Robin Vilain
videogame designer
author of commotions
créateur de jeu
Blog
Blog
Blog

Hunter’s pistol

I finished Bloodborne two days ago, and very much liked it. It’s an interesting game in many ways, some of which I’ve been wanting to write about since I started playing. I’ll eventually talk about the story – I’ll let you know when I start venturing into spoiler territory, in case you haven’t finished it yet. Before that, though, I’d like to have a look at the game’s most advertised new feature, its main differentiator when compared to its predecessors: the gun mechanic.

When you start the game, you’re given two and half tools for killing: one fairly large cold weapon that has two equally murderous forms (you can usually trade swiftness for reach and/or brutality) and a firearm. When it was revealed that this shooty, noisy device would go into the protagonist’s left hand instead of the expected shield, many players were surprised, worried even; I know I was. The safety provided by holding down the left trigger to hide behind your shield in Dark Souls (the previous game from Bloodborne’s developers) allowed for a slower, more measured play style, and that was exactly what I loved about it. Bloodborne, on the other hand, seemed to be all about constant side-stepping and frantic shooting – twitchy, nerve-racking, focused on precise timing, and indeed without a moment to rest. There is a shield you can equip in Bloodborne, but its in-game description reads like a design note from director Hidetaka Miyazaki, directly addressed to the player: “Hunters do not normally employ shields, ineffectual against the strength of the beasts as they tend to be. Shields are nice, but not if they engender passivity.

What’s interesting about this note is how it almost explicitly reveals why guns have replaced shields in Bloodborne. The developers felt there was a problem with Dark Souls that they needed to fix – passivity – and the firearms were their solution. Let’s fictionally go back in time and pretend we’re attending the early design meetings for Bloodborne… it probably went something like this. The starting point was the removal of the shields; that much is obvious. However, shields did come with a mechanic that didn’t encourage passivity at all: parrying. To parry in Dark Souls, you opt not to keep your shield up, and instead try to raise it right when an attack is about to hit your character; it’s risky (half a second too late and the attack does hit) but if done successfully, it staggers the enemy, leaving them vulnerable to a powerful riposte. That’s the kind of gameplay mechanic that requires so much focus and concentration that you can usually feel it affecting your whole forearms, as they tense up waiting for the exact right moment to press that button. It’s anything but passive. How, then, do you keep it without the shields? What kind of combat-oriented tool could complement a main weapon and allow instantaneous parrying, but not passive protection?

The answer’s definitely not ‘guns’. Daggers, maybe, or some sort of steel gauntlet… but not guns. That doesn’t matter, though; let’s call this mysterious parrying device the Tool X for now.

Meanwhile, in a different meeting room, here’s what designers are struggling with: bows. Dark Souls featured bows which you could aim freely; as a result, players would find spots in the game’s level from where they could shoot arrows at unexpecting enemies. Videogames being what they are – simulations that rely entirely on flawed programming – if the player’s character was far enough, said enemies would fail to register their existence at all, despite having just received an arrow in the chest. Players took advantage of this and defeated monsters meant to be extremely challenging opponents, simply by standing just far enough and shooting hundreds of arrows at them. In other words: they found a way to beat the game while staying passive. Obviously, this would not do.

Bows, then, also had to go. Just like shields, though, they did have their usefulness: especially when facing another player, the possibility of them taking out a bow kept you on your toes, even at long range. Finding an alternative that avoided the aforementioned pitfalls seems easy enough: you need a certain Tool Y that can hit from a distance, but not too far.

Let’s recap. Tool X is very useful, but it’s only used in one very specific situation: in close quarters combat, when an enemy is about to attack. Tool Y is also very useful, and it’s also used in only one context: when engaged in combat against a foe that’s out of range of the main weapon. The genius of Bloodborne’s designers is that, looking at these two absolutely unrelated issues, they realised that Tool X and Tool Y could be one and the same, since their usage didn’t overlap. There would be no circumstance under which both would be needed simultaneously; which meant they could be grouped under one label, associated to a single input from the player. The question then becomes: what can hit foes from a distance, is quick to operate, can stagger a vulnerable enemy, and provides impactful feedback when doing so? A gun. That’s why I love the mechanic: in the context of Bloodborne, it makes perfect sense. It elegantly solves many issues at once, and efficiently communicates with the player: the weapon’s booming noise suddenly becomes invaluable when you’re in the middle of a tense face-off against a boss. Each press of the trigger results in a loud bang that supersedes every other piece of information generated by the game, to reach the player’s senses through a high-priority lane; if that shot also happens to be a successful parry, the gun’s noise is accompanied by an even louder, heavier audio cue. As a player, you learn to identify those sounds, to the point where the appropriate reaction almost becomes a reflex – which is exactly what Bloodborne, with its unrelenting hostility and difficulty, demands of you.

