This video is supported by Patreon.
Video Game Design. Sometimes we think we have it all figured out. At least to the point that we can safely say what design decisions are bad and shouldn’t be done in games. After all, there’s enough history to go around with and tons of established conventions, right?
Like, for example, it is bad design to not put frequent automatic checkpoints in the game. If you fail a particularly tricky challenge, or it so happens that you need to exit the game, we don’t want to push you too much back, that would be too punishing. Therefore, infrequent checkpoints are bad… But they’re also good, like Alien: Isolation shows with its manual save point system where save spots are placed a fair space apart.
Ok, well, we know that games without animation cancel are bad. If you start punching a henchman and see another one attacking, you need to be able to press counter and cancel out your hit to counter successfully. That just makes sense, right? You didn’t know the other guy was going to attack, you need to be able to adapt. Lack of animation cancel is bad… But it’s also good, like in Dark Souls where the commitment to your actions is what adds so much depth to its combat system.
Hm… Well, you know, Alien: Isolation is a horror game and Dark Souls is, well, Dark Souls, it makes sense for these games to subvert the established norms and still be fine, right? Oh, I know, let’s take progress tracker. Games can be a hefty time commitment, so it’s just good design to show how much of the game you have seen in some shape or form and if there’s something you’ve missed. Not allowing to track your progress is bad… But it’s also good, like in The Stanley Parable, where the whole joy of the game is about discovering all the different endings without knowing how many are left exactly.
Wait, so if all established norms of good design also have examples of good design that do absolutely the opposite… what IS then Good Game Design? This is what yours truly, Stanislav Costiuc, is going to discuss in this month’s edition of Farlands, a series about Video Games and Video Game Design.
To start answering what good Game Design is, though, we need to get back to the core of the medium… and ask, what is a Game in the first place? There have been many great definitions across the years. My three favorites are the following:
“A game is a system in which players engage in an artificial conflict, defined by rules, that results in a quantifiable outcome.” – Eric Zimmerman & Katie Salen
“A game is a type of play activity, conducted in a context of pretended reality, in which the participant(s) try to achieve at least one arbitrary, nontrivial goal by acting in accordance with rules.” – Ernest Adams
“A game is a series of meaningful choices.” – Sid Meier
I truly believe that these are the definitions every Game Designer should know. They are great for theoretical discussion and teaching of fundamentals. But, over the years I started to find them somewhat… dry. One thing I feel is missing quite often when dissecting game design and what makes it tick is a holistic view of a game – which is probably the most important aspect of defining what makes a particular design actually good. After all, what stays in people’s minds is not the mechanic itself, but what it made them feel.
Games consist of many things outside of the systems and rules in place. The meaning, the tone, the sound, the visual design, the narrative, be it pre-defined or created by players via interactions with mechanics, and many more things. And it all doesn’t work in a vacuum, it lives together in an ecosystem that defines the feelings and emotions that players experience after interacting with the game.
Which is why I have a liking for the following definition: “Game is an Experience.”
Admittedly, it is pretty broad and doesn’t help with explaining to somebody new in game design or development how a game works… but it is immensely useful for the process of design itself – both in terms of creation, and evaluation.
The thing is, a video game can meet all the definitions of being a game and yet provide an awful experience. Likewise, a game can be a collection of excellently designed systems that as a whole don’t really build something cohesive or special.
The target experience is defined by developers. It doesn’t have to exist from the get go, it can be found throughout the pre-production process, and it can be as broad or concrete as we’d like, but it must be done with an understanding that every choice we make in every part of the game will affect the resulting experience in some way, be it minimal or critical. The players do not have to be able to explain what the experience is exactly, but they will feel when something is off or doesn’t quite fit right.
For example, let’s take Journey. As an experience, Journey is all about the Hero’s Journey paradigm. Which is too specific for players that are unfamiliar with narrative structures to be able to explain, but they don’t have to – they still get the whole array of emotions from Journey that the paradigm entails. The evolution of gameplay mechanics throughout the game, the level design progression, how the color palette changes, the music, everything works together perfectly to create a distillation of the Hero’s Journey in game form and provide that experience to the player. And if thatGameCompany would’ve made different choices, the players might’ve received an experience that was either different or disjointed.
