Presentation Takeaways – Gearbox Software

This past Friday, three game developers from Gearbox Software visited the SCAD Savannah campus to discuss their work on the Borderlands FPS franchise.

Katelyn Pitstick presented a talk titled “More” Missions for Borderlands 3 in which she discussed her career as a level and mission designer. Kevin Powell discussed his career as a systems designer in Sticking to Your Guns: Borderlands 3 Systems Design, then Kate and Kevin took audience questions. Finally, art lead Jimmy Barnett discussed the evolution of the Borderlands gun generation system in The Weapons of Borderlands 3: When a Bazillion Guns Isn’t Enough… and answered some audience questions.

For people who couldn’t make it to the event, we wrote up a summary of each presentation.

Katelyn Pitstick – “More” Missions for Borderlands 3

• Kate graduated from SCAD in 2010, and designed levels for game genres from MMO to FPS to MOBAs before joining Gearbox in 2016. She started as a level designer for Battleborn, and then moved to the Mission Design team on Borderlands 3.
• Kate clarified the difference between level design and mission design:
  • A level designer creates a (typically) three-dimensional virtual space, and plots the progression and movement the player will take through that environment.
  • A mission designer creates a small, often self-contained story, then makes that story interactive with Combat, Obstacles, Rewards, and Objectives. They also script that story with tools like Blueprint and Kismet to make all of those elements work together.
• The day to day experience of a mission designer includes: writing design documents, scripting interaction, iterating your scripts, and communicating with members of other teams. The main purpose of a mission designer is to come up with interesting, different ways for a player to use an existing set of game mechanics.
• To keep their design documents organized, the Borderlands Mission Design team uses a document collaboration tool called Confluence (https://www.atlassian.com/software/confluence). Sharing documents live with collaborators allows the team to quickly give feedback and make changes.
• In Borderlands as well as many similar games (especially RPGs), there are Plot Missions and Side Missions. Kate discussed the difference in methodologies for making these types of missions.
  • For Plot Missions which concern the main story of the game, Mission Design works closely with the Narrative team. Their main concern is to take the Narrative team’s story and make it interactive for the player.
  • For Side Missions, Mission Design has much more narrative freedom. They start from scratch, planning out the objectives and beats of each mission. The whole time, the team keeps in mind why the player would want to invest time in a non-essential mission: for Borderlands, the answer to this question is in humor, character development, and loot.
• Like any other game developer, mission designers constantly hand off their work to other developers, and don’t always be able to explain their work in person. From Kate: “Design as if someone else is going to be handed your work.”
• Kate also expressed the importance of imposing restrictions on design. Although it may hamper creative freedom, putting specific restrictions on aspects such as console performance, narrative continuity, and development timeline helps a design team avoid “design loops” and stay on task to ship a game on time.

Kevin Powell – Sticking to Your Guns: Borderlands 3 Systems Design

• Kevin graduated from SCAD in 2010, and performed a variety of roles including level art and character design before becoming a Systems Designer at Gearbox in 2016.
• Kevin presented some of his pre-Gearbox work, including his Character Design experience on SMITE. He discussed the process of designing the Rama character’s Rolling Assault ability: “Find the fun first, then balance later. If you try to design straight for balance, it might not always be fun.”
• Systems design encompasses many different types of content. One system Kevin designed was the new “player rooms” feature in Borderlands 3, which combines the systems of a home base, inventory management, trophy and loot displays, and visual customization. Since systems design is such a broad field, Kevin explained it as “building upon existing technology to drive a new feature.”
• Some other systems Kevin has worked on in Borderlands 3 include a redesigned SDU vendor system, puddles, and barrels. These carry the same ideas of gameplay from previous games, just presented in a different way; the changes made by the Systems Design team to user interaction and player feedback make them feel completely different.
• The rough steps of system design are Documentation, Prototyping, Iteration, Ratification, and Management.

