands atop a second nearby pillar and shoots fire arrows at the dragon.

The Ender Dragon encounter is primarily a player-directed boss fight, meaning that the pace and rhythm of the fight is directed by the player, rather than by restricting rules of the boss’s environment. The player directs their interaction with the environment, rather than the other way around. The Ender Dragon is a large black winged beast that breathes fire and flies around an isolated island in an empty void. Her behavior consists of several states which she cycles through, with a preference for certain states over others depending on the status of the structures surrounding the outer edges of her boss arena, called End Crystals. The dragon will dive at players occasionally, and being buffeted by her wings can send players careening off into a lethal fall if they’re not careful. Whilst circling above, the dragon will launch fire balls that leave pools of deadly fire-like substance that persist for some time. Minecraft does not have extremely robust combat mechanics as I mentioned, and dodging usually just means moving to where the danger is not. The fire pools can be irritating, still, them lingering encourages players to dig and build around persistent obstacles.

From a first person perspective, a person digs into the side of a stone wall with a pickaxe, then further up to reveal purple sky above the dug hole. Outside the whole, several blocky minecraft guys scramble around a pool of purple fiery gas.
Burying yourself underground is a viable strategy, like some sort of gopher or marmot

So essentially what’s going on, is that these Ender Crystals want to be destroyed. Some can be sniped with projectiles, but some have protection that needs to be dug through, and to do that you need to build a scaffold up to their elevated location. If you don’t destroy these, the dragon stays in the air and restores her health. The design is suggesting the player utilize their building skills to solve the problem, which is good. Building is one of Minecraft‘s most fun and robust features, so it’s good to lean into for a boss fight. It’s a little straightforward though, there isn’t much developed knowledge of building technique needed to build up to these crystals. What’s most restrictive is simply the intimidation factor of the dragon, and outside of that there’s not much to it, as far as the end crystals are concerned.

Another strange quirk of this fight is the presence of many Endermen monsters in the area. Endermen are some of the strongest, most deadly, and most aggressive enemies in the game, but only when provoked. They can be provoked by attacking them, or by centering one’s camera on theirs. The Enderman’s relatable aversion to being perceived translates to a need for avoiding eye (camera) contact with them. I can’t speak too much for others’ strategy, but I always find myself look down at my feet during this encounter, as though my character is just really embarrassed to be there. It’s strange challenge – having to be wary of where your camera is pointing, not just where you character is standing. It does fit with the dragon being an Ender Dragon, as the Enderman monster, which can be encountered outside this fight, and isn’t all that uncommon in the overworld, always abides by this eye-contact rule. It does give the player another plate to spin, but it may feel claustrophobic and a little artificial to some. I can’t help but wonder how this fight might be improved if the designer can rely on the player having full freedom of motion with their camera.

From a first person perspective, a person shoots a bow and arrow at a black dragon made of blocks, flying around in circles below. The person stands atop a tall obsidian pillar. In the bottom left corner of the display, a dialogue states that several allies have been killed by Endermen.
The Endermen live up to their name

One of the things I find most commendable about the design of the Ender Dragon is how open to various forms of problem solving it is, and how it accounts for the many methods a player might used to dethroning it. Minecraft offers players a huge number of tools with which to solve problems as they explore the world. For instance, the dragon usually is flying around the sky, out of reach of Minecraft‘s most common weapon, the sword. It can, of course be shot with range weapons such as arrows, but the dragon will also frequently land to rest after the End Crystals are dealt with, making it vulnerable to melee attacks. This is one solution. Since the dragon will land and stay stationary for a time, it’s also possible to load up her roost with explosives and set them off like a trap.

From a first-person perspective, a person runs up to a roosted Black dragon made of cubes and places a white bed on the ground, which then promptly explodes in their face, catapulting them into the sky.
Yup. Working as intended.

