Choosing Your Inputs and Avoiding Overwhelm

If you’ve read my blog recently, you’ll know that not long ago I discovered Seth Godin and his genius wisdom. At one point in his interview on the Tim Ferriss podcast, he talks about controlling your inputs in order to overcome a feeling of being overwhelmed. In fact, not controlling your inputs can lead to overwhelm, stress, ignorance, dependency, and more, so I’d like to talk about this idea today.

Everything you read, hear, see, and do comes to you somehow as an input. The news you read in the morning, the stories your friends tell you, the tasks you choose to do at work, the art you choose to go see, these are all inputs that you engage with. But what happens when these inputs are too many, too time-consuming, or too demanding? We feel overwhelmed and stressed. What I’m about to discuss can apply equally to your work life as it can to your personal life, and I find that this is a problem I’ve faced while running the business side of my game development studio.

Many inputs!

In Godin’s words, it’s a systems problem. We’re not controlling or managing the inputs we have in an efficient or effective way, and it leads to stress. The imagery he used (which I absolutely loved) was that “drinking from a firehose is a really bad way to get hydration. It’s a dumb choice to drink from a firehose,” even if it could hydrate you. I, of course, pictured someone grabbing a firehose, turning it on and carefully trying to get their lips in the stream only to have their face blown sideways by the ridiculously high pressure water over and over again. So how do we manage these systems?

Removing and Rebuilding Inputs

The first step toward diminishing stress from overwhelm is to remove all unnecessary inputs, to the best of your ability.

In Seth Godin’s case, he says that he doesn’t have a television, he doesn’t use Facebook or Twitter, and he doesn’t go to meetings. Not every input can be removed of course, but many of them can. When I say unnecessary inputs I mean anything that you have control over, even if it’s something you enjoy and feel that you need. Then, you can properly start from zero.

Zero inputs!

The next step is to add carefully picked inputs back into your life. By choosing the ones you care about most, you may realize that some of the things you do are done without you really wanting to. Again, I’m not talking about the things you don’t have a choice about, but rather the things that you do have a choice about but you might not realize it. Once this breakdown and rebuild of inputs is done, you will almost surely end up with less inputs than when you started.

Say No to More Things

In both business and in our personal lives, a lot of us (myself included) have trouble saying no. We think we’ll be missing opportunities or that we’ll offend someone, but I’ve found that saying no to things frees up my time and energy for much more useful things. I almost never miss much, and I’ve always found people to be more receptive to “no” than I thought they would be.

More specifically on the work side: do you really need to follow that business lead? How long will it take you to look into the new software they’re offering? Who told you about the software? How long will it take you to learn it if you do follow through with using it? For our company, I try to make sure that we choose the inputs we see when it comes to things like software. We don’t just open our ears to anyone selling us anything, we seek out a solution to a problem we have only if it’s an actual problem we actually have. Otherwise, I try to defend my team from the noise firmly and unapologetically.

Prioritizing Inputs

Sometimes we feel like we just have too much to do. Maybe this concerns work, maybe it concerns our social lives, and maybe it concerns hobbies. We’re overwhelmed, and we don’t know where to start. I can think of two good ways to prioritize these inputs: starting small and pushing past discomfort.

Starting small is the best way to get past any kind of procrastination, and is an equally valuable strategy to get started when feeling like we have too much to do. When you look at a messy room with clothes everywhere, dust on shelves, and papers on a desk, you can start by choosing a small corner, working on it until it’s done, and moving along from there. This is an over-simplified situation, but it’s an example of a negative input that you didn’t choose and you need to deal with. In the past, I’ve written very brief outlines for documents I’ve had to prepare. This was as simple as starting a 90 slide presentation by writing six bullet points on a page. That was my first bit of work on the presentation, and that’s all I did on the task that day. From there I could go back another day and put more bullet points between the ones I had already created, and start moving along from there. The tiniest commitment to the start of a task can make a huge difference.

The second situation when prioritizing is to decide whether or not you have the energy to push past what might be a negative or difficult input. In Godin’s (paraphrased) words:

Do I care enough to experience discomfort to get to the other side? If I don’t, then I should turn off the input. Because sitting with an uncomfortable input when we don’t care enough to make things better is just a formula to be unhappy.

As I mentioned above, sometimes it’s not worth the effort to deal with the input. Oftentimes, the seemingly impossible-to-turn-off inputs can actually be turned off.

Control Your Inputs on Social Media

This one relates more to our personal lives, but could have applications in work as well. In our everyday lives, assuming you’re using social media, you’re being bombarded by ideas and thoughts and suggestions of what to do, where to go, how to dress, and what to think.

As much as Facebook’s algorithms are good at serving you information that you already want to see, you can’t stop there. If you do, you risk complacency and ignorance. Facebook, Google, etc. do not control your inputs, and I find that blaming them for your lack of balanced information is irresponsible. You actively choose to use their platforms to get your information, and even within those platforms if you stick to only the things that are being served to you, you will end up missing out on finding greater knowledge and truth.