And in the same way that the game requires intense focus and accepts nothing short of perfect timings, it itself shows incredible focus and perfectionism. The same philosophy that probably led to the firearms can be found in all of its mechanics, in an incessant bid to eliminate passivity: most notably, the ‘rally’ feature, which allows a player to regain some of health lost after being hit by immediately hitting back. When, in previous games, you would retreat (or hide behind a shield), wait for a safe moment, then drink a potion, here the best strategy is often to strike back right away – although, as you can imagine, that exposes you to more attacks, more risks. Again, the stakes are raised, the tension is cranked up a notch. What I find even more interesting, though, is the more subtle side-effect of the main weapon’s transforming abilities. As I said earlier, it can switch between two forms – one meant for quick and agile movements, and the other one for more heavy-hitting blows. (It’s worth noting that one also bridges the gap between close quarters and long range: between the cold weapon’s two forms and the gun, there’s always one tool to hit an adversary, no matter where they stand in relation to the character.) There are three primordial aspects to the transformation process: it only takes a single button press, it looks cool, and it can be done while the character’s moving. Combined, these characteristics mean that most players probably keep pressing that transformation button for no particular reason other than to keep their fingers busy; the effect is satisfying and comes with no penalty. And just like that, even in the mundane act of traversing an empty level, passivity is vanquished.

If anything, Bloodborne’s too focused. (If you don’t want to know anything about the story, avert your eyes now!) By the end of it, I was growing bored of its insistence on exploring increasingly dark, morbid, gruesome themes; not once did it stray from its initial stance. It only kept on diving deeper and deeper into madness and horror, evoking Lovecraft’s works more and more – to the point where it seemed like Yharnam was From Software’s interpretation of Arkham. That’s my one complaint with the game: where Dark Souls kept a constant hint of wondrousness in its tone and settings, and could surprise you at any moment with majestic landscapes and the nagging sense that you were an intruder upsetting a melancholic harmony, Bloodborne lets you know from the get-go that there will be no respite. It can be hard to keep on slashing your way through endless frenzied beasts, when you know that more gore awaits beyond the next door, that the next environment will be yet another bleak delirium.

And still, you do persist, because you know persistence and stringency is what it’s all about. Bloodborne presents a story, but it doesn’t force it upon you – far from it. It asks you to get it yourself. It merely complies with your inputs, as a neutral performer; press O, and your character rolls to the side. That roll is always the same: it takes a very specifically tuned time to be executed, makes your character travel by a very precise distance. Shoot at the exact right time, and your enemy’s animation is instantly interrupted, leaving them vulnerable for a definite amount of microseconds. Perform a counter-attack: see the counter-attack animation play out. Always the same. Contrast that with the videogame incarnation of Batman, for instance: a single button never does the same thing. Press it repeatedly, and the character jumps in all directions, throwing kicks randomly selected by the game from the pool of possible kicking moves; if the conditions are met, Batman uses the décor to beat down criminals, pushing them through windows or throwing them against walls. If you successfully trigger that game’s counter-attack equivalent, you never see the same animation: Batman may dodge to the left, or he may jump above his enemy, or he may block a blow with his arm. Only the result is the same (the attacker’s knocked out). Batman’s game (like countless others) interprets the player’s inputs to create a choreography, but it comes at the cost of precision and player-driven behaviour; it is spectacle, and the player’s the audience. Bloodborne rejects that notion by leaving all of the character’s actions entirely up to the player, without any filter to smooth them out or make them prettier: any behind-the-scenes processing would take away from the feeling of accomplishment. ‘Rigour’, Bloodborne seems to be hammering home. ‘Strictness’. As you realise that the game expects you to adopt its austerity for yourself, it all becomes clear, and you embrace the authors’ crusade against passivity. And as you contemplate how deep into darkness you still have to go, one thing does become clear as day: giving up would be quite a passive option, wouldn’t it?

Work
Work

Hi! My name’s Robin. I’m super passionate about videogames.

I think they can be meaningful on many levels – from the authors' ideas to the players' interpretations, from the intimate harmony of mechanics echoing each other to the booming contribution to greater social or cultural contexts.

I love games that are deep, inspiring, yet still engaging enough for the players to seek out and even analyse these meanings. When the overarching purpose shines through, you know you’re playing something unique and beautiful.

This is what I aspire to create: experiences that make players stop and think. Below is some of my past work, in which I've tried to put this into practice.