The dragon’s fire can be dangerous to those walking across her arena’s surface. Of course, Minecraft, being a game about building and digging, it’s totally possible to dig a safe series of tunnels under the arena to move around, or build up cover to protect from the dragon’s fire balls. Climbing up to reach the End Crystals poses another challenge, one that can be tackled in several ways. Shooting the crystals with projectiles is possible, but some are encased in shields that need to be removed. Building up to them with blocks is an option, or placing water (which can inexplicably be swam up in Minecraft) also works. Gravity is an ever-present threat in this encounter, so it’s wise to have some way to address it. Careful use of water can break any fall, or enchantment to one’s boots renders great falls less lethal, or you can drink a potion to fall more slowly. The point is, in Minecraft there are a myriad of ways to approach any given problem, and the dragon neatly accommodates nearly all of them.

From a first person perspective, a person is falling down the side of a tall obsidian pillar. As they approach the ground, they dump a bucket of water onto the side of the pillar, whose contents miraculously breaks their fall. The person then eats a slab of steak.
The tools of dragonslaying: steak, a four poster bed, and a bucket of water

The existence of the crystals is evidence of this. They exist apart from the dragon, at stationary, elevated points. If you want to destroy them up close, they require you to reach their elevated location, likely with building. Building during a dragon attack is a pretty compelling idea, and having to build up to your target seems like a natural fit for a boss fight in Minecraft. There is some inherent risk and reward to how one approaches building up to them. You can build straight up, or try for a more cautious staircase, or perhaps build up to one crystal, then create a bridge to the others.

Another advantage of Minecraft‘s immense suite of problem solving tools is how it frames the fight against the dragon. It is a difficult undertaking, even with a group of friends cooperating. There are a lot of hazards to look out for, so preparation is key. Minecraft is, in a lot of ways a game of preparation. Each night, hostile monsters will appear and overwhelm the unwary, but a solid roof over your head, and an ample supply of food will get you through. Bigger challenges in Minecraft harbor greater risks, and protective enchantments, armor, potions, special food, animal companions, building materials, etc. can all lighten the load. There’s a special sense of camaraderie when all of your buddies convene at a central location to swap enchanted bows and arrows, high-quality armor, and buckets of water in preparation for a great adventure. The Ender Dragon’s accommodation for all these varied modes of tackling problems is one of the best parts about it, regardless of other shortcomings. It’s the context and framing of this boss fight, I think, that creates its greatest moments.

Blocky minecraft guys in armor run around an enclosed space made of gray bricks. There are boxes, chests, and supplies piled around the room. One guy runs up a staircase at room's center and repeatedly crouches and stands.
Just messing around is truly the essence of this game

Whether Minecraft players wanted it or not, the Ender Dragon is the unambiguous overriding goal of a Minecraft ‘playthrough’, its final boss. Is the Ender Dragon a culmination of the skills and knowledge a Minecraft player accrues during their journey to the End? Well, kind of. The levels of preparation and supplies gathering players can engage in is huge in scope. There’s all sorts of stuff you can gather and build to prepare for this fight, and gathering/building is one of the most fun things you can do in the game. In my opinion, the experience is only heightened when a group of friend communally pool their resources to maximize the chances of success. The fight itself is a little convoluted, with the disparate Ender Crystal targets, apart from the dragon itself, but also somewhat dead simple in the very limited suite of behaviors the Ender Dragon can actually engage in. Having to utilize building or similar methods to reach the Ender Dragon’s crystal weak points is compelling and fits with Minecraft’s gameplay, although it’s a somewhat blunt and simplistic implementation. I find the Ender Dragon’s overall design commendable in a lot of ways, and I do look forward to the challenge as a fun communal activity with friends, but it feels under-designed in several ways. Sometimes putting a dragon boss fight into a game like Minecraft feels like hammering a square peg into a round hole. Still, I think there’s a lot to praise about and learn in regards how to approach these sorts of challenges in games that don’t primarily utilized combat as a central game mechanic.

A black dragon made of cubes slowly rises into the air as it disintegrates, a purple light emanating from it, then it explodes.

And the game was over and the player woke up from the dream…