Read opposing views of news stories that involve conflict between two sides, do the fact-checking on topics you’re going to consider sharing with others or being passionate about, stop spending time scrolling through your personalized news feed, stop falling into YouTube holes watching silly videos… these are all ways of reducing your inputs to things you actually control.

Seek Out Inputs, Don’t Let Them Seek You Out

To conclude, the common thread in this whole post is that you’ll be better off if you actively choose to seek out specific inputs. If you can avoid inputs that throw themselves at you and more purposefully seek out inputs, it can lead to less stress, less feelings of being overwhelmed, less time wasted, less ignorance, and less complacency. Generally speaking, this will create space and time for you to improve in the way you want to improve and live your own life, and not the life someone else tells you to live.

From Funnel Marketing to the Teacher-Student Dynamic

I wanted to talk today about two opposing concepts in marketing and how they might apply to game development. This article is strongly influenced by Seth Godin and I highly suggest reading some of his work (his blog, TED talk about the spread of ideas, his book “This is Marketing”, ) if you haven’t already. I’m going to try to discuss what I learned from him in a recent podcast and apply it to game development in a way that hopefully is instructive to some.

The marketing legend and wizard, Seth Godin

The first concept is the traditional view on marketing, which is the idea that you start out trying to reach the widest audience possible, and some of those impressions become clicks, some of those clicks become further clicks, and eventually some of those become customers. This funnel method is how TV ads, web page takeovers, and other mass impression strategies work. According to Godin, this is what happens when young and inexperienced marketers have money to play with, and this method is on its way out in many markets.

You may be already saying to yourself: “but I don’t do that kind of marketing, plus we don’t have the budget to do that kind of thing even if we wanted to”. Unless you’re a massive AAA company, that is likely true. But even without paying for TV ads, you may be finding yourself analyzing your follower growth under a microscope, paying for Facebook ads to get page likes, or trying to plan how to get your Reddit post on to the main page of the Gaming subreddit. These are all things that use the approach of “how do I reach the biggest possible group of people and pick some off as future customers”. What I’d like to propose is another way to look at the problem.

Seth Godin calls this the classroom method, or at least refers to a teacher / classroom dynamic when describing this. The people who like your game(s), buy your games, join your Discord communities, and come to meet you at expos, these are the students in your classroom. Remember your favourite teacher in school? Probably not the most conventional, and probably not everyone’s favourite teacher. But this teacher had a style and a way of teaching that you liked, and they surely showed that they cared about you as an individual. As Seth describes, your goal shouldn’t be to produce the most appealing thing to that giant funnel and hope you get some customers in the end, it should be to

cater to the students in your class who are already
paying attention to you, who are there for you, and who would
be disappointed
if you didn’t give them something interesting.

In an age where we fight not only for attention but for trust, we need to focus on serving the customers (gamers) a high quality experience that doesn’t try to please the masses. By doing our thing, doing it well, and caring about the people we do it for, we can actually create greatness which appeals to a larger niche than we expect. By dumbing down our content and product and trying to hit ten million people, we risk not creating something that’s special to anyone. If you can’t capture the single person and make them passionate about your product, you won’t be able to do it for the masses either.

Image by Vectorpocket

This is not to say that we don’t need to have a lot of people see our work in order to have financial and critical success. What it encourages is changing the focus, and in focusing on the small, personal aspect of what we do, we have a better chance of achieving that wide market and critical success.

This is a massive shift from thinking of your target audience as all the people who don’t already like your product to thinking of your target audience as the people who you already know like your product and trust you.

These are individuals with personalities, thoughts, feelings, and lives and by respecting that you give yourself the best chance to succeed.

Your goal should be first and foremost to make a quality product, and then to serve your customers (in this case, your game fans) in a way that makes them want to come back and makes them want to talk about you. By focusing on this smaller core and caring about them at the smallest level, you create an atmosphere that is highly valuable.

I think this new approach is especially instructive when it comes to the community and has a very clear link to the development and sustenance of the community around your company and game. The goal is to create the experience (or product) for that person in your classroom that makes them feel like they’re getting value from it. The value one gets from being part of your game community can be many, many different things. On the surface, it can include fun and exciting content, new people to play with, direct communication with developers, new strategies and play guides, etc. On a deeper level, this community can also provide: comfort, excitement, emotional support, friendship, and other more fundamental human needs. We need to take special care to offer whatever value we can to our community and make sure that our players know that they’re appreciated and taken care of. If you treat people right who are already your in classroom and who are willing to learn, the word will spread because these people will feel truly special. If you advertise that anyone and everyone could enjoy your community because it pleases everyone, you’re likely to 1) miss that audience because it’s not offering interesting value, or 2) appealing to people who you don’t even necessarily want in your community.