Logo JMC

At JMC Academy, I taught game design and development, mentoring students as they worked on their final project: pre-production, planning and management, production, promotion. This also involved teaching them how to approach design critically and meaningfully, rather than conventionally or arbitrarily. I also supervised prototyping and programming classes focused on getting a better understanding of Unity and C#.

Teaching was an opportunity for me to articulate and expand on my design aesthetics, and to introduce students to kinds of games and interactive works that they’d not heard of before. Seeing their horizons broaden was truly rewarding and led them to experiment with new ideas, while still maintaining a sense of ownership and uniqueness. This was a group of people with different outlooks on games and different goals, who taught me as much as I taught them. I encouraged them to cherish their own perspectives, to cultivate their own individuality, and to imbue their projects with it.

Logo Leda

During the same period as when I teaching at JMC, I also worked with Productive Procrastination on Leda, a web+mobile platform for teaching ethical leadership practices and empathy, with a focus on inducing long-term positive behaviour change. While I was initially brought on board as a game designer, my role quickly evolved to include development, art direction and management.

This project was a massive undertaking backed by a university and several major companies. It was crucial to balance the needs of those investors with our mission. I strove to design games that were accessible to anyone (including people usually put off by them), and that included Leda’s principles as intrinsic play elements. This approach then extended to the scripts and storyboards of animated shorts, the motion design of interview videos, and more broadly the UI and UX of the whole platform; all led by the push for consistency and pedagogy across all of the content.

In 2015 and early 2016, I worked on a variety of playful live events: murder parties, pop-up escape games, historical investigative mysteries… I also designed and/or consulted on small-scale videogames and board games. During this time, I focused on how to better introduce play concepts to different audiences, and analysed reactions on more experimental projects. As part of my push to broaden my skill set, I designed and developed several websites (including this one).

This exploration of new fields and desire to step out of my comfort zone led me to move to Melbourne in April, where I was welcomed by the local game dev scene. I continued progress on personal projects and started working at Mind Games, a board game store, which provided yet another opportunity to get a better understanding of the kinds of audiences interested in games – from experts to newcomers.

Logo Innovation

Innovation: Age of Crafting is a mobile game that was recently released on Windows Phone, and will soon be out on Android and iOS. Tiles representing scientific and cultural landmarks of humanity’s evolution through the ages are arranged on a grid; the player must slide them to combine them, discovering more advanced technologies.

I worked on the game as a freelancer; when I arrived on the project, only the basic concept had been determined. I fleshed it out and expanded it, then designed the game’s rule variants and all of 125+ levels they’re used in, aiming for a constantly renewed experience and a welcoming learning curve.

Logo The Crew

In 2013, I worked as a mission designer on The Crew, at Ivory Tower. I collaborated with the environment artists to find beautiful locations within the game world, then defined mission rules that were adequate for the player’s level. I tried to subvert the existing mechanics, to suggest emergent narratives; I took advantage of the game’s features, such as off-road driving, but was careful not to be confined to them either.

I also worked on the exposition of the game’s structure, which had never been seen before. The tutorials and informative screens had to effectively teach the player how to navigate the many systems, without being an obstacle to free-form play. Predicting the player’s progression and timing the delivery of explanations accordingly was crucial.

Logo Rainbow Six

My post-grad internship was at Ubisoft’s Editorial, which oversees the development of the company’s games and ensures a high level of quality, providing advice and feedback to the dev teams – I was assigned to Rainbow Six. I studied the portrayal of crowd panic and tactical squad interventions in films and games, and derived key points and guidelines. I also played each build and offered suggestions on how to make the experience more cohesive and intense.

While I focused primarily on Rainbow Six, I also intervened on other projects, among which Far Cry 3 and Splinter Cell: Blacklist. I helped balance the multiplayer modes of both games; on FC3, I also played the whole game several times to give feedback from an external point of view, and specifically whether the introductory sections properly explained every aspect of the game.

Logo Flux

FluX is my final school project; it’s a 2-player digital board game, played on Microsoft’s PixelSense touch table. Each player owns a base that releases ink; by placing wooden pawns on the screen, they create currents and obstacles. The aim is to lead the ink towards neutral bases scattered on the board to convert them into new ink sources, and then to capture the opponent’s main base.

My contribution was quite broad: I elaborated of the main concept, designed board layouts, programmed a vector field and fluid mechanics to handle ink propagation, and implemented the pawn identification system – using PixelSense’s virtual reality API within Unity 3D. This project started out as an art installation motivated by a question: how to make human-human interactions through a digital medium feel natural, transparent and even physical?