A detailed look at the roots and the challenges of the level design world in an interview with Max Herngren from Mojang.
Everybody knows what doctors, journalists or engineers do. You could just come up with a definition and it’d pretty straightforward. But who are Level Designers? It appears that an attempt to define this profession might give you a headache. We present to you an incredibly detailed look at the world of Level Design in our new interview with Max Herngren.
Hi! I’m Max Herngren and I’m currently working as a Level Designer at Mojang in Stockholm, Sweden. You might know Mojang as the creators of Minecraft and Scrolls and the publishers of Cobalt.
My primary experience before I joined Mojang was working at the indie studio Right Nice Games as a Level Designer on the 3D platformer Skylar & Plux: Adventure on Clover Island. That game was heavily inspired by the platformers of the 90’s and 00’s such as Jak & Daxter, Crash Bandicoot, Spyro and Ratchet & Clank. It was a fairly large game for a small team of only five people. I was responsible for all the levels in that game in one way or another and worked on it up until its release.
Before that, I was actually a student of game and level design at Futuregames in Stockholm where I worked on a number of student projects of which one went on to win “Best Execution in Art” at the Swedish Game Awards, a student competition in Sweden.
During my studies, I also collaborated with Teotl Studios and Sjoerd “Hourences” De Jong on The Solus Project, a survival exploration game where I helped out at the end of the game’s development along with some other students in my class.
I’ve also worked – and continue to from time to time – as a designer and writer at the traditional game developer and publisher RiotMinds making tabletop roleplaying games. I’ve primarily worked on the game Trudvang Chronicles which is systemically a BRP continuation with its own twist on the system coupled with – in my opinion – one of the most elaborate and deep campaign settings out there inspired by Norse and Celtic myths and the Finnish national epos Kalevala.
I got into level design almost by chance, I’d say. When I started studying games I didn’t know what field I wanted to go into. I knew I wasn’t going to be a programmer or artist since I was never good at programming and I can’t draw to save my life. However, coming from designing my own tabletop games since a very young age, game design seemed very natural to me. I quickly learned that being a game designer isn’t easy and most companies out there consider it a more senior role. Sjoerd “Hourences” De Jong – who was my head teacher at Futuregames – started talking to us about level design and I quickly realized that this was something for me. Ever since my early days of playing World of Warcraft, I’d always been more drawn towards the worlds of games rather than their characters and systems. There’s something about the way you can tell a story with an environment that always seemed way more powerful to me than the stories we tell with words.
So, at that point, I decided I was going to focus on level design and that’s the role I took on in all my projects at the school. I quickly found out that level design was a way for me to fit into the pipeline of game production in a way that felt natural and comfortable for me. I was working both with artists and scripters and never found myself doing the same thing two days in a row. I found that I could make a difference and impact the player’s experience in a really creative and flexible way just by using the editor. It was very direct and powerful for me. And I’ve not looked back since.
I think level design is a really hard role to nail down. There’s just so much that you have to be good at in order to be a good level designer and there’s no way that everyone can be good at everything. That’s generally why you often have a team of several level designers doing different things on one game, or sometimes even one level. And apart from just the very different skills that you have to master, you also have to take the studio you’re working at into consideration. How close do you work with environment art and lighting artists? Do you just block out a level and call it a day or are you there for every step of the way? You’re sort of poking at so many different disciplines just by the very nature of what it is you have to accomplish as a level designer. It’s very hard to define where you draw the line and how you actually define what you do. I think you’ll often find that level design doesn’t mean the same thing at every company and that can sometimes be problematic as there can be some disconnect with what is expected of you and what you’re actually capable of.
I do think there are some fundamentals that every level designer has to really be good at and find natural in order to become successful. You have to have a great sense of pacing and player experience. You have to, no matter how tired or uninspired you’re feeling, be able to put yourself in the player’s shoes and try to imagine how they’re going to experience your content. If you’re not able to do that I think you simply won’t be very successful. No matter how much user testing you do, if you don’t have that ability to understand what a player wants at any given time or how they’re being affected by your pacing curve, you’re going to have a hard time.
