Tips from a combat designer: The art of boss design

[In this reprinted #altdevblogaday in-depth piece, Vigil Games' Mike Birkhead ooks at how movie boss fights satisfy audiences more consistently than games, and presents tips for designing game bosses.] Boss fights are the linch pin of your entire combat experience, possibly your entire game, as one bad experience is all it takes to bring your game to a screeching halt. Something so important, so integral, demands effort and attention to detail, but for the majority of projects they are relegated to the very end. Why? I was recently posed a question: why are movie boss fights superior to game boss fights. The question was framed through the stance that movie fights are, in general, more visually (and I would add, emotionally) satisfying. Yet I am sure, at this very moment, you are conjuring up your favorite video game fights; and, chances are, you are well on your way to skipping this entire post in order to rail me with examples – just hold up. If I took the stance that all video game boss fights are bad, then that would be an untenable position, yet that is not what I am saying. What I'm saying is that movies are just so damn good; more importantly, they achieve their greatness with higher consistency, which is what we want, isn't it? So, after much thought, I realized we had the wrong question, and the real question is thus: how do movie boss fights satisfy the audience with greater consistency than games? A difficult question, but one I think we can answer. The journey starts by learning what movies are trying to accomplish. Movie boss fights

Movies are a visual storytelling medium. Stating the obvious, I know, but the path to unraveling their consistent mastery starts by first understanding their intent: tell an honest story. Cinema always strives for asymmetry; meaning, movies want visual diversity. A movie must not repeat itself — visually, or otherwise — because repetition is boring, and a bored audience is a lost audience. More than this, though, is the fact that opportunities to deepen, or enhance, a movie's storytelling in a visual manner must be taken — MUST. A character cannot simply "overcome" a situation. No, he must act in a way that is honest to his character, or the illusion is lost.

Clip: Big Trouble in Little China, Jack Burton vs Lo Pan

There is a reason that this has the title "Quite Possibly the Best Scene Ever Filmed," and I would add, best boss fight ever filmed. Jack Burton's character acts in a way that is completely honest to his character: both overly confident and utterly hopeless, yet still managing to get things done in a bad ass way. Compare and contrast that with Egg Shen's battle with Lo Pan earlier in the movie.

Clip: Big Trouble in Little China, Egg Shen vs Lo Pan

Ignore, for a moment, that the two battles are completely different in terms of execution and focus on Egg Shen's battle. Notice that Egg Shen controls his avatar through a crystal, whereas Lo Pan does not; both do mythical battle, but their methods of evocation are visually different. As you can see, these fights with Lo Pan provide mountains of information about people involved, and the net result is that you know these characters. Stomping out boredom and cliche from your movie is about being honest to your characters, but it is also about defying the viewer's expectations; or, what Robert McKee would call Defining the Gap. This is an important point to understand, because boss fights are the ultimate gap — and, when combined with asymmetry, defines the complete picture of a movie boss fight:

Movie boss fights are so satisfying because they take this list of three goals and do everything to maximize their impact; which is to say, they maximize their visual storytelling. Cool camera angles, bad ass stunts, the hugest of explosions, witty lines, and unexpected turns of character are all paths to this goal. The writer, director, composer and cinematographer do everything they can to make the boss fight as mighty and honest as possible. What does all of this have to do with Games? Well, games need all of these things, too, but they have their own concerns that change the dynamic. Games vs movies Participation is our double-edged sword. It is our greatest strength, yet it is also our greatest weakness. In the right hands, participation enhances our attachment to the game, for we realize that we are not being told a story, but instead becoming a story. However, if you treat participation like a crutch, then, like a stubborn child, you will slam that square peg into that round hole until the player's interest shatters into a million pieces. Instead, you must embrace participation, and to do so means motivating the player — either extrinsically, or intrinsically — while avoiding two major pitfalls. Compulsion and extrinsic motivation This is the dark side. Its power is great, my friends, and it will tempt you with its ability to keep the player participating. But users of extrinsic motivation must always be wary of its partner in crime, the Justification Effect. That article makes mention of an interesting study, wherein they gave prizes to children who chose to draw during free time, and what's so fascinating is that weeks later, when given free reign, the children showed a sharp decline in their interest in drawing. Humans are so crazy! The takeaway, to me, is that giving planned — that's the key, remember — rewards to people for things they already like doing kills the enjoyment. It distorts their user story. Don't you see how that can be potentially damaging to your game? Please, people, think of the children. Look: I'm not saying you must remove compulsory rewards in your game. Hi, I love Diablo just as much as the next person. What I'm trying to tell you is that extrinsic motivators have a downside; a very powerful one, and you can not, should not, dangle carrots in front of every single part of your game. If you keep asking yourself, "BUT THEN WHY WOULD THE PLAYER WANT TO DO X?" then, well, I'm going to go out on a limb and say you have bigger problems in your game. Designers that constantly worry about trying to compel the player through their game have stopped asking the right question, which is, of course, wether the game is any damn fun at all. (Chances are the answer is no, and you are afraid to admit it). Accomplishment and itrinsic motivation The righteous path is usually the harder path, and leaving the player feeling accomplished is difficult work. It is the ultimate result of a pyramid of factors, but it comes, most of all, from allowing the player to make meaningful choices in how she resolves her conflicts. The thing about accomplishment is that it is addicting, right? Mastering new things, as Daniel Pink tells us, is one of the three big motivators in our life, and once the player feels like a complete bad ass, is she really going to stop right there? No! She's going to find that new hill to climb, and climb the ever loving shit out of it, so to speak. In my design process, fun and accomplishment are interchangeable concepts — if I'm having fun, then I am probably accomplishing something, and vice versa. Despite their interchangeable nature, I spend less time thinking about fun than accomplishment, for I find levels of fun harder to quantify; consequently, when playing the game, I often ask myself, "do I feel like I've accomplished anything," which usually provides a very stark answer: yes, or no. These are the two major paths to keeping the player involved, and they are complicated topics in their own right; for this discussion, however, I must move on, as I have a lot more to cover. As stated, motivation is great, but there are also two major pitfalls that must be avoided, or everything will come to naught.

We have covered a lot so far, so let us review. Boss fights in games are satisfying when, like movies, they maximize their intent, and our list of requirements is similar to what we wanted with movies — but with some differences.

Our list of requirements is longer, but I would argue that it is not necessarily harder. It is just different. Viewed in this light, though, we can clearly spot where most games fail. We rarely, if ever, allow our characters to act in visually unique ways, unless it is in a cutscene; yet still more damning, we far too often introduce, or shoe-horn, functionality into boss fights that complicate the experience.

Clip: Uncharted 1, final boss fight, an example of how not to do it.

I find that avoiding complication and requiring participation are problems that, once aware of their need, make for easy bed fellows. This leaves asymmetry. It is clear to me, and maybe to you, that one of the greatest hurdles in this list is the goal of overcoming adversity in honest and visually unique ways. Which brings us to one of the greatest problems we face in game development. The problem Our oldest foe has always been that of content. I know it, you know it, most everyone knows it; and, if you don't know it, then you are one of those unfortunate people that screws whole teams over, so you BEST learn it. All too often in my career, when discussing certain ideas, the worry has been about the One-Off: work that is not duplicated. It is the greatest crime you can make in a world of constant reuse, for we must stretch our content as far as we can in games — to the breaking point, honestly. Every seasoned designer has grappled with the One-Off problem. You have a great idea, but your engine doesn't support that exact kind of gameplay, or the enemy doesn't have that animation. Therefore, regardless of the quality of the idea, it must be rejected. Both the time requirements and quantity of content in a game dictate that one-use content is a waste of time. This is our reality, right? One-offs are not inherently evil, though; I am here to tell you that, when used appropriately, they can inject that much needed variety. But you must do it in a way that neither explodes your content, nor overcomplicates the experience. In short, you must learn the Art of the One-Off. The art of the one-off The art of the One-Off is all about planning, compromise, and clarity. If you go into your designs planning for one-offs; if you know exactly when, where, and how to use them; if you understand that you will have to compromise on certain things; and, finally, if you ensure that everything is expressed to the player in a clean manner, then you are ready to introduce very meaningful One-Offs into your game. God of War is riddled with One-Offs, but the goal is usually to hide their simplicity. In Chains of Olympus, for example, we had a few world interactions where we wanted Kratos to knock over large columns, or other large objects. With poor planning, these might have easily exploded into tons of unique animations for the character. Instead, we used planning, compromise, and clarity to ensure that we had the best experience.