Katelyn Pitstick and Kevin Powell – Q&A (answers are paraphrased)

• Q (Both): What are some of the differences in pipeline and crunch time between HiRez and Gearbox?
  • Kate: Live games constantly need monitoring and balance tracking. There is a high demand for new content all the time with a live game like a MOBA. With Borderlands however, a fully-shipped game creates different expectations for content-at-release; once the game ships, it’s done unless there’s something horrifically broken. Production pipelines have more to do with what kind of game you’re making than the studio, especially whether or not it is a live title like SMITE or Battleborn.
• Q (Both): Do you ever interact with people from the Cogs program?
  • Kevin: A lot! We give them tasks and make sure they’re being set off on the right foot. It’s a collaborative environment.
  • Kate: For those of you who don’t know, the intern program at Gearbox is called Cogs. It’s also a mentorship program, so professional designers will work with interns to familiarize them with software and processes which are common across the industry. We typically start new interns by handing off bugs to them. By the end however, they have the chance to develop content on their own.
• Q (Both): As a level designer, what do you spend most of your time doing?
  • Kevin: Have a couple of assigned system, go through queue and do tasks; doors and interactive objects; each object goes through standardized list of requirements, each of which involves a different person/team such as sound, modeling.
  • Kate: Iteration and bug fixing; goes from paper design to test map within two weeks. Spends lots of time figuring out if something is playable and fun. Not designing in a bubble; tools are constantly changing. Be aware of what you’re utilizing in a mission, and continue to iterate.
• Q (Both): What defines a system designer, and what should be in a portfolio?
  • Kevin: System design is taking a design and making it scalable and repeatable, and making it work with other systems. Making something that can react in a predictable way that doesn’t break when players do things unexpected. A system designer facilitates that. From a portfolio, I would expect to see different systems, like an inventory system, interactive lighting, health management, so on. It’s less about the complexity of an individual system, and more about how they add up and work together.
  • Kate: Show your blueprint for that too; I would want to see clarity (such as with apparent parent/child hierarchy organization) and documentation.
• Q (Both): How do you avoid feature creep and stay out of design circles?
  • Kevin: Don’t add stuff unless there’s a good reason. At a certain point in the project, say “No new stuff!” and stick to it.
  • Kate: Make a list of all the ideas and requests, and sit down to prioritize them all. Figure out “I can do this amount of things in this amount of time.” Make the realities of production clear to the other members of your team.
• Q (Both): What are some of your favorite missions from other games?
  • Kate: I play games for specific purposes, especially the recent God of War and Tomb Raider reboots. Their missions, puzzles, and environments blend seamlessly into each other. I like how they incorporate gameplay into their environments, in that “game elements” don’t always demand an explanation – players already know they will have to do game stuff. The elements just need to work internally.
  • Kevin: It’s important to stay current as a designer, since there’s so many mechanics in games that you can steal. Playing games gives you ideas for things you can do, and the new and interesting ways you can incorporate content into your game.
• Q (Both): In light of the current controversy over working conditions in the game industry, what are some of the positives that make you want to work in games?
  • Kate: It’s fun, I like what I do and the people I work with. I like to see people watch trailers for the first time; seeing people excited is satisfying. I also enjoy the “puzzle solving” aspect of doing scripting, figuring out how to assemble a scene just the way I want it. There’s a great community in development, where you all celebrate each other’s accomplishments.
  • Kevin: You just can’t dwell in the negative stuff on the internet. Learn how to disconnect and do what you know how to do. Stay positive; we get to make games!
• Q (Both): What was the biggest change between SCAD and industry, and how do you cope?
  • Kate: I started working right out of school, so there was zero break between school and work. I don’t get nice long holidays, or established days in the week where I know, every week, I will just rest on that day. Gearbox is good about health leave, but the weirdest thing is suddenly not being able to count on summer/winter break.
  • Kevin: Take care of yourself. It’s hard to get out of the cramming mindset you get into at school, with midterms and finals. If you’re hired, it’s not like you have something to prove; don’t burn yourself out.
• Q (Kevin): When you came to SCAD, did you expect to be a systems designer this far out?
  • Kevin: I focused on doing art, but ended up being able to jump to different positions since I developed multiple skillsets.
• Q (Kevin): Do I need to learn C++ to be a systems designer?
  • Kevin: I work completely in Blueprint, but this may be different for other studios. Ultimately, learning C++ won’t hurt, and sometimes C++ coders are needed to make stuff like custom nodes.