I also think a good level designer has to master composition. This is something that really comes into play in all stages of the level creation process. If you don’t understand composition you won’t be able to guide the player through a space and you won’t be able to make a level look appealing. Guiding a player is arguably the most important thing that a level designer has to do and if you don’t master composition you’re going to struggle with that. Without an understanding of composition, you might also end up creating very technically messy levels that won’t run well as you’ll often struggle and throw the kitchen sink in to accomplish the vision. If you master composition you’ll have a much easier time finding the best way forward when you’re composing your layout and art and thus won’t clutter it with unnecessary geometry or lighting or what have you.
The third fundamental I’d say is an artistic eye. Take it from me, you don’t have to be able to draw or make 3D models but you have to have an eye for it. You have to intuitively be able to look at a space and have an idea of if it looks good or not and how you can make it better. You have to know what looks natural, plausible and not random.This I’d say comes with experience but also it comes from playing games. I analyze almost every game I play and try to understand why they put that rock just there and why that cave is laid out in this or that specific way, and how would I have done it differently and what would that mean for the player? You know what appeals to you and chances are it will appeal to others. Having an artistic eye and understanding of depth and space will make your levels look better and play better.
Of course, there are many other things that you can say that a level designer should be good at but those three things I feel are really fundamentals and are things that I would look for in a level designer.
Other than that, I’d say that almost all bets are off. I have friends in the industry who do nothing else but script and they’re still level designers. I know that others simply block out the levels while I like to take a more broad approach. It will depend heavily on the studio you’re at and what game you’re working on. Ultimately every studio will find the best way to make the game, and that’s what it’s all about. However, I do feel that the industry would be a healthier place both for level designers and companies if we could branch out more call things as they are. Because often people are looking for someone to fill a specific space in the pipeline and do a specific job. And with games becoming bigger and bigger and teams growing across the board I think it would be healthy to use other terms than just “level designer”. Level scripter and level artist are some terms that I feel could be more widely adopted and then perhaps a handful more terms could be necessary. Much like you would specify if a programmer is needed to work on systems, networking or gameplay. However, I’m no authority on the subject but I do feel it’s something that we should be talking about more.
However, in the end, a level designer will always be working on the world of a game. I think that’s the thing that is truly universal to all level designers. A person who creates worlds.
So, speaking of the things I said earlier, I’d say my role is some hybrid of a level designer and a level artist. I concentrate my work on everything from conceptualizing early on all the way to optimizing the level’s performance before release. I design the level on paper, block the level out, art it up with assets, light it and iterate upon it. I’d say that the only thing that I most often don’t handle is any scripting or programming that goes into a level. If it’s something very intricate that needs my attention and input I’ll, of course, give that but I’ve been in a position during my projects where I’ve had coders and scripters doing most of that work. Other times a coder would come to me with a system that I can use to get along without bothering him in the future if there’s something we need to do a lot of and kind of automate.
Instead I’ll early on be working mostly in a document where I very briefly (excessive documentation is your enemy) work out things like the time and place of the level, what the story is, what the primary objective might be for the player in that level and what kind of mood and feelings I want to evoke in the player. At this stage, I’ll also be looking at references from movies, other games, comics or traditional art or architecture to find things that I can use as a starting off point. All this is useful not just for yourself but for any environment artist who might be building assets. It’s also very useful stuff to have noted down so you can communicate it to your leads or directors.
From that point on I’ll usually grow as the production grows. I’ll try to prioritize an art test if I’m waiting for systems or mechanics to be designed and implemented. During an art test, I’m not as much worrying about the layout or design specifics as much as I’m worried about the color palette of the space and what assets we’re going to need and how we’re going to use them in the level. The only design related thing I’m really thinking about during the art test is player guidance and composition. What colors can we use that will draw the player’s eye? Do we have enough movement in the scene? What landmarks do we have? And so on.
Then it’s business as usual. I’ll make a point of staying connected to the team and collaborating with other departments where I need to. This often comes naturally in the level creation process where you find that you need to talk to the artists about art related stuff and to the programmers about technical things and events you might need. I’ll go through the usual passes such as laying it out on paper, building a blockout, testing the blockout, doing an art test of the blockout, testing some more and so on. I also do a lot of cutscenes and cinematic work as that has often been done in the engine on the projects I’ve worked on. Then it really goes back in a loop after testing and when new systems come in. Iteration, iteration and more iteration. A game is an ever evolving thing and as a level designer you have to be very flexible and get good at “killing your darlings”. Things will change and you’ll have to adapt to that which takes up quite a lot of time. Re-working things as the game changes is definitely a part of the job.