And this all is a build-up to the very first point of what defines Good Game Design. Good Game Design is Contextual. How good a design choice is can’t be defined without knowing the intention of the overall experience.
Let’s get back to Alien: Isolation for a bit. It wants to recreate the feeling of the movie Alien where you feel hunted by this creature and rarely feel safe, which is why the manual save point system is better than automatic checkpoints. Hiding in a locker as the Alien walks by just wouldn’t feel as intense if you knew that if you die you just respawn at the room entrance.
During development, though, Alien: Isolation did in fact have an automatic checkpoint system, as its Lead Designer, Gary Napper, tells. However, there were couple of complications that the team had on the way related to its dynamic Alien AI.
“It was the question of what to do with the alien when we save that raised the debate of save points again. “When I get to a soft checkpoint and the alien is about to kill me… What happens when I come back in? Is he still there?”
We had two clear options. One option meant not doing anything which would trigger an infinite loop of dying and respawning. The other option was to remove the alien on the load, which would mean sprinting to a checkpoint was a valid tactic — as after he killed the player, they were safe again. Neither of these were good enough, and neither supported the core gameplay.” – Gary Napper
A designer, Simon Adams, throughout development proposed a manual save system several times, and this is what happened after the team at Creative Assembly decided to test that out as a solution to their problems:
“Our design team quickly put together a simple scripted push-button save point that would trigger a manual save. The first mission to use this suddenly took on a different feel. No longer were we walking around, without a care, knowing that if we died, we wouldn’t lose much progress. We were afraid. If we didn’t make it to the save point and successfully save, we would lose our progress. Even in a single mission this was tense.” – Gary Napper
How the game is saved is a decision that has defined what players experience and how well everything fits together. And it’s not that horror games can’t use automatic checkpoints. Dead Space works just fine with its mix of manual and automatic saves, but Dead Space is also not about a single monster with a dynamic AI hunting you throughout the game. A choice that is good in one context is not necessarily good in a different one.
And even if overall intentions make sense, specific designs can still act very weird in context. The Trail tries to create a Frontiersman experience, so it utilizes sensible concepts like crafting and trading. Yet these concepts are designed in a very weird, and I would even say sometimes off-putting, manner. Crafting is this almost slot machine-like endeavor where you place resources and pull a lever, while trading is a strange speed-based contest mini-game with a belt that pushes everything into a grinder. And it’s not that in abstraction these are bad mini-games, but if in context you’re as confused as I am, then you see what I mean.
It’s also important to note that what is absent from the design is as important as what’s in it. For example, in the same Trail game, in a situation where somebody collapses on the road, your only two options are to either just ignore them and move on, or steal their things… even though your bag can be full of stamina-restoring food, you can’t help anybody in any way, and that also affects the experience – mainly that there’s nothing that allows you to befriend somebody in that wild environment you’re exploring.
All this leads us to another aspect of Good Game Design. It is Deliberate. It’s not enough for design to be ‘just cool’ or ‘this is just the way it’s done in these sorts of games’. It doesn’t mean that it’s bad to follow conventions, but it’s also important to understand that every convention exists for a particular reason to solve particular needs, so you need to ask yourself at least once if that reason fits YOUR needs. In general though, you need to be able to argument the inclusion of every design decision.
It doesn’t have to be something related to core experience. For example, mini-games in Assassin’s Creed IV: Black Flag exist to flesh out the historical world and allow for some relaxing activities should the player desire one. Mini-games do not improve the core gameplay, do not affect the overall narrative, but there’s a reason they’re there that makes sense in the context of the world and the main character.
And I know I’m jumping back and forth between different games, but let’s get back to Journey. There’s an hour long GDC talk called ‘Designing Journey’ which is a worthwhile watch for everybody. Jenova Chen talks about the process of developing Journey, starting from the intentions of focusing on cooperation and emotions of awe and mystery, and ending with how different elements were applied to fit the Hero’s Journey paradigm.