What about with respect to the game or game design itself? I’m not entirely sure where I stand on that, so if people have additions to this please feel free to write in the comments… but I’ll give it a shot. I think that intelligent, consumer-facing design decisions should do the trick here: intuitive UI, proper difficulty curves, in-game moderation, etc. These will be done differently for different games, but should always be done to serve the small classroom of people who have come specifically for you (your game in this case). Making something interesting, exciting, and different is a way to engage this classroom in the same way your favourite teacher growing up did for you. By making something that is so inoffensive that you think no one will be turned off by it, you try to reach this massive audience and it could lead you to making decisions which alienate your biggest and most dedicated fans who are also your champions. Where this becomes complicated with respect to the classroom model is that I’m not sure if the classroom is already established or if the classroom gets filled with fans of your game. That is, are you creating a product to serve these students or are you creating a product and the students come to see you?

I hope this blog post was useful to some people in helping them focus on the smaller-scale, person-to-person care that goes with creating and sustaining a good community. In our company, we will continue to focus on this approach and hopefully the word will spread about our game and community.

Planning and the Unknown Unknowns

Last November, I gave a talk where I described the story of our video game Ultimate Chicken Horse from early conception to release and beyond. My goal was to amalgamate the answers I’ve given to people when they asked about different parts of the story: “how did you get funding? How did you get your game on PlayStation? How did you find your partners to start your company?”, etc. The format was to stop the story at each point where I felt I learned a lesson, and share those lessons with the audience.

One of the lessons that I drew from the chronological story of this journey was to try to think of everything. It seems obvious of course, but it seems like you couldn’t possibly know what you weren’t thinking of, because obviously you weren’t thinking of those things.  So given that “try to think of everything” is a bit hard to act on, I now use a new phrase: plan for the unknown unknowns.

https://encuentro.gr/wp-content/uploads/2016/10/023_CNT1_03.jpgWhen you don’t plan, you end up with half-built things.

This phrase is taken from a book by psychologist Daniel Kahneman called Thinking: Fast and Slow, in a section where he describes the planning fallacy. The planning fallacy is used to describe plans and forecasts that “are unrealistically close to best-case scenarios” and “could be improved by consulting statistics of similar cases”. The result, of course, is that plans take longer than planned, costs go way over budget, people get unhappy, products under-perform, and everybody loses. Sound familiar? Game development, and surely most other fields of work, are plagued by the planning fallacy.

I’m going to focus today on three proposed ways to mitigate the planning fallacy:

  • Establishing a baseline
  • Taking the outside view
  • Planning for the unknown unknowns

Establishing a Baseline

Kahneman and his partner Amos Tversky did a lot of research into many topics in behavioural economics from a psychology perspective, and eventually won a Nobel prize for that research in 2002. After coining the term planning fallacy, they offered a solution of how to solve it by establishing a baseline.

The idea is to avoid the common tendency to neglect the statistics of similar cases to yours. To use game development as an example: “Most MMOs take four years to make with a team of our size, but we have industry veterans and we’re really well organized, so it will only take us two years”. That’s obviously not a very forward thinking mentality, but it happens to all of us whether we realize it or not, for small tasks and large ones alike.

This idea was eventually formalized and given the name reference class forecasting, and it works in the following way:

  • Identify a fairly vague appropriate reference class (indie games, platformer games, online games, games made with X people, games made with Y budget, etc.)
  • Obtain statistics from the reference class (how long did these other projects take? How much did they cost?)
  • Base your predictions on the stats from the reference, then use specific information about your case to adjust the baseline prediction

You might only realize that you’re the little duck after establishing your baseline

Often, you may find, that you might have to stray away from the baseline prediction by increasing it, not decreasing it. What are your resources like? Does your team have experience with this kind of project? Are there new technologies that can quicken the process? Are there new technologies which need to be learned, which may slow it down?

It sounds so obvious to look at statistics from our surroundings, and yet we all forget to do it and rarely catch ourselves forgetting.

Taking an Outside View

The next thing to do is to take an outside view, and to step back from our situation. We naturally take an inside view by focusing on what we have, what we’re doing, and the experiences we’ve had with regards to the situation. We extrapolate from what we know, we reason using small bits of data, and we get caught up in emotion when making decisions about ourselves and our plans.

The easiest way to get around this, and the way that has worked for me and for my company in the past, is to ask others. Find people who will give you straight up, no-bullshit feedback about your plans and ask you the tough questions that you’ve likely been ignoring.

Asking others means you’re no longer forecasting based on information in front of you, and gives you a more complete picture.

Planning for the Unknown Unknowns

Remember the example about the MMO that was only going to take your studio two years? We already discussed that it’s not the smartest thing to assume that it will take you less time. But most people will take an extra step and actually plan; they will thinking of all of the things they can, take an outside view, ask others, and even look at comparative projects to make their estimates. Once they’ve done all of this, they will sum up all of the things they think they need to do, assign times to them, plan using current and future expected resources, etc. After this whole process, they’re still left with a little over two years in their plan for this MMO project.

This is because they didn’t think of the unknown unknowns. These are the things that can come up mid-project: bureaucracy (and boy do we know about that one in game development), illnesses, divorces,  technical delays, dependencies on contractors, change of personnel, people moving, etc. etc. etc. As I mentioned in the introduction, you can’t know what you don’t know, so it’s hard to plan for it.

