Players want to kill the unkillable, solve the unsolvable, and explore the unexplored, but most of all, they want to feel they have accomplished great things. The greatest goal of God of War combat is to ensure the player never deviates from his quest of awesome. This goal, this process, is tied into your understanding of combat encounters and how they fit into the flow of your game. Everything about the encounter, what you face, where you face them, how you complete it, all of that is just a means to an end. The best fights challenge the player tactically, strategically and intellectually, but they never cross the fine line that leads to frustration.

After 4 years and two god of war games that is what I have learned: our job is not to prove how smart we are or how good we are at our own game, it is to prove to them how smart they are and how good they are at our game. Enough talk, it’s about time I get down to the specifics: what works, what doesn’t work, and what are some classic styles of encounters.

What Works

I love action adventure combat. When it works, when things come together, there is no greater feeling than getting the best of a tough opponent. A lot of my favorite moments revolve around well designed combat encounters, and though the characters you play and the monsters you face are rarely the same, there are several common traits that all great games share.

Killing The Unkillable

Taking on the unkillable, while intuitively awesome, is one of the hardest fights to execute effectively. We have seen a lot of games attempt larger than life bosses, but rarely do they turn out very fun in practice. Do not think that scale is some simple equation to be applied to the player’s enjoyment factor. Observe the difference between Shadow of the Colossus and Castlevania: Lords of Shadow. The former is memorable and fulfilling, while the latter is hamfisted and frustrating.


The larger things get the more effort and thought must be put into the encounter. Even “oversized” monsters like the Cyclops are not a simple matter of scaling up a regular monster and calling it a day. How they react to the player and how the player reacts to them are going to be meaningfully different than other monsters. When it works, though, it is so awesome. The Colossus from God of War 2 – a joy and a thrill to fight – owes his experience to the designers’ expertly hiding his simplicity behind a glamour of showmanship. Does the player notice the colossus’s attacks do barely any damage to Kratos, or do they notice that his collision box covers almost the entire screen? No, all the player sees is that he survived like a badass, and in the process just destroyed something awesome.


Grand Set Pieces

Set pieces here refers to both the spaces you occupy (arena) and the spaces you can see (vistas). Your set pieces can, properly handled, dramatically enhance your combat experience without impacting it’s play. Some have a grand visual impact, such as the first boss battle in the star wars game, and others have an important gameplay impact, such as being inside the colossus in God of War 2. Having grand set pieces is important, but never forget the most important rule of scale: if everything is big then nothing is big. Large things only appear large when you put them up against things that are small. When you forget this, you end up with a game like Too Human where things that are meant to be “epic” end up seeming “normal”, leaving the players and characters to appear as miniature dolls in a (not-so)funhouse.

High Pressure

Nothing kicks up the adrenaline like the threat of impending death, so at times you must allow the player to let his mass murderer out of the bag. Kill them before you are crushed, kill them before you fall through the floor, kill them before they evolve, and so on. This is a misunderstood style of encounter. Yes, it is true that “no one likes timers”, but when it is done right, when it is done well, this will be a memorable moment that turns a boring fight into a very achieving moment. Sometimes the player needs a little push to see how far they have come, and how much they have mastered.


Great fights have a flow to them. They start off strong, sure, but they also build; eventually building to a final crescendo. Great fights never overstay their welcome, and they never leave you unsatisfied. Flow, however, is more than studying tension within a single fight. Your encounters are not independent microcosms, and you must keep an eye on where your encounter is placed within the grand scheme. If the player just left a very difficult fight it is probably not appropriate to go right back into another. Flow is important not only for your fight, but also across all fights.

My Favorite Fight From Chains

These common traits are all different paths to the same goal. Big monsters, grand set pieces, and high pressure design are all geared towards accomplishment. I mention accomplishment a lot, as it is such an important concept. Note, however, that giving the player a sense of accomplishment is not the same as “making things difficulty” or even, noble it may be, as “making it challenging”. Challenge is a MEANS to achieve the feeling of accomplishment, but it is NOT the same.

So, let us make the assumption that you agree and take to heart these concepts. This is the first step. Talking about what makes a great fight compelling can help, but these commonalities in good design will come to naught if you do not avoid the common mistakes.