I think level design always helps a project evolve in a really concrete way. When you do your work right and you have the things you need you’re often able to do a lot of work in the background as a level designer. You’ll be off on your own for stretches of time as you come to the first iteration of something and then when you hit the “commit” button on whatever version control you’re using, everyone really notices the impact. Often times coders are prototyping early on in a dull gray test world with not much to look at. But when you bring them into the level that you built and you let their systems be integrated into their actual “live” context you’ll really see that your work is appreciated. As you continue to work the game becomes the actual game, not just a mixed bag of mechanics and ideas. As a good level designer you’re laying down the structure in which things happen and the game goes from a prototype to an experience where you can get your first sights of the end result. Level design just adds so much to the game and from my experience, it shows in a really palpable way. It helps the game take that next big step.
When does your work usually start when it comes to game production?
Well, it depends on the project. At Mojang, we work in rather small teams and thus every developer is really important even early on in a project. Trying to work out a vision and idea together is a time where everyone gets to wear the game designer hat, be it for a short while. So in that working environment the level designer, meaning me, wouldn’t be allocated to another team or anything like that whilst a core group works out the broad strokes of the concept.
Once an initial concept has been designed I’d say it’s healthy for a level designer to join in. At that point, we should know if we’re making a shooter on an alien space station or if we’re making an RTS in medieval England if you understand what I mean. Early on in game production, most things are very loose and high level and I’d say that the same is true for level design. My goals during these early stages of a project are basically trying to establish a broad structure for the game. How many levels are we going to need to build? Are we working in an established IP where we have to build everything from scratch or can we use locations from an existing franchise? What kind of different environments are we going to visit, is there enough variety here? What’s the overarching goal/story for the player and how can we translate that into a set of levels? Researching other games in the genre that you’re working with is also very essential. You have to play a ton of games and understand what has been done before and how other people have solved the same problems that you’re about to face. Some genres are more rigid than others and thus doing this might help you more or less, but I find that drawing inspiration and learning lessons from your peers is very helpful this early on. If you’re doing a sequel to a game I guess you’d do the same type of exercise but with the previous title. Seeing what people liked and didn’t like and so on.
As far as tasks go I think it’s hard to say for sure since it depends so much on the game. You have to be open to what the game needs and what you and the team need and sort of build from there. At this stage I’ll mostly be using a pen and paper, writing things down and drawing ideas up. Sometimes I’ll jump into the editor or a program like Sketchup if I need to visualize something in 3D but just writing and drawing with a pen and paper or in a google doc will most often do the trick.
I’d say the main objective of a level designer in “pre-production” if you will, is to build a good foundation and base from which you can later build the game into a sequence of levels that are good and make sense in the context of the game. Now is the time to figure out your goals and try to work out an initial strategy of how you’re going to reach them.
You’ll often hear the term “fail fast” being used in the industry. I think this early stage is a great and safe space to do just that. If you do fail or struggle to find the right story hook, mood board, references, ideas and so on you’re only wasting time and maybe a few arcs of paper, not actual assets and most importantly you’re only wasting your own time and not someone else’s. The further down the line you go in production the harder it will be to change things. So really think it through early on and be thorough.
What are the essential tools of a level designer?
The short answer: whatever get’s the job done. There’s not really a right or wrong way to do things in terms of software and tools. I can only speak from my own experience but generally, I find that I’ll spend most my time in the actual engine and level editor. Whether that’s Unreal Engine 4 which Skylar & Plux ran on or some in house tool that your studio has developed. I’ll be in there early on working with really primitive shapes such as BSPs while I do the blockout and later on, I’ll replace that blockout with actual assets that have been imported from whatever 3D software the environment artist is using. Most times you’ll be working with something coming from Maya or Zbrush. You really can’t go wrong here or choose the wrong method. What engine you will end up using will surely not be your sole decision and from that point on you’ll just have to adapt and work with whatever you are given. Some engines like UE4 are great and, judging from the various horror stories out there, some in house engines are not so great. You don’t have power over that decision so I’d urge you to get used to it and learn to excel in spite of the tools. You should really strive to be the expert on the editor on your team as you will most likely be the one spending the most time in there. If you learn the tools in and out, you’ll make your life a lot easier. There are great documentation and tutorials out there so you’re definitely not alone in trying to learn this.