This is why we add contingency in our budgets, and why we should add a hell of a lot of contingency time to our plans when starting projects or agreeing to deadlines. We also try our best not to promise anything before it’s ready; many companies (and individuals) run into problems when they can’t meet deadlines for deliverables, but often these dates are self-imposed and do not need to be so fixed. There are, of course, cases where a client is dependent on you, or a project needs to reach a certain milestone because of timing with a season or sale, but I’ve seen many self-imposed deadlines set up by the “suits” for no apparent reason, and this can cause unnecessary stress and perceived failure due to missing those deadlines.

Beyond asking other people what unknowns you might run into (as was the case when taking an outside view), you can also study other projects or companies and see what kind of issues they ran into. Even if you don’t expect to run into the same exact issues, there’s a good chance that it will inspire you to think of potential issues for your own situation.

So first, think of everything you and all of your peers can think of, and then plan for the things you haven’t yet thought of.

Why is this Important?

This is relevant to anyone in a management position or anyone who is making decisions about planning.  In video game development, every single project I’ve ever heard of in the history of games has been late. If it wasn’t late, it was shipped at a way lower quality than it should have. I’m not sure if other industries are as notorious for delays, but I imagine it’s a common issue across the board.

While none of these suggestions are hard science or give you concrete steps to take to ensure success, they should help guide you in a way that can help prevent failure. As I read this section of Kahneman’s book, I realized the direct application that this psychology could have to the game development world and thought I would share, so I hope you enjoyed reading.

If you have comments, please feel free to leave them on the Gamasutra article here.

Thinking Fast and Slow, by Daniel Kahneman
Note: the idea of “unknown unknowns” is originally attributed to Donald Rumsfeld, as Kahneman states in his book.

Daniel Kahneman: Beware the Inside View

 

Is Anything More Important Than a Game’s Consistency?

Hello friends!

I’ve been thinking for a while about what makes a game successful, or rather what makes a game unlikely to fail. My definition of success, in this context, is simply not failing. That is to say, a game has been successful if it can earn its money back and not be a financial and critical failure. I’m not looking for a formula to solve game development, because I don’t think that exists. But I do think we can mitigate our failures by keeping some important points in mind, and I’ve been trying to discover what these points are.

I don’t think we can predict success if we define success as selling 1M copies or having your company bought by Microsoft… but I like to think that we can identify some key properties that will make a game unlikely to fail. This is a hit based industry and I would be a multi-millionaire if I knew which games would be critically acclaimed monumental financial successes… but I’m not. And I won’t pretend that I know the answers to what makes a hit, but I will throw an idea out there to be pondered within my definition of success. This idea was created in hearing tons of game developers talking about their failure stories, and I think there might be something to it:

Consistency is one of the most
important predictors for a game’s success.

That’s my hypothesis, and hopefully it can be discussed, debated, and refined or refuted. I am by no means an expert in this subject and don’t intend this to be purely informative, but I think this is a point worth considering and discussing, and that discussion may help us achieve a deeper understanding of our craft.

I should add, before we get to definitions, that a crappy game won’t succeed regardless of how consistent it is. I can’t claim to know where exactly how high that minimum quality bar is set, but the idea is that many games that could have had success ended up not finding it, mostly due to a lack of consistency.

My Definition(s) of Consistency

When one creates a game, they create a universe. They create at the bare minimum a visual style, a set of rules of physics, a soundscape, interactions between parts, a user experience, and a universe. Beyond that, game developers may create mechanics, rules of interaction between systems, stories, cultures, languages, customs, and more. These systems need to be consistent with how the game is talked about, consistent with one another, and consistent in themselves.

I’d like to consider three kinds of consistency here, and then we can discuss whether this is actually a strong predictor for success (i.e. non-failure). Personally, I think the order I’m presenting them is their order of importance.

Marketing Consistency

We talk about and pitch our games in different ways, on different platforms, and to different people. We try to explain our entire game in a catchy sentence, we show a screenshot, or we make a trailer that tries to convey the player’s experience in our games. Unfortunately, we often break the consistency of the universe we’re attempting to create even in the way we pitch or market out games; we show off static shots of cute characters when really the in-context animations are what give them their charm, we talk about procedural generation in a rogue-like when the real fun comes from the combat mechanics, and we talk about open-world and crafting when the best part is the story.

From a failure workshop talk that Hugh Monahan of Stellar Jockeys gave at Full Indie Summit in 2016, he talks about his early access trailer for Brigador: “The problem is that it looks like a twin-stick shooter. Brigador is anything but a twin-stick shooter.  For some games, Brigador included, the subjective experience of playing the game is totally different from what it looks like watching a video of gameplay or watching somebody else play.” He goes on to explain how the feel is different from what it looks, and how much deeper the gameplay is than it looks: “This isn’t the game I was sold”, is what many players were saying after playing it. “Because this game looked close enough to existing tropes, or existing genres of gameplay, a lot of what was creative and unique and different about Brigador got completely wiped by this instinctive ‘oh, it’s a twin-stick'”.

