I’ve been doing a lot of interviews with industry professionals about their roles, how they got started and discussing details about their workflow and the more technical side of working in the games industry. Throughout these I noticed that job descriptions vary from studio to studio and that there are a lot of misunderstandings about what you end up doing when you apply for a certain position, even from people within the games industry. So this new, (hopefully) series is made for readers that are aspiring to become games industry professionals. We’re not going to provide advice on how to enter the games industry because that has been done to death already. Instead, we are going to talk what’s it like once you are in. Our very own Will “Vilham” Josephy, is prepared to share his experiences working as a Level Designer for Crytek. Together we will try to paint a picture of what it’s actually like to work as a Level Designer for a large company in the games industry.
Hello! Thank you for agreeing to do this kind of feature with me. I’m sure the community will appreciate it. Before we start though, could you tell us a bit about yourself?
Hey, glad to help. I’m a Level Designer with 6 year’s experience in the industry. I have worked in 3 different studios getting my hands into about 7 different projects of which 2 have shipped so far and some which were cancelled and will never see the light of day. I started out making levels for Starcraft back in about 2000. I gradually started making mods and maps for various games I enjoyed like Morrowind, Age of Empires 2 and Warcraft 3 before I really got into making maps for Counterstrike:Source and Day of Defeat:Source. That’s when I really started to learn. There were a lot of early attempts that were failures but good learning experiences. I eventually went on to study Computer Science and spent my projects focusing on game related things and during my spare time I continued to map for Counterstrike:Source and Day of Defeat:Source. I secured a job out of university as a Junior Level Designer and have progressed from there to Senior Level Designer at Foundry 42.
Let’s assume I’m a very talented Level Designer who has made some well received custom maps. I apply at Crytek Frankfurt and I get hired on the spot. I’m feeling confident about making it when I show up for my first day. What can I expect?
The first few weeks will probably not be what you expected. They will generally be spent being introduced to people, the project (you will probably play the game until you are dying to dive in and start working on something) and the processes. This whole period can feel quite slow, however it is very important that you get a proper grasp on how things are done and what the game is trying to achieve. My personal experience at both Crytek and Frontier were pretty similar in that regard. Frontier, being my first job, meant I really wanted to get started, but in the end I was reading lots of documentation, playing parts of the game that were playable and sitting in on meetings where I felt incredibly out of depth. However, people quickly made me feel welcome and treated me as one of the team.
Ok, so I have survived the initiation. I’ve mopped the floors, cleaned the toilets and brought everybody coffee for weeks on end. What is expected of me as a Level Designer? Am I only creating white boxes and setting up gameplay sequences that are predetermined or do I have more freedom than that?
This very much depends from company to company and the particular design role you have applied for. At Crytek there is just one type of Level Designer, who deals with everything level related. For Ryse that meant I was given a high level story goal and general world location, from that I created the whitebox and worked with my Environment and Concept Artists to define the desired end goal visually. That is interspersed with reviews with leads and other designers, gradually building up what the goals and flow of the level are.
At the beginning your level can change significantly at times. For Ryse that meant changing the entire location of my level during early whiteboxing stages as we were running into issues with the initial location. For example, in these development shots you can see there used to be a Roman camp on the path of the aqueduct. This served as a choke point for the men you are supporting on the road below and you were basically going to open up that choke point for them. But we wanted to bring the Boudicca to the forefront in the level. So the Roman camp was removed and an introduction to the Boudicca involving a mini boss fight was added at the top of the aqueduct. Depending on whether you are working on a new IP or sequel there can be other issues that affect your whitebox development. In the case of Ryse lots of systems weren’t online during the early whitebox stages. This meant you had to design ahead and calculate the spaces you would need and rely on other departments delivering that feature to you.
Some time has passed and everybody has gotten used to my antics and I’ve proved myself capable enough to secure my position. What would my typical work day look like? What would I be doing all day?