When it comes to tools outside the editor, it’s the wild west. I often write a lot in google docs before I even open the editor just to try to straighten out whatever idea or vision I have for myself before I get cracking. I want to have a plan and make sure that it’s thought through well before I attempt to implement it. If you go in blind you’ll most likely end up with something messy. Sometimes I’ll prototype something with BSPs or primitive shapes in the editor while other times I’ll use Sketchup instead and then I might bring that into a google doc along with some reference to organize a plan. Sometimes I’ll sit in a quiet corner of the office with a pen and paper and try to draw some ideas up while another time I might have a chat with the art director or game designer if there’s something specific that they require of the level.
For reference, I’ll try to play other games and take screenshots, search around the internet or simply take a walk and photograph place with my phone. Really anything goes here and there’s not a right or wrong way to go about doing something in terms of workflow. Of course, you want to stick to some key stages such as finishing a paper design before you move on to the blockout and you probably want to finish that blockout and nail that design before you even begin arting it up. However, those stages might blend together more than you’d think and I don’t want to say that that is necessarily a bad thing. Do what feels best for you and what get’s you the best end result.
I actually work a lot with references. I’m a huge geek and a consumer of pop culture and you’ll find that a lot of game developers share those qualities. So we all sort of speak the same geek language and have the same geek radar that we can communicate through and describe quite abstract things to each other with. So for that, I find that using reference from movies, comics, music, and games can be quite useful not just for yourself but also when you’re trying to communicate something to someone else on the team. If you’re making a shooter and say that you want the player to feel like Robocop I think most game devs would understand what you mean. Just by using that character, Robocop, you communicate so many nuances and feelings to someone else in a really effective and clear way. So I’d say that reference is great for you when you’re looking at how you should go about building something, but it’s also great when you’re trying to communicate your idea or vision to someone else.
When I’m off on my own and I’m actually trying to construct an encounter or trying to pace something out I’ll often use a lot of reference from movies or comics. To see how a director and editor have chosen to construct a scene to reach its maximum potential I find to be quite inspiring. The same things go with comics. Those panels go together in a specific way because someone intended it to be so, there’s nothing random about it. I think the most inspiring thing to me in terms of comics is Frank Miller’s The Dark Knight Returns. The panels in that book are so masterfully laid out and constructed that it just makes you want to do a better job pacing your levels when you look at it. Film wise I’d recommend every level designer out there to look at the first John Wick movie. The way they build that character up is something that is super applicable to level design and if you look for it you’ll see that the latest Doom game has got some cues from that film in there. All this stuff might not be things that would be super obvious but they really help me on an everyday basis to think outside the box and strive to do better work.
Of course, I use more traditional reference as well. If I’m building a tropical shoreline I’ll go look for what the real thing actually looks like, but that goes without saying I think.
How does one get better at level design?
There are so many areas of expertise that a level designer can deepen their knowledge in to become a stronger level designer. It can for sure seem daunting at first but you have to realize that this is a process that takes a lot of learning by doing and failing along the way. As I have said before: being a level designer at one company won’t be the same thing as being a level designer at another company. Different studios will favor different skills and you can’t possibly cater to everyone. So, do your research and find out what the companies that you’re interested in value.
Generally, a level designer will need a solid understanding of the visual aspect of environments and how they are composed. This alone is something that you can spend a silly amount of time becoming good at. Even if you’re only going to be blocking out levels with simple shapes you’re still laying the foundation and groundwork for everyone else coming after you. If you make a really visually unappealing blockout that doesn’t make sense and isn’t plausible and can’t be translated into game art, then it won’t work and it won’t sell. If you’re handing your work off to an environment artist later down the line he or she won’t be able to just flick their magic wand and make it look pretty or make it plausible. They’re going to work of what you give them. Even the most visually simple games need to look visually appealing and be well composed. This is also something you have to balance against the gameplay and the pacing. All the bells and whistles that make a level into a level have to still work and be visually appealing. You’re not just making a static scene so the difficulty of making something look good whilst still working in the context of the game is a huge task. If anything, a junior level designer should strive to master this before anything.