Now, I know that anecdotal evidence isn’t going to prove my point, because for any example I give I’m sure others can be found that were inconsistent in their marketing but still did well, but I think this shows how this inconsistency can be dangerous.

Genre Consistency

Players have certain expectations about your game based on other games they’ve played in the genre, other games they’ve played with the same art style, and other games that list similar mechanics or features to yours. If a game is a hardcore strategy game, it needs to look like a hardcore strategy game. If it doesn’t, it needs to be made abundantly clear to players why it doesn’t look like they would expect it to. I should add that we have to be careful when making games, because any hint that our game is similar to another game or similar to games in a genre will be picked up by players, and the expectations start to creep up.

This is not to say that we should make games that all look like one another… absolutely not. But I think we need to be aware that player expectations exist regardless of what we want, and that making a deep strategy game that looks like the image below is setting yourself up for an uphill struggle of trying to change player expectations that are already pre-established.

While I haven’t played the game yet, there’s a pretty wide agreement that Yooka-Laylee didn’t live up to expectations. I don’t think it sold as well as was anticipated, and its review scores for metacritic were not too favourable. The issues that reviewers talk about always include the idea that the nostalgia element was good, but the game brought with it all of the annoying things from those old games like Banjo Kazooie: the camera, the pointless currency collection, etc. I’m not sure the Yooka-Laylee team could have prevented this, since the expectations were set high as soon as Playtonic mentioned that they were making a 3D platformer. But at the end of the day, people expected something more than what they got.

Just as a quick aside, I know and respect both of these companies whose “failures” I’m highlighting, and I still look up to them as game developers despite using their games as “failure” examples.

Game Consistency

I touched a bit already on what game consistency is comprised of: things in the game cannot contradict what other things in the game say, how they work, or how they feel. Within the game itself, the mechanics and everything that you’ve created in your game universe need to be consistent and predictable.

A rule that has been established and conveyed to the player cannot be broken by the game, otherwise the game risks losing consistency. If the game is set in the year 2093, the font used in the menus shouldn’t be Times New Roman (unless there’s a damn good explanation, and even then, why are you using Times New Roman??). If the player plays as an extremely kind-hearted, benevolent, peaceful person, they shouldn’t be killing in cold blood in the next level. If the player can always grab ledges in a platformer, there shouldn’t be similar looking ledges that can’t be grabbed. The example here is in Zelda (in every Zelda game I’ve ever played, in fact), a cracked wall means you can bomb it. It never, ever means that anything else, as this would bring the player out of the experience and lead them to question the rule they learned, that all cracked walls are bombable.

We teach our players the rules of our universe in many ways, and if we ever contradict the rules that we established or the rules that players believe we established then we break the immersion and we create a bad experience for players.

Why Do We Care?

Consistency is key, in my opinion, because

player expectations are created by consistency and those same expectations are shattered by inconsistent marketing and game design, leading to a bad experience.

These player expectations are created by you, your universe, trailers, screenshots, menus, game mechanics, art style, website, and everything else that has anything to do with the game. These expectations are also created by preexisting genre tropes, and anything your game does has to be aware of those preconceived notions.

This is why, in my search to find some properties of a game that will help it to avoid failure, I’m pointing to the idea of consistency as a proposed indicator.

Some Good Examples

I’m going to give some examples of some projects that I think achieved the consistency I’m discussing. As is the case with any argument, using anecdotal evidence is not a strong way to provide “proof”… but I’d just like to demonstrate some strong consistency examples after talking about a couple of weak examples in the above sections.

To provide an example of success, I present the game Shovel Knight by Yacht Club Games. That game promised retro, old school, challenging gaming and that’s what people got. What we also got was the innovations in new games (save files, responsive input, longer game, etc.) while not ruining any of the old stuff that we found so charming. Plus, everything about the game kept you in its world by being consistent: the music, the level design, the art style, the menus, the sound design, etc.

Firewatch from Campo Santo is another good example. First I should say that the trailer for Firewatch is a work of art. This is just one of the trailers they made, but they all seem to be consistently amazing.

It gives you every feeling that you’re going to feel while playing the game: suspense, discovery, relationship building, fear, relaxation, everything. This really sets the tone for the game, and it doesn’t fail to deliver. The music, the art, and everything else about the game reinforces this core point.

Like I said before, finding these examples doesn’t “prove” my point, but it can help illustrate why I think consistency might be a major factor in determining success of a game.

Suggestions to Improve Consistency

I can think of a few ways to try to ensure consistency, some harder than others.

Make sure your trailer conveys how the player will feel when playing your game.

M. Joshua Cauller has a great article about this on his blog. This is probably the consistency issue I’ve seen most: developers will create trailers (or worse, have trailers created for them by other companies) which don’t properly explain what the player experience is like. The trailer might completely miss the mark and focus on something that the developers find interesting, but that isn’t the real thing that makes the game fun. Sometimes, trailers can even be good on their own accord, but not linked to how the game makes the player feel. For example, if your game is interesting because of the flow and precise shooting and movements, creating a story-heavy trailer that doesn’t show those elements might cause people to expect something very different from what you’re providing… that lack of consistency leads to bad reviews.