ThatGameCompany has put the emotional arc of the Hero’s Journey and three-act structure in the form of an emotional level graph. A lot of things were designed based on that graph. The elevation levels of different parts of the world mimic it, as well as color palettes change depending on what kind of mood or emotion the developers would like the players to go through. More than that, the freedom of movement was also based on the graph, as it can cause different emotional responses from players as well. In the lower parts of the world and the narrative where a struggle was desired, navigation was slowed down, animations changed, and in general things were at a slower pace, while the catharsis of the whole story has allowed players full freedom of movement.
I’m not going to retell the talk in full detail, I think it’s best if you go watch it, I’ve put the link in the description, but the point I’m trying to make is that good design always has reasons behind it, be it big or small. If that’s not the case, it’s very easy for the game to take on a kitchen sink approach, especially in sequels to successful video games that try to put in a lot of new things.
An example of that is the Batman: Arkham series. The first game, Arkham Asylum, was an excellently designed focused experience. But ever since then, the series started taking the ‘bigger and more’ route, which doesn’t necessarily mean ‘better’. Of course, it also doesn’t mean that the games become bad, but gradually over the course of the series a fair share of flaws piled on, culminating in the “biggest and morest” game of the franchise – Arkham Knight.
Arkham Knight had the biggest city, the most amount of villains in the series, the widest set of gadgets and skills… and yet in that biggest city there was a very small amount of interesting interior locations to explore, which is actually Arkham series biggest strength, a lot of side-quests involved you doing the same thing over and over again with little variety, and a lot of gadgets and skills weren’t really needed that much in gameplay.
And of course, there’s the Batmobile – the biggest addition to the series that due to its nature of having a lot of content created around it, dilutes the experience even more. The tricky thing here however is that the Batmobile is something that both makes sense in context, and the reason behind its design and inclusion are quite deliberate, so why has it been the biggest point of discussion regarding Arkham Knight?
Well, here we get back to the point of holistic view of a game – how every system interacts with one another, how nothing lives in a vacuum and how all parts of a game are essentially an ecosystem that affect each other. It all must be Cohesive. And cohesion is where viewing a game as an experience helps the most.
The issues people had were not related to how Batmobile plays, it is quite excellent from gameplay point of view, but how it was incorporated into the game as a whole. It’s very typical of Arkham Knight to create levels where you can’t progress without the Batmobile, but where you also need to regularly get out of Batmobile to open up a pathway for it.
This goes against several things – the established mechanics and gameplay in previous games of the series where Batman in similar situations was very capable without the Batmobile but in this game can’t do without it. It also goes against Batman as a character, being so reliant on that piece of technology whereas usually he wouldn’t have been. Plus there’s the fact that sometimes situations where you need to use the Batmobile just feel forced. All this leads to an experience that can feel disjointed, which doesn’t make the game bad of course, but it does affect players’ emotions and enjoyment.
That said, there’s one other reason why cohesion is very important: it tramples whatever flaws one can find in your game.
I really like using Assassin’s Creed: Brotherhood as an example of a flawed game that through its cohesion still becomes something great. If you start analyzing the game in detail, you’ll find a bunch of things one would consider ‘alright’ at most – a combat system with a single dominant strategy through killstreaks, overpowered tools that make others redundant, a weak black & white narrative, and essentially a ‘win button’ in the form of calling the Brotherhood.
But when people play games, even though they can discern particular parts they like or don’t like, their emotions and experience are not based on how everything works on its own, but how everything works as a whole. A game is a sum of its parts, and Assassin’s Creed: Brotherhood still builds a great experience where you, wounded and defeated, get to Rome, and set on a goal to liberate it from the Borgia. Almost every activity in the game ties into that goal, even such a little thing as collectibles – which are flags of the Borgia family, so essentially you remove their influence by gathering the flags. The main campaign alongside all the side missions and the economy loop altogether create a very strong focused experience of liberating a city and training up a brotherhood to defeat the oppressors. This is the exact reason why, even to this day, Brotherhood is considered a favorite by many Assassin’s Creed fans.
Of course, let’s not forget about the other aspects of Assassin’s Creed, like all the UI which is based on the Animus concept to keep it in line with the modern day part of the game, as well as music that incorporates a little bit of electronics in it to evoke the notion that this is not just a historical world, but a recreated simulation.