What Doesn’t Work

Studying and striving for what you know works is a great start, but even with years of experience playing games and studying what I loved there were countless mistakes that I made when designing encounters. Some are hard to avoid, and some, once you know, are easy to avoid.


The greatest and most common mistake when designing combat encounters is simply including too many damn enemies. Everyone does it. Do not fret, because it’s an easy mistake. The problem stems from our perception of good encounters that exemplify flow. Think back to your favorite fights, picture how many enemies you faced, and then replay them. Take special note of how many enemies you truly face, and I am certain you will come to realize it is fewer than you remembered. That’s ok! That means they were designed well, because a well designed encounter FEELS like swarms of enemies are pouring out of everywhere, and it is that feeling that gives you a sense of accomplishment when you survive.

The best defense you have against this problem is to play your own encounters. Not just by themselves, but within the context of the level it is within. Even if you have nothing to do with the level itself, you MUST play your fights within the surrounding context. I cannot stress this enough. If I could, I’d send up fireworks on your screen so that it spelled “context”. Even if your encounter feels as if it has just the right number of enemies (probably wrong), if the previous / next fight is also a significant encounter then you are going to give the player combat fatigue.

Front Loading And Waves

Front loading enemies is another grave mistake, and is most commonly made in games that insist on enemies all appearing “naturally”. Laugh all you want that enemies magically come out of the ground in God of War, but there are numerous reasons to allow your designers to do this. Remember what I said about Flow and it’s importance to the feel of an encounter. Controlling the difficulty and intensity of your fight is integral to managing the flow of your fight; however, if all of your enemies appear right at the start you are, in essence, starting off at the most intense and slowly deflating as you fight – not what you want. Where’s the surprise? Where’s the intensity? Designers must have control over not only where enemies appear, but also when they appear. Think of the start as an “oh shit” (OS) moment for the player. If you front load your fights you are missing out on multiple OS moments as new waves of enemies are spawning in. Yet it is more than simply understanding that you NEED to spawn in waves, you must also understand how to trigger these waves.


Death is not the only trigger at your disposal. Yes, death is a good metric, but health is also a very important trigger, especially when dealing with larger, more difficult enemies. A favorite trick of mine is to pair a large (like a cyclops) with two or three smaller pests (hoplite) and then, when the cyclops reaches less than half health to trigger a second cyclops. It instantly ratchets up the intensity of the fight (OS), without adversely increasing the difficulty. You see, the player sees two cyclops’s and she feels overwhelmed, but in actuality one is almost dead. If you wait until the cyclops died before spawning the second you lose out on that intimidating sight of two (even though if you combined their health bars it’s 1.2 cyclops). It’s a slight change in flow for a massive payoff in feel.

The Second Cyclops Is Triggered On The First’s Health

Zoning And Pulling

Something still feels off. They are spawning in waves, they are triggering on things other than death, but somehow the fight is missing something subtle. You forget zoning. The player is not a static entity in your arena, and where he stands affects where things should spawn – especially if it’s a big monster. The player is always moving, so why would you use the same spawn points throughout the fight? You wouldn’t. You must design the system to understand where the player is located and adjust where things are spawned accordingly. If the player is on the left side of the arena, spawn the archers on the right; if the player is on the right side, spawn the archers on the left. To do otherwise creates a flat experience.

This does not mean, however, that you should always do this. You must also understand the concept of pulling the player. The player is naturally drawn to enemies – assuming your enemies are fun to kill – so you can use them to draw the player around your arena space, if you so choose. Ranged characters are especially effective at pulling the player, specifically because he MUST go to them. Let’s say you wish to ensure the player is in the north end of an arena so that you can do a grand reveal of a big monster. Spawning archers in the North will naturally pull the player.


Variety is good, but keep your fights simple. Having too much variety (within any single fight) is a very real mistake. Just because you can pair certain enemies in your cast, does not mean that you should. This is where the importance of breaking your cast down into Classes and Roles comes into play, as it is generally not a good idea, within any given fight, to combine more than two Roles or more than two Classes. I know the thought process: ok the cyclops is awesome, the medusa is awesome, and the hoplite is awesome; obviously, if I combine them all, it will be three times the awesome. Wrong! Your enemies are playing too many mind games with the player. Instead of creating a really challenging scenario you have created something that is frustrating. You don’t need to PROVE to the player how awesome your enemies are, and how devastating they will be in certain combinations. The player isn’t going to say to himself, “Man, ya know, getting my ass handed to me really IS awesome! These monsters suuure are tough to kill.”