Of course, if you’re going to be a level scripter you’re gonna want to focus more on the technical aspect of things and how you make stuff work and happen in an efficient way. However, I do believe that being visually trained is something that will be helpful for all types of level designers.
What are the most essential things any level designer should understand?
I think what I’ve mentioned so far is pretty fundamental to all aspects of level design. These will always vary from project to project but there will always be something that you can apply those skills to. You have to be able to look at the game as a whole and use the tools that you are given to make the game happen and be fun. It’s a huge task that isn’t easily cut up into smaller parts since they so often blend together into one thing. How can you make mechanic X and mechanic Y be fun for twenty hours by combining them together in different ways? How can you in the best way possible construct a certain event and pace it so that it hits home with the maximum effect? How can you technically make what you want to happen actually happen in a way that works and won’t be buggy or performance heavy? You have to be able to bring up your technical skills, your artistic skills, and your design skills to a certain level where you can pull off the task at hand, often times by yourself. There are no easy answers and there’s no point where you can say that you know it all. I think if you have the passion and the will to learn and you put your mind to it, you’re going to be fine. However, it is indeed a tough job.
What if…Designing Levels for a Sandbox Game
I believe it’s always a level designer’s job to enhance the core mechanics of a game. If you’re, for instance, building a shooter you might be able to get quite a few hours of enjoyment out of a square room with nothing in it if the core mechanics are solid. However, I believe that if you apply level design to that and find a way to enhance it the experience will be better for it. So in this case that you’re suggesting I would go in and look at the core mechanics. What is the player doing over and over again? Are they shooting? Are they building? Are they hunting? Whatever it might be you have to find a way to enhance that experience and allow for moments of gameplay that vary. I think you have to look at it as a theme park kind of thing. You can’t just have one roller coaster over and over again. You need to offer up a platter of different activities and attractions that vary from one another and then the player will choose what they want to do. This again has a lot to do with pacing. I think The Hunter games do a good job at this if you look at it from a level design and environment design perspective. You have open fields, dense forests, country roads, watering holes and so on. You’re still putting sights over targets and pulling the trigger but the environment and the world around you will give you certain advantages and limitations to alter how you use those core mechanics. And the player will seek out those differences depending on what they’re looking to do. The recent Zelda does a great job at this as well. It’s sort of a choose your own adventure approach to level design.
I think you should be able to do almost anything style and environment wise if you’re a good level designer. However, what always gives me a headache is realistic or semi-realistic architecture. The main reason for this is that there’s no wiggle room. When you’re doing more stylized fantasy or sci-fi environments you’re dreaming things up and sort of make your own rules. Of course, a stylized fantasy tower still has to look like a functional tower but there are so many areas where you can put your own touch on it and make it a bit personal. When you’re doing realistic architecture, in my experience, there is no such thing. A bridge is going to look like a bridge and it needs to have the proper foundations in place and every little steel bar has to connect at the exact right angle and way and so on. It just makes me nervous since you have to plan it so, so well and the smallest mistake will show right away and literally screw the whole thing over. The closest I’ve gotten to this is when I helped out on The Solus Project which was a semi-realistic looking game. I built an architectural piece in that game and it was the most irritating and annoying thing I’ve ever done. There were so many small parts that had to connect together and it was a very modular kit of assets. Which meant that there were no shortcuts and that everything had to be thought out very well and placed with purpose. It all has to fill a very specific function, most of the time. I was sweating through the whole thing and I spent a lot of time on it even though it was quite a small area.
I wouldn’t say this is applicable to all level designers. This is just the main thing that I dread when I close my eyes at night.
However, you need to learn to cope with those problems and the way I deal with it is to plan it all out well ahead of time. I make sure I have enough time to do what I need to do and that I’m not going in blind. Doing things step by step in a very meticulous way helps me deal with such more complicated tasks that don’t come very naturally to me. As I said: you should be able to accomplish almost anything and the way you do that is by being well prepared.