Get an artist.

Probably not this one…

I’m not an artist, and I wish I was better at this… but people need to have an artist with a good eye look at their graphical elements. I’ve seen too many games with strange menu fonts, colours that don’t match the theme, UI elements that look like they came from Hearthstone in a game where the rest of the screen looks like Fez, etc. Strangers and other developers will can you if your visual style is inconsistent… leading to the next point.

Show the game often, and show it early.

By soliciting player feedback early, you practically ensure that you’ll catch the major issues before you get too far. Showing the game to other game developers, artists, film people, designers, architects… all of these will help you to understand the consistency of your art style and your game in general.

Be aware of everything that is in or related to your game.

I might just be picking patterns out of nothing here, but I find that often the inconsistency I’ve seen in games is linked closely to outsourced work that wasn’t well monitored. Art asset creation assigned to other companies, completely outsourced trailers, and far removed audio teams could contribute to this. If we are careful about the details about all of the things that are going in our game or are related to the game’s universe, we may be able to mitigate some of that.

What Do You Think?

Do you agree with that consistency might be a good predictor for success, above a certain low quality bar? Do you disagree? Do you think I’ve missed something important and shouldn’t be focusing on consistency? Do you think I might be on to something, but misinterpreting it? I’d love to hear what you think, so feel free to email me or discuss on Twitter, or leave a comment on the Gamasutra version of this article!

Thanks for reading, if you’ve made it this far! 🙂

 

My Thoughts from E3

Hi everyone!

In case you aren’t in the video game world, or if you are in the gaming world but are living under a rock, E3 happened last week! E3 is the annual game conference where all of the big companies announce new games, release dates, and general hype stuff. Here are a couple of the coolest things that I saw, and some reflections about those things.

The first piece of news that stood out to me about E3 this year was Microsoft buying Compulsion Games along with three other studios, and opening a studio of their own called The Initiative. The reason that stands out is because we’re close friends with the folks over at Compulsion, and they helped us get off the ground as a studio and figure out how to exist in the games world from the very start. We’re super happy for them, and super proud of the Montreal community that helped spawn them (or that they helped spawn, really). Beyond our happiness for Compulsion, I think this marks a pretty big shift in focus for Microsoft. The days of larger companies (Microsoft, Sony, Ubisoft) buying small studios was rumored to be over, and I think this proves otherwise. To me, this is just another step Microsoft is taking to show its dedication to the indie or mid-level (some would call it AA or iii games) studios. You should really check out Compulsion’s new story trailer for We Happy Few:

Nintendo showed some amazing stuff in their E3 video, as usual. Most notably among them was a long segment about the new new Super Smash Bros game, which includes all characters that have ever been in any Smash game! You can tell by the amount of time and focus they spent on it that they aim for this game to be the next Smash Bros Melee, and aren’t going to be discounting it for future competitive tournament play. They also showed a new Mario Party game, and announced a million things that will be coming to Switch, including Fortnite. I think the inclusion of Fortnite along with these other games will be huge for the continued success of the Switch.

Also, Overcooked 2 was announced with online multiplayer! If you liked the first one, this one is similar but has more features such as throwing food items, more dynamic levels, and of course online play. This is an interesting move for a couple of reasons; first, the choice to make a sequel instead of provide free updates to the existing game is one that many indie studios are not making. The aversion to sequels has been described by some people in the industry as being very silly, seeing as sequels almost always sell better than their original counterparts if the first game was a success. Why we tend to avoid sequels is a huge question that could probably justify a whole other article, but we’ll leave it at that for now. Second, the addition of online multiplayer to a game that performed well despite having only local play is an important one that emphasizes the idea that a local multiplayer game simply doesn’t have the potential to do well in the current game landscape. Additionally, it might show that it was easier for them to rewrite significant portions of the game and make a new second game rather than make adjustments to their current one, a strategy that many indie studios have avoided (much to the chagrin of some busy programmers).

Apart from those things that were most important to me in terms of announcements, I was excited to see a Cuphead DLC which I will surely buy, and I might be interested in getting back into the Tomb Raider series if I can get over the fact that the games tend to be more like movies than games, and just enjoy the ride.

There are a bunch of big announcements that I totally skipped because they simply aren’t the kinds of games I would play. Fallout ’76 was announced, a new HaloForza 4 and about a million shooters that will probably be uninteresting but will surely make millions of dollars. You can check out some pretty in-depth reviews of each company’s press conference on any of the big press sites: IGN, Kotaku, GamesRadar, etc.

More posts coming soon!

Reflections and Lessons from GDC 2018

For the last 4 years, one of my favourite times of the year has been the week in March marked by the Game Developer Conference, or GDC for short. It’s the biggest game conference in North America, and attracts a bajillion extremely interesting and inspiring people. This year, we were fortunate enough to bring the whole team. Our goals included some team bonding, learning, keeping up business relationships, and more. I definitely think we succeeded, and wanted to share a few of the most important things that I learned. I hope this post will be useful to game developers and non-developers alike.