Literally anything to do with the level. Being a Level Designer from my experience is by far the most varied job in the industry. Your job as a Level Designer is to put together everything everyone produces for your level. Even if you aren’t putting it in yourself, like art assets, you still need to be tracking that process and communicating with your team to ensure everything remains working. You will communicate with code, both gameplay and technical, audio, lighting, art, system design, cinematics, etc. You are literally in charge of getting your level done, so it’s important to stay on top of everything. You also need to keep your level maintained and working the whole time, even when things are changing, that is very important for the testing of the game day to day. So an example day assuming you don’t have a big backlog of tasks might be:
Get the latest build of the game and ensure you have all your work files. Read your emails while you wait for the build to download. Load the level up in the editor and start working on any minor tasks you might have, let’s say improving the combat setup at the beginning of the level. While playtesting that section you have an idea of how you could shift parts of the area around to improve leading. Take that idea to your Environment Artist, get their perspective on it. He agrees it’s a good idea, so you go talk with the Lighting Artist to see if he can adjust the lighting in the area to heighten the leading, killing two birds with one stone. You make the adjustments to the level and it might be about lunch.
You decide to get an update from audio about the levels ambiances and sounds, so you message your Audio Engineer and pop over for a quick talk about what they want to do and where they are at with their side of it. That little meeting reminds you that some of the dialogue isn’t sounding great at the moment due to some changes that have happened, so you write an email to the writers suggesting some changes that would better suit the current setup. Quality Assurance have just finished testing your level today and an influx of bugs pops up, you start working on fixing the most critical ones to ensure the level is playable beginning to end. During the process you tweak and fix a few more minor bugs that you notice. You do a final test in game and commit your fixes and changes for the day.That could be an example day, once work has started on the level rarely do I have two days alike. Only towards alpha where it becomes a case of fixing bugs, without major changes or additions, does the day become a bit more regular.
For this new format I’ve also reached out to the community to find out what they would like to know. If you want to participate in future events be sure to list your desired role in the games industry here. Below, Will “Vilham” Josephy handpicked some of the questions I collected and answered them.
How long should projects last? Is there a set timeframe in which each level needs to be developed? If so, how do you know if you reached the highest possible quality? I take it that’s decided based on testing. But who tests the level and how often is that? Who ultimately decide it’s done?
Kinect Disneyland Adventures took about 9 months. Ryse took 2.5 years. Projects like Fallout 4 apparently take 5 years. World of Warcraft was 4-5 years. Basically it can vary hugely based on the scope of the project and the resources behind it. Timeframes can vary hugely, some games I have worked on multiple levels simultaneously, on Ryse it was just one. Then there is obviously the question of how large a level is. You could make multiple small levels in the same time as a large epic level. Development on a level normally doesn’t cease until a period of time before release called content complete. At that point you will focus on bug fixing and minor improvements to “polish” the level.
I don’t think any developer ever looks at one of their levels and thinks that it was the best it could be. It's generally part of the creative process that you see flaws within your own design over time no matter how minor and insignificant they might seem to the player. It’s actually a good reason to take a step back and stop at some point. After a level reaches a certain stage (where we think it is worth testing) they are tested daily by multiple QA testers. Towards the end of the project the publisher will normally stick outsource testers on it too. At the end of Ryse I saw bugs being submitted by dozens of different testers.
What would you consider the normal process for the overall level design of any type of game?
Ok, this is a fairly long one that varies studio to studio. I have heard plenty of terms for the different stages and there doesn’t seem to be a solid consensus on it. Not every project has been completely like this but the general gist has been the same. You will start out with some high level criteria that the creative leads and narrative leads have decided for the level e.g. a story beat; capture Oswald, your place in the story and therefore the pace of the level; after a huge epic level filled with a beach landing, roughly what mechanics you will have been exposed to; first half of the game have met some basic enemy types and will reveal a new one.
From that you build a pitch which will inform the direction of the level; mission statement (what is the core aspect about the level that distinguishes it from others), setting (location, atmosphere, time of day), general flow of events, reference images and a simple top down layout. Once that is oked you will begin a whitebox, this allows you to begin defining spaces and how they connect, where the events take place within and what mechanics are used where. This stage will involve lots of iteration and you will slowly build up areas to give a better impression of them. You will begin to add small bits of AI encounters.