Clip: Chains of Olympus, interaction one

We created one shoulder pushing animation and ensured that it would be used in any world cases, planning; since he would be using the same animation, this prompted the world interactions to be slightly adjusted, compromise; lastly, all interactions use the same button to activate, and they are all indicated with twinkles in the level, clarity.

Clip: Chains of Olympus, interaction two

This is a pretty simple example, and all of this is easier said than done, I know, so let's study a more complete and fantastic example of what I am talking about: Mega Man X. Mega Man X Mega Man X is an awesome game. More than that, it was a transformative game. It blew minds, people, and it has more than earned its place in my All Time Favorites. If you have never played it, which would be a CRYING SHAME, then I shall direct you to this wonderful video — Ridin' on Cars! — that fully captures the awesome of this game. Mega Man is a series designed around boss fights, so it makes for a great test case. The stages are constructed to train you — which they do, masterfully — for the skills that you will eventually use to defeat the bosses; additionally, all five major actions you can perform in the game are simple:

That's it. The key awesome part here, the part that makes this the great One-Off win, is the brilliance in allowing the megabuster to steal attacks from the bosses, and then coupling that with bosses having a unique weakness to specific weapons. That feeling when you use just the right weapon on just the right boss… exquisite. Check this shit out. Watch Spark Mandrill get totally owned by the player's use of Chill Penguin's ice cannon.

Clip: Mega Man X, Spark Mandrill fight

That's what I'm talking about, people. How do you not scream at your TV, "I GOT YOU," the whole time that is playing out? It never feels cheap, by the way, because you worked to get Chill Penguin's gun, and chose to use it on Spark Mandrill. I know, you're sitting there all like, "damn, Mike, that is kind of bad ass." But dude, just wait, because it gets even better. Every boss has a cool unique intro; and, like any good movie, none of it is wasted. Each intro tells you a little something about the guy you are going to be facing. That way, when they use their cool little tricks — always give your enemies a trick — you never feel cheated, because you say to yourself, "yeah, fair enough, I saw you do that."

Clip: Mega Man X, Boss gauntlet run

I know, now you're sitting there all like, "Mike, seriously, I can't handle this awesome." BUT WAIT. After their intro, every boss does his little pump-up stance, and we get to see his health bar fill all the way up. This is important, as it is the final piece in the puzzle. It very clearly makes a statement about the boss's initial health advantage. This all sounds awesome, but let's compare it against our goals:

Look at that laundry list of ownage. Nailed it. Point being, they could have easily designed all of those fights without all that One-Off work (unique intros, unique boss rooms, and unique weaknesses). But the fights would not be anywhere near as satisfying, and that's what matters. As we said in the beginning, the art of the One-Off is in knowing when the benefits outweigh the cost, and in knowing how to avoid complication. In the case of Mega Man X, since it was a long running sequel, they clearly went into it knowing that Mega Man was going to be stealing and using powers to kick Robobutt, and it shows in the quality of the boss fights. Read that sentence again. Here, I'll highlight the important part: "…they clearly went into it knowing…" Roll that around in your head, because you'll realize there is nothing that prevents us, too, from crafting such cool boss fights. All you need is good planning. The six P's My grade school teacher made me constantly recite the Six P's: Prior Planning Prevents Piss Poor Performance. It drove me crazy. Maybe she was on to something, though, because it's forever burned into my brain. I hate it when teachers end up being right. In our case, seemingly, a piss poor performance is a crappy boss fight, and prior planing is our documentation, right? Well, as you are probably aware, game developers are rarely stricken with sparse documentation, and often we find ourselves laden with pages and pages of laboriously stipulated notes. Been there, done that. But that's not working, obviously, so clearly documentation — at least, the kind we're doing — is not the kind of planning you really need. GRIP for boss design Writing a boss document, marking off a spot near the end of the schedule and labeling it "boss fight" does not cut it. Good planning starts from well crafted goals that are clearly stated; and, understanding your goals is the first of four cycles — GRIP: Goals, Research, Implement, Polish — that all systems go through, even boss fights. Define goals Thankfully, understanding and defining our goals has been the plan throughout this entire post. It has guided us every step of the way, and we know the right goals for any boss fight:

