July Update: Dev Diary by James Henley
Hey folks!
This is James Henley, Lead Level Designer on the System Shock reboot. We thought that we’d give a little bit of insight into the level design process this month! Being a creative process, there’s a lot of variation from person to person but what follows is my personal approach. I’m going to speak in generals so try not to read too much into game content specifics from this.
First off, however, let’s address the elephant in the room...
System Shock’s levels are already designed, aren’t they?
This update is about process rather than intent, so I’ll keep this brief! The purpose of a reboot is to leverage an existing foundation while still allowing the freedom to re-envision, clarify, or otherwise expand upon a work. You could view it as building a new body to house an old soul.
The answer to the question is both “yes” and “no.” The original levels were designed under specific philosophies and restrictions that have grown or otherwise evolved in the years since. To that end, much of my work involves translating old intentions and bringing them forward to work in tandem with the technology and principles we have today.
My goal is to create a believable world space that retains System Shock’s original sense of exploration and freedom.
I won’t get into the specifics of my design philosophies at the moment but let’s just be clear that I’m not talking about the modern trend of corridor-shooter level design; I’m an advocate of choices, exploration, and the freedom to tackle challenges in a variety of ways. My job is to develop a believable world space for the player to explore, not a cinematic corridor to be raced through.
Initial Planning
Before we can place so much as a single piece of flooring, we must first consider the goals and themes of an area, addressing questions such as:
- What section of the station does this level occur on?
- What are the major narrative beats it must cover?(General objectives, Key events & interactions)
- Is it a new tileset / major set of assets?
This information is compiled in a short document to provide material for the larger discussion meeting that comes next.
A level designer’s work can often be viewed as the end product of an assembly line, loosely speaking. Everything in the game feeds into it and much of what we rely on are the fruits of another department’s labors; creatures, tiles and art assets, hackable objects, etc. Accordingly, stakeholders meet to hash out some of the more detailed points of the level, such as:
- What kind of artistic themes and points of interest can we leverage?
- What is the player’s toolbox? (Abilities or powers, Weapons and other possible gear)
- What is the designer’s toolbox? (New and returning creatures, Mechanics to introduce, Gating mechanisms, Features of the area itself, thematically or geometrically)
Paper, Please
Even in the digital age, I’m still a fan of working with paper first and foremost! I have a small notebook that I cram full of bullet point notes to clarify thoughts, sketch out room designs, etc. There’s just something about not being bound to my computer that lets me feel more creative at this stage.
Sorry, gotta scrub all the sensitive information!
I like to start by establishing the overall gameplay beats -- the pacing and flow of both narrative and game elements within a level’s critical* path. This means determining what the major objectives are at a more granular level. Where do you go? How do you get there? What affects your ability to get there and what options do you have for solving that?
* The specific events, be they branching or linear, that must be encountered in order to progress from beginning to end of a level
I write the major events, objective updates, etc. on cue cards. Yes, cue cards. I spread them around on the floor and start grouping, ordering, and reordering them to get a feel for strengths and weaknesses in the flow of events. I enjoy this approach because iteration is simple, rapid, and I can sort through and make adjustments anywhere from my office to a coffee shop -- though the latter is less appreciative about the whole “spreading them out on the floor” thing.
Spatial Relations
While the cue cards have helped hash out a lot of information at the high level, that doesn’t necessarily tell a lot about the physical layout. I have one more step before I dive into the in-engine side of the process: conceptual mapping. It’s a fancy term for a pretty simple thing; a flow-chart mapping out the spatial relation of level locations and events, as shown below. It gives an opportunity to identify potential for shortcuts, space requirements, and helps clarify a few other bits and pieces.
Even in a sci-fi setting, there are still a number of equivalent locations that we already have ingrained context for in the world around us, such as factories, hospitals, department stores, etc. During this stage, I do a lot of research into floor plans and layouts to determine what existing contexts can be leveraged in order to help make the space more believable. Would this kind of room really be near that kind? Are these features logical or are we placing them this way strictly “because plot?” There is a point where you have to draw the line in favor of the “it’s a game” argument but there’s little reason to abandon environmental believability prior to that. Ultimately, it’s a balancing act of realism vs. intended player experience.
The Gray Box
To be honest, all the doc preparation in the world doesn’t change the fact that level design is ultimately an organic process once you get into the engine itself; it’s strictly a set of guidelines to help propel the work along. Planning for the space isn’t quite the same as working within it. New ideas form, plans change, and feedback is taken into account. That’s just the way things go.
So why all the initial work on paper? I think Eisenhower said it best:
“In preparing for battle I have always found that plans are useless, but planning is indispensable.”
Personally, I build much faster and more effectively with a handful of guidelines established.
The first stage of actual level creation is known as the gray box. Temporary assets are used to block out the various rooms and corridors, stub out points of interest, and ultimately create the framework for a playable area. It’s not pretty, but it’s the bone structure of the game world’s physical space. A designer’s job is to make it functional and effective; an artist will make it look damn good later!
A quick gray box demonstration room.
Gray boxing is a critical phase in the life of a level. Iterations can be performed fairly quickly at this point without as much risk of the changes impacting other departments. We play through the level frequently to feel out travel times, sight lines, room designs, verticality, and all manner of minutiae. Side paths, alternate routes, and the like are implemented and tested.
Gray box sections can even be provided to art as reference for more accurate concepting of props and space purposes.
Feedback is taken and adjustments are made accordingly. As a general rule, I assume that nothing is right the first time; iteration is the key to a polished design.
Stubbing in the Gameplay
Once the gray box geometry begins to stabilize, encounters and event markers can be stubbed in. Even if the correct creatures aren’t available, something can always be placed as an approximation. This allows us to:
- Plan encounter locations
- Plot patrol paths
- Arrange stealth routes or alternative solutions to combat
- Test the overall pacing in a variety of playstyles
From a non-combat perspective, we can also start plotting the specifics of:
- Key item placement (Weapons, Progression items or stand-ins, Information provisions critical to narrative)
- Event areas for scripting
- Environmental storytelling
Polish and Iteration and Iteration and Iteration and...
As you’d expect, all of this is run through another round of feedback and iterations! This cycle of adjustments continues until design signs off and hands the level over to the artists so that they can work their own magic. While they do that, we can continue to iterate on and polish the non-geometry gameplay elements -- though the occasional change to geo may still be necessary. Nature of the beast, and all.
Throughout the entire level design process, communication lines are kept open with the Art department and ideas are shared about the look, shapes, or visual goals. It’s not enough to just pass them a box and say “make this pretty.” They have their own goals to meet and intentions to express, so a strong dialogue between level designers and artists is critical to making sure that everyone is getting what they need out of the world that’s being built.
I hope you’ve enjoyed this peek under the hood at how a level initially takes shape. I’m sure Art’s side of the process would have cooler things to show but, hey, a level’s gotta start
somewhere!
CS:GO "Oceanic Ordinance"
Calling all CS:GO players! Come vote for the "Oceanic Ordinance", a custom P90 with the Nightdive Studios logo paint job.
Thanks Tom (DonHonk), for creating such an awesome design!
http://steamcommunity.com/sharedfiles/filedetails/?id=1091098759
If you have any other questions or support needs, please let us know here:
Support.NightdiveStudios.com
See you next time~!
(。・ω・。)ノ♡Karlee Meow