To sum up, cohesion is important because flawed design that is wrapped into a great experience will affect more players positively, than great design that is wrapped into something dry or contradicting each other.
But there’s still one very important thing that I haven’t talked about yet when it comes to good design. Regardless of how well thought-out your system is, it doesn’t matter if Implementation is lacking. Let’s continue the trend of jumping back and forth between games and get back to the Batman: Arkham series again.
Its combat system has been an inspiration for a lot of games ever since 2009’s Arkham Asylum, and yet, while it may seem that it can be enough to just copy paste the system and it’s going to be fine… it’s not.
Arkham combat system is great not just because of its design on paper, but also because of its implementation and feedback loop.
- The response timing-window is fine-tuned.
- The sound effects are crunchy and satisfying.
- The enemy variety calls for the use of different moves at different points in time.
- The combo counter and available skills and gadgets create a mastery element of the system in which you can learn to go beyond the basics and become a really efficient fighter.
- The camera allows you to view enough of the surrounding area to take note of where your enemies are.
There are games where the Arkham-style combat is implemented great, like the Shadow of Mordor series, but there are also a lot of titles where it’s functional enough to be ok but doesn’t play as well, like let’s say the Mad Max game. The camera there doesn’t provide a good enough view of the surroundings, the mastery element is not as sufficient making the combat seem monotone, and other smaller flaws that together can really affect how the game feels.
And look, the reality of game development is that in the midst of creation there are so many moving pieces that sometimes you can’t get a feature to a certain level or the things you thought would do that didn’t actually fix anything. The sometimes sad reality is that games start taking a real tangible shape only close to the end of development cycle. Good implementation is important, but it’s also the aspect that’s easiest to evaluate in retrospective.
Anyway, I’ve been talking for quite a while so I think it’s a good point to start summing things up. The following are the four main points that define Good Game Design.
- Good Game Design is Contextual – whether a design decision is good or not can’t be defined in a vacuum outside the context of the game you’re making.
- Good Game Design is Deliberate – however small or insignificant, there must be a reason behind a particular design decision, otherwise it’s just blindly throwing things into the mix.
- Good Game Design is Cohesive – together with other parts of design, as well as other aspects of the game, it builds a wholesome experience for the player.
- Good Game Design has Good Implementation – it’s not enough for an idea to be good on paper, it must also be good in the game proper.
Everything else lies in details of the context you’re working with. You can be making a platformer, but is it:
- A platformer based on jump timing and precision? (Super Meat Boy)
- A platformer based on time travel elements and puzzles? (Braid)
- An atmospheric platformer with environmental puzzle-obstacles? (Limbo)
Even though the basic mechanics are the same – you run and you jump, the best design decisions are different for each particular case. But as long as you define and know what experience you want to provide and strive for that, and evaluate your decisions based on the broad principles defined above, then I’m sure you’re going to be fine.
You could be asking though, well what about the theory of Flow, the Magic Circle, player motivation psychology, good learning curve, and everything else that you usually keep in mind during development – doesn’t that all also define the quality of design? The devil is in the details, after all.
Well, kinda. Those things are important, but not as much as the big picture. Look, I’ve been officially in the industry for a relatively short amount of time – a bit more than 5 years. So I don’t have the wide breadth of experience of somebody like Ernest Adams, Raph Koster or Warren Spector.
But I have seen that the best design decisions come out of less, one could say, tangible questions: “What do we want player to feel?”, “Does that fit our experience?”, “What will the player remember?”, “What is the emotion we’re going for?” These are all not very technical questions, but they’re very human, and that’s what really drives good design forward.
Thank you all for watching. A special thank you goes to my Patreon supporters, Paolo di Stefano and Comissar Doggo. And if you’d like to join them, feel free to support my campaign at patreon.com/farlands.
What do you think are the aspects of good design, and why? Do you disagree with the principles I’ve listed? Feel free to leave any comments below, I’ll be happy to participate in a conversation. And if you have enjoyed the video, hit up that like and subscribe button. Thank you very much!