Collision And You

Next you gotta keep that floor flat. Flat flat flat. Especially if your game has any kind of throw. Avoid stairs, avoid ramps, avoid giant random ledges. This is (and forever will be) an ongoing battle between you and the artists. Maybe some day we will design combat systems that can handle the nuances of sloped ground collision and not look like COMPLETE ASS, but that day has not yet come, and the flatter you keep your combat arenas the better things are going to look. You are not making an FPS, and while adding some verticality to your fights is a great addition, in an action game it has to be added with great care.

Whoops! It’s OK, it’s my level =(

Ok, look, it is not that you cannot design your arenas with ledges everywhere and stairs galore, but you are just going to discover you miss judged the complexity of the situation. Collision in action adventure games – both wall and ground – is a very complicated mistress. I would guess that of all the bugs I ever receive about a level I built, a good 50% of them were collision related. Players falling out of the world, monsters falling out of the world, throwing a monster out of the world, things spawning above the ground… it goes on and on. It is not a matter of making a polygon and saying, “this is a wall”. In God of war, for example, here are just SOME of the wall collision types we used:

  • Ignores Kratos (kratos can pass through but enemies cannot)
  • Ignores Enemies (kratos cannot pass, but enemies can)
  • Ignore Projectiles (kratos and enemies cannot pass, but their projectiles can)
  • Ignore Kratos, Ignore Projectiles
  • Ignore Enemies, Ignore Projectiles
  • Ignore Jump (while running, walking, or attacking you collide, but as soon as you jump you pass through)
  • Ignore Walk (running, attacking, and jumping collide, but you can walk through)

You get the idea. Respecting and understanding collision separates the men from the boys.


Last, and this is a big one (ha), is scale. Don’t make your arenas so damn big! If Kratos is “2 Hero Units (HU)” tall, then the starting size for an arena has a diameter of around 15 to 25 HU, which might sound ludicrously tiny. Do not be afraid of small spaces! Your big arenas might LOOK cool (visually) but they are going to feel like SHIT to play in – trust me. The bigger the space, the more your player can run away; the more your player is running away, the more downtime you have; the more downtime you have, the shittier your fights are going to feel. You want the player to feel as if they are escaping by the hair on their nose, and then they are immediately back into the fray. They have plenty of time to cool down when they are traveling to the next fight, or better yet, solving the next puzzle. A fight is a fight! Not an excuse to run around. This goes for your big enemies as well, don’t be afraid to throw one in a tight space with the player to see what happens. You might be surprised!

Look At The Size Of That Arena!

Only One Of Many Mistakes They Made

Is That It?

No, there’s more! These are all pretty important, but avoiding all of these pitfalls will create structurally sound but homogenous fights. Regardless of how cool and fun your cast is, and regardless of how many combinations of them you can create, you will bore the player eventually. You must vary not only the cast in your encounters, but also the style of the encounters.

Encounter Style

Having a good variety in your cast is good, but the more significant problem is that many games fail to have a good variety in the style of the encounter, and this is an important enough topic to warrant it’s own section. If every fight starts or plays the same, regardless of how fun or diverse your cast is, you are going to bore the player. Some styles are less about the arena space and more about a trick, while others are tied directly into the specifics of your arena space. I have identified five (certainly not all) styles of encounters: Intro, Arena, Pincer, Trick, and Surprise.

The Intro

When you introduce a monster it is not just an opportunity to add extraneous cinematic work. Introductions can be an opportunity to sell a new trick to the player, they can enliven the flow, and they can reinforce the threat of your enemy. Let’s say you have a sub-boss class enemy – a tougher style enemy – but your player’s are not approaching it carefully. They rush in and die against him. Having a special intro that shows this monster smashing several npcs, ripping them in half, and drinking their blood is going to make the player stop and say, “whoa – this guy is not to be messed with.” Or maybe the enemy has armor, and the player is having trouble understanding. Having an intro showing soldiers raining blows upon the monster while he laughs at their futility is going to help your clarity issues. The Intro is a lot of work, but when used correctly is very useful.