First off, talking to the amazing devs at GDC for me thinking about Ultimate Chicken Horse and about its future. I’ve had a bit of a feeling that we’ve been working on the project for a long time, and I want to start working on new stuff… but on the other hand, the game is doing well, the community is great, and there’s still a lot of potential. So what’s next for the future of UCH? I didn’t outright ask people their opinion on this, but it was somewhat obvious that it is / was on my mind, so I got a lot of feedback on it. Do we want to go more casual and community-heavy? Do we want to add more mechanics to level the playing field, like Mariokart-style? Do we want to go more competitive? What’s required if we do that? How can we improve the tech? What does the community want? Do we have the funds to hire more people, and if so, what will they work on? I won’t go into too much detail here about my thoughts on the matter, because I don’t want to get anyone’s hopes up (or down) before the team talks about it and decides what’s next, but it’s definitely on my mind more than it was before.

Many, many games come from game jams or quick prototypes.

It feels like the majority of the successful indie games that I saw at the conference started off as game jam ideas. This was the case for Ultimate Chicken Horse, and it doesn’t really surprise me that it’s the same for many other games. It seems like game jams are a good way not only to practice skills, but also to come up with great ideas. There’s some fairly common wisdom that it’s easier to be creative given some constraints, and without thematic or timing constraints I think it can be easy to stare at a blank page forever, waiting for the next revolutionary idea to pop out of your brain. Even in the AAA studios, some of the games came from quick pitches from someone who wasn’t an owner or creative director at the company. These were as simple as a short presentation with some mechanics and some concept art, and they were off to the races. Of course, I should emphasize that it’s absolutely critical to be able to kill the project early on if it’s not working, but that’s the case whether it came from a game jam or any other method.

Ask. Just ask!

We’ve been pretty good at this, as I’m somewhat shameless when approaching people for help. But it works! You’ll never know how much you could be missing if you don’t ask other developers, publishers, platform holders, friends, family, etc. for help. The industry has a lot of wisdom that it’s very willing to share if you’re able to overcome the shame of not wanting to bug people… so do it!

Different ways work for different people, there is no right way. This point is actually what my talk with Tanya Short (Kitfox Games) was about, though we looked at it from the business side. We talked about general things that should be considered: burn rates, revenue sources, projections, diversification of studio into multiple projects or not, etc. But I also spoke to people about the creative process, and it was interesting to hear some pretty opposite views.

The big one that stood out to me was when I was talking to someone about coming up with content updates. He said that he tries to envision what the trailer for the update will look like– that is, what’s interesting that the public will latch on to, before starting work. The idea there is that a patch with a bunch of bug fixes and user interface improvements isn’t enough to get people excited, and the trailer helps guide the production toward something useful and exciting. On the other hand, some people like to go the more organic route and play around until they find something that works well. It’s not marketing-driven and is easier to get early feedback on, but whether one way is better than the other is really up to the studio. And as with anything in this industry, many different strategies can work!

You need to build a community before launch.

This is actually a bigger topic, and I’m going to write a full article on Gamasutra and my personal website about this sometime soon. The basic idea is that you give yourself a much better chance at success if you’ve created a community around your game before the game launches.

On a less educational note, there were a couple of highlights that I wanted to point out from the conference this year, that include talks and just general feelings.

The first one of those was my favourite talk, the ‘Composer Confessions 2’ session. The talk was done last year as well, and it brings together five composers to talk about some times that they’ve screwed up and what they learned. I’ve always had a very strong personal interest in game music, and I like to write some myself (even though I’m not nearly professional), so it’s really inspiring to hear people like Austin Wintory (Journey, Assassin’s Creed, Tooth and Tail), Gareth Coker (Ori and the Blind Forest, ARK), Darren Korb (Bastion, Transistor, Pyre), Peter McConnell (Hearthstone, Psychonauts), and Gordy Haab (Battlefront 1 & 2, Halo Wars) talk about their craft.

I only mentioned a couple of the games they each worked on, but there’s a ton more and these guys are absolute legends in the field. Some of the main lessons included making sure to delegate work and not be a control freak, learning to accept what the client wants even if it’s against what your musical instincts tell you, and learning to show completed examples instead of work in progress because producers can’t imagine the finished product in the same way musicians or the composer can. Beyond that, the talk was hilarious. Each person had a 10 minute slot to talk about whatever they wanted, and somehow they all ended up being hugely entertaining and funny.

Another thing that I really liked was THAT Party, a party I hadn’t been to before because tickets sell out super quick and I had other, more “businessy” parties to go to each year. This time I went and I found it really nice to see and meet some indies that I haven’t met, but also to be able to “party” in more of a traditional sense, with drinking and dancing and such. This isn’t because I’m a natural born party animal, but rather because I like the idea of moving from the business contact mentality to the friend mentality, so a mix of that combined with more professional cocktails was nice this year.

Beyond that, I feel like our team had a good chance to bond– not necessarily all of us at once around a table, but in pairs that would split off as we walked places, shared hotel rooms, and talked about non-game stuff together.