You will have another review after an agreed period of time, if that goes well you will move into production. You will begin to get whitebox assets that an artist will be creating for the spaces you have defined, normally starting with the large structures then moving onto smaller details. Meanwhile you will be working on integrating encounters and scripted events. Everything else will be improving in parallel. You will probably end up making changes to areas as more things come online and you notice things that are not working anymore. There will normally be regular reviews and playtests of the level to ensure its staying on track and nothing needs changing. Eventually you will hit content complete. At this point everything should really be in and you will focus on bug fixing and polishing things.
Depending on the type of game, what are some parameters you would consider? For example: ceiling or jump height for first person shooters, line of sight, pacing, etc.
Figuring out metrics is an important part of early development. It’s partially a case of looking at what you know about games and trial and error. For example FOV and movement in games make what are realistic sized doors feel tiny. Seriously, look at most games, the doors are 3-4m tall, in reality most doors are ~2m. Unfortunately on every project I have worked on it's had teething problems. For example, on Ryse it took a while to get the heights of the vaults, drops and climbs feeling right (i.e. not too insignificant but not too high to make). It’s generally down to the requirements of mechanics and the scale of the world (God of War has huge environments and the player is basically superhuman so the metrics are very large. Ryse has huge environments at times but the player is meant to be grounded in reality so the metrics are fairly reasonable).
Must you know scripting and/or programming for level design? If so, does this apply to both single and multi-player?
As a level designer you should learn how to script in various engines; written script and visual scripting. You may land a position that doesn’t require it such as a world designer or a multiplayer designer on a project that has no scripting within the levels, but that isn’t a given and you sell yourself a lot better if you have a full complement of technical design skills. You should expand your knowledge to cover other areas like game design (which inherently I don’t think you can be a level designer without, levels are the environments within which you use mechanics) and environment art, learning a bit about composition, complimenting colours, etc is very helpful. Architecture is another thing to read up on.
How do you address draw call concerns during the design process? For example, tight corridors mixed with open areas?
You need to find out the limitations of the engine you are working in. If it's an engine you have used before then great, you probably already have a good idea. If not, it requires a bit of trial and error and information gathering. Once you have a good idea you need to design within those limitations. If you have designed something that doesn’t run at a decent frame rate then something will need to be cut. Cutting that element may compromise the design itself and make that portion of the level weak.
Who or what inspires your level design?
Primarily games and travel. I play lots of games from most genres this gives me tons of inspiration from settings to mechanics. A game in a totally unrelated genre can still inspire work on another. Travel is a great way to get inspired, Europe itself is great for this, lots of countries with varying architectural styles over hundreds of years and, beautiful and varied terrain. In terms of known names I don’t really think there are any. Level design is often the overlooked aspect of games, there are numerous great “Designers” or “Creative leads”, but generally you don’t know the names of the people that have created great levels. In terms of actual levels, I hear lots of praise for the Uncharted 2 Train level. Then there are often conversations about aspects of different levels, lots of conversations in the office recently have been about The Witcher 3’s environmental storytelling and Wolfenstein's level pacing and ballsy locations.
How does the environment impact the level design? Does it start after the initial white box version of the map? Or is it taken into account from the beginning?
In my experience it comes in early, the setting is a key aspect of a level. In Ryse for example my level started as a siege camp outside York, this obviously had huge implications to the sort of encounters I could create and what sort of events could happen. After struggling to get the level feeling right I pitched to move the level out to the forests further from York. The level evolved from there. During that time the high level goals of the level never changed. You were a reinforcing army moving up from Dover to lift the siege of York and during the level you must capture Oswald.
Do you sometimes feel that your level design ideas are held back by the game design and/or marketing for the game?
You can be held back by the game’s design. You have to roll with the punches and argue your case. If there is nothing to be done, then you need to do your best with what you have. You will never get everything you want for a game, there are too many dependencies and opinions for that to happen. Plus, who is to say the idea is actually good in practice? Maybe you are working with people who have more experience and they explain why your idea might not work.
That concludes this feature. A big thank you goes out to Will “Vilham” Josephy for making it possible. If you appreciate his effort and if you want to see this feature return in the future, be sure to hit the like button. As always, discussion is encouraged below! Last but not least, come find me if you are a games industry professional interested in helping me out with this (in a different games industry role of course).