Notice How It Both Sets Her Up (audio)

And Sells Her Abilities (goes through walls)

The Arena

The “big” fight. Trap the player in a space and send in waves of enemies. I’ve spoken about this throughout, but the most important thing to remember is context. The context of where this arena sits in the level, and the context of where this fight takes place in the flow of the game. Planning on a grand, memory-intensive vista along with your arena means you will blow the budget having 3 different enemies; similarly, if you just came out of a big fight you will probably exhaust the player by going right back into another. Context is imperative when considering the Arena. Also, more than any other, having a good grasp on flow and zoning is critical. The bigger and longer you wish your fight to be, the more important it becomes to not front load your enemies.


Context Is Important

The Pincer

Fights of this style are about dividing the player’s attention and forcing commitment. As the name suggests, one of the classic ways to do this is to spawn enemies on opposing sides of the arena, but that is just one example. This style of fight requires the designer to fully understand and make use of the concept of pulling the player. Some enemies, like archers, are designed to pull the player, and if you do not commit to them, they are going to rain death on you with impunity. The Pincer style of encounter includes any fight where the player is pulled in more than one direction. What about three directions? You can try it, but you aren’t going to like it. A player can handle a tough choice between two threats, but the second you add that third dimension it just feels like you are getting bent over by the designers (you might as well admit that’s what you were trying to do).

The Trick

Worried you are boring the player with simple fights? Enter the trick. Here you take your basic encounter and you spice it up with a simple puzzle mechanic. The key here is SIMPLE. You cannot combine visceral and cerebral. It just doesn’t work. Trust me. The actual space of the arena is less important in these fights, because unlike the chaotic focus of regular combat, the puzzle mechanic will become a focal point. These encounters can be very effective when done well, but be warned: these fights are the quickest to go diving off the edge of challenging into that horrible abyss of frustrating. Some classic tricks that I always find useful.

Gate: You must kill a certain enemy, possibly in a certain location, in order to proceed. This works well when it is paired with another enemy that pulls the player.

Delivery: Moving something from point A to point B while being harassed. When it works, it works (pushing the caged man up a hill in God of War 1), and when it fails, it crashes and burns. The only solution to nailing this trick is loooots of play testing. You gotta find that right balance.

Activate: Use something while being harassed. Though I list it, I feel I should add that this can quickly become your “go-to trick”. People will get sick of this shit really fast if every other fight is “turn this crank while dudes punch you in the face”.

Countdown: Finish the encounter before a countdown. This style requires a very important point: the countdown doesn’t always need to ACTUALLY be counting down, it just needs to FEEL like it’s counting down. With careful scripting you can give the impression that the sky is falling without actually fucking the player with a falling sky. These are definitely some of my favorite fights.

The Surprise

I like the surprise, and you should always find the budget in your levels to squeeze in a good surprise now and then. It keeps the player guessing. The surprise is generally not a full fight, and they usually don’t take place in things you would normally consider “good” arenas. In fact, that’s the whole point. Arenas tend to have that “look” to them. A player sees a space and says, “well there’s gunna be a fight here…” Not what we want here. The surprise should happen exactly where you don’t expect it, and the goal is to drive it off a player action. Player opens a chest: a guy jumps down from the ceiling. Player breaks down a wall: oh shit what’s that hoplite doing in there! Obviously, like all things, this can be over abused. Doom 3′s “monster in a closet” is a classic example. Do it too often and players will quickly call bullshit on your shenanigans.

The End

Ultimately, when designing your combat encounters, your greatest asset is not yourself, but the people around you. You designed it, so of course you are going to be good at it. You need to remember that YOU are not the audience. Your job is not to prove to the player how awesome and smart you are, it is to prove how smart and awesome THEY are. Have other people play your fights, and spend less time listening to what they say, and more time watching what they do. I say this not to deemphasize their spoken feedback, but to emphasize the importance of studying their actions. Every person might says, “damn that cyclops is super tough!” when in actuality they were all getting killed by the pesky hoplites.

I learned a lot living with Kratos for so long. He’s a demanding bastard, but he knows his shit. If I could pick one thing to end on, it would have to be that what player’s really want is to feel they have accomplished great things. If you stick to that, you can’t go wrong.