Alright, so it looks like this post has become huge and I should probably stop writing before everyone falls asleep. Thanks for reading if you made it this far, and I’ll have more articles coming soon so check the website or follow me on Twitter @RichMakesGames for updates.

Some of My Favourite Design Elements in Super Mario Odyssey

Recently, I’ve been trying to look at games through more of a designer lens. In order to improve my design abilities, I think it’s important to look at games as a designer. This means thinking about the choices that were made while designing the game, the levels, and the systems, as well as trying to understand what an average user would experience in your position.

That being said, I finally started Super Mario Odyssey last weekend. I know, I’m late to the party. If you’ve ever played it, you already know that it’s great. But I wanted to talk about some of the things that I think the average gamer might not have noticed, but are really fantastic design elements or layers of polish. These are things that even independent developers on a tight budget can do in their games, so I think they’re important to talk about.

The first two involve the level design, and the clarity with which the player is directed through the level. Firstly, players can always see where they need to go without looking at a map, and yet don’t feel forced to follow a path because of the opportunity for exploration. The thing is, the opportunity for exploration is very carefully given to you. When the game wants you to go somewhere and not step out of line, it shows you clearly. In the below image, this is the part right before encountering a boss. Behind Mario is a ton of open space to be explored, and in front is a clear goal without distractions or options:

The game strikes a nice balance of making you feel like you can explore and wander off and discover things on your own, while still giving you a very linear path to follow, similar to how Ocarina of Time on N64 felt.

The second level design point is that the game doesn’t just aimlessly add places to explore in the levels. I’ve seen some games where it seems like the mentality was “the level looks too much like a boring square, let’s add some nooks and crannies”… and it’s awful.

In Odyssey, every time you explore a place you haven’t seen before,
you get rewarded.

These rewards can be seeing cool stuff, finding things you’ll interact with later, finding a Power Moon, or finding a mini-game (which leads to a Power Moon, as shown below).

Another game that did this really, really well was the new (2016) DOOM game. The level design was obviously quite a bit more complex than in Mario, but it trained the player well in the same way. It gave the player options to explore that were somewhat hidden, but once the player had decided “okay, I’m going to check this out” it always gave a reward. In the image below, let’s say the pink arrow is the story / linear path, the green path would be shown with some hints or a semi-broken door, and the player would explore.

This trained the player to look for hidden things and secrets (apart from the game obviously telling you that there are “secrets” in the levels), and so when a player felt that there was a hidden path, they would explore it.

This is where DOOM and Mario both do magic; you never have the disappointing experience of “Oh, I was sure there would be something here” or “Why can’t I do that? I expected it would lead me somewhere”. Those feelings are bad for design because they break the immersion by breaking the rules that the designer had created previously for the universe / game / level. Both games reward your exploration and you don’t even feel like exploring in the places where the game doesn’t want you to explore, because it directs you through those places well.

The next main point I wanted to talk about was the purchasing system for outfits. In Mario Odyssey, you can purchase things from a shop at the beginning of the kingdom you’re in. There are two currencies, the regular coins and one that is specific to the kingdom.

Some of the options in there include Power Moons, but also include custom outfits or souvenirs from the region which you keep with you on your ship as you travel throughout the game. This is super, super cool. Character customization stuff is always cool, and seeing it in Mario is certainly unexpected, but that’s not what I find so interesting about this game design choice. What I find interesting is that

The game rewards you based on the effort you’ve put in, leaving you feeling more accomplished than simply having “beaten a level”.

This ability to purchase allows you to choose your reward, choose what you value in the game, and creates a feeling of a collection challenge. Some people like to try to finish everything in a game (they’re called completionists). I’m not one of them, but, with this new feature I really want to go back and get all of the kingdom-specific currency in order to buy all of the cool things in the region. As it stands, I moved forward but made sure I had at least one physical souvenir from each kingdom and that I unlocked all of the outfits; that left me feeling satisfied and accomplished, much more so than simply “you beat the level”… which is easy enough in most Mario games.

As far as polish goes, there were a couple of things I noticed in the first couple of hours of gameplay. Personally, I love it when games throw me off and react to something that I do when I don’t expect it.

When you encounter this sphinx (it makes more sense in the desert level, even though it’s kind of Mexican themed and there are ancient Egyptian style ruins and a sphinx…), it asks you a “riddle” to pass. I tried climbing on it to get somewhere at one point, and your magical hat says to you “Er, Mario? You’re not a hat.” There are a few minor things like this that I found really great.

Another one I noticed was that when you move into 2D mode with Mario wearing an outfit, he keeps the outfit on and they made a pixel art version of this. It’s somewhat logical, but I wouldn’t have been surprised if it just defaulted to the regular Mario… I was pleasantly surprised. Surely there are many other little things like this, but I really appreciate that extra effort that the developers made.

That’s all for this time, I just wanted to jot down some of my ideas and hopefully some other game designers can read this and get inspired by some of the choices made by the Nintendo team. And if you haven’t played it, I highly suggest it… I’ll be getting back to it as soon as I can!