These goals, like most things I talk about, are not hard rules that you must follow. They are a tool that allows you to ask the right kind of questions:

These, and others questions, are just starters, but asking the right questions will help you help you to narrow in on the Intent of your boss. This is important, as you will see when we get to Implementation, but for now, we are only concerned with putting it to words. Intentions, in general, are the specifics of how you plan to accomplish a goal. For example, my goal may be to brighten up a room, and my Intention would be to flip on a light switch; I could, however, achieve the same goal by yelling loudly and getting my roommate to flip the switch. Same goal, different intentions. Therefore, in our case, we know our goals, and the Intent of our boss is the specifics of what we are trying to accomplish. An example:

What is important to note is that there are a lot of ways to implement that boss and still maintain the same Intention, but we will get to that in a second; meanwhile, having a solid understanding of your boss's intention is the best way to hit the ground running when you hit the next cycle in the process, Research. Research other bosses I swear to you, there is not a single boss idea you can come up with that has not already been done. Unless the underlying mechanics of your game are totally wacky and new, chances are damn near 100 percent that you can find a good example of what your want in another game. Do us all a favor and check it out. Note: If your company gives you crap for playing games at work, then you are probably working at the wrong kind of studio. It is in this phase that you, and I, usually end up with some wonderfully crafted excel file that lists all the moves, frame windows, damage numbers, and all the other data in their perfectly organized and number-crunched beauty. This is great, and necessary, but get ready to leave it behind. Implementation is about feel You are about to become the best friend and worst nightmare of both the animators and programmers; but, since we've already nailed our goals and done all the necessary research, we are way ahead of our contemporaries. You will discover a world of difference in how people react to you when you approach them with a good head on your shoulders, and the first step is in understanding that implementation is all about Intent. This is where our hard work from before pays off, because I guarantee, as soon as things start getting implemented, your carefully laid plans will divert; however, since you understand the Intent of your boss, you are ready for one of the most critical parts of being a designer: handling the flux of development while maintaing your intent. I can't even begin to tell you how important that is to your job, but I'm trying. It's super important! Things divert because it's hard to know how something is going to feel, and that, in the end, is what this phase is all about: things have to feel good. Which is why, by the way, most documentation is so inadequate, for it rarely discusses intent, and feel will always distort your original plans. Polish with the right questions Vocalizing your thoughts on the feel of a boss is difficult, so I like to ask myself the following questions, which I get from the principles of animation:

These won't do your job for you, obviously, but they are great questions to ask. Generally, if you feel like something is missing, then you can usually find the problem through asking those kind of questions. Closing Games are not movies. It's such an obvious statement, but it can be very easy to find yourself caught in that mind loop; consequently, from what we've discovered, it is the reason why some video game boss fights lack satisfaction. To return to the original question, I feel that if our greatest problem was a lack of visual expression, then we would solve that problem. We totally would. Game developers, as a group, are pretty damn smart people, and when they collectively put their mind to a problem they find the solution. Games have different goals, though, and we have to give up that battle in order to achieve some of our other goals; namely, avoiding complexity. There is no sure fire formula for a great boss fight, unfortunately, but greater consistency is something that we certainly achieve. Great boss fights are achieved through good planning, and good planning comes from a strong, clear understanding of the goals, from clearly stating and understanding the intent for your boss, from researching bosses in other games, from successfully handling the inevitable deviation in your design, and from asking the right questions to nail the feel. [This piece was reprinted from #AltDevBlogADay, a shared blog initiative started by @mike_acton devoted to giving game developers of all disciplines a place to motivate each other to write regularly about their personal game development passions.]