Jimmy Barnett – The Weapons of Borderlands 3: When a Bazillion Guns Isn’t Enough…

• Jimmy graduated from SCAD in 2009 and spent ten years at Gearbox, where he is currently the Lead 3D Weapons Artist.
• Jimmy managed the evolution of the Borderlands series’ weapon generation system between Borderlands 1 and Borderlands 2. They used the same basic system and worked “upwards” – improving visual fidelity, establishing manufacturer identities, and beating the first game’s figure of 17.75 million guns.
• On Borderlands 2, Jimmy’s team accomplished most of their goals over their two year development timeline, but had to find creative, unexpected ways to make the end product work.
• Jimmy expressed the importance of asking the question “Is it performant?” – basically artists need to develop for the lowest common denominator, i.e. the console with the worst specifications. This encouraged the art team to find creative techniques to overcome the performance and memory limitations of the Xbox 260 and Playstation 3.
• For example, because texture memory is taxing on these systems, Jimmy’s team used a technique known as “gray packing” to minimize the amount of data needed to store all the visual information. Also, every weapon in the game draws upon a single “gestalt” file which contains all the possible parts for that weapon type. These techniques allowed Jimmy’s team to put 53 million different guns into Borderlands 2.
• For Borderlands 3 however, Jimmy noted that bigger would not necessarily be better. Although his team did work to increase the number of guns higher, there is a point of diminishing returns. Numbers in the millions are hard to comprehend, and at a certain point it becomes impossible for a player to experience all that content, so there is no significant difference in the player experience. For that reason, the gun design process for Borderlands 3 built “out” instead of “up”.
• Jimmy’s team started their design at the level of player experience. They wanted variation between weapons to feel meaningful. To do this, they designed “alternative fire” modes for the weapons in addition to making more of the typical components.
• They also “leaned into” the existing aesthetics which differentiate the Borderlands series’ gun manufacturers. For example, the new alt fire mechanics enhance the familiar qualities of Borderlands’ guns, such as Dahl’s burst fire and Jakobs’ high critical damage. The whole time, Jimmy’s team focused on making the gameplay experience feel satisfying and thematic.
• Finally, Jimmy noted to keep in mind that any decisions an artist makes will impact the departments around them; for example, clipping issues often arose when modular gun shapes would interact with player animations.

Jimmy Barnett – Q&A (answered are paraphrased)

• Q (Jimmy): There’s a lot of amazing stuff on ArtStation, and it seems like there is a set style of what the game industry wants. Should I cater to that, or try to stand out?
  • Jimmy: I would approach it in a slightly different way. Artists and tech are both improving, so you do have to compare your work to the talent level of the industry as a whole, not a specific person. Ask yourself if what you’re doing is in that realm. It’s a small industry, so that bar is literally your competition. Don’t let it intimidate you, but use it as a motivational tool. You can find somewhere in there to fit. Also, don’t tailor a piece specifically to a studio’s style, make your portfolio to just look as good as it can and showcase your strong points.
• Q (Jimmy): I noticed that there isn’t a Cogs listing for my specific talent. Should I still apply?
  • Jimmy: The Cogs program is slowing down at this stage in the project, since it’s nearing completion. It’ll ramp up again once the next project starts. You’re more likely to get picked up if you fit into a specific box, but we may accept generalists. We pull Cogs to work on specific tasks, so it’s good to have a specific skillset.
• Q (Jimmy): When and how do the guns get generated?
  • Jimmy: The gun generation happens instantly on the fly; the system makes a gun immediately after the loot request gets fired. Actually, all the meshes exist in each gestalt, and it just decides which parts are visible. That’s how we’re able to handle big mesh data quickly.
• Q (Jimmy): As a leader, does it help to be a generalist?
  • Jimmy: I use general skills often, but since technology is improving and increasing in complexity, everybody is starting to specialize. This is good, because the content benefits as a result. But even if you specialize, you should still know what the people on either side of you are doing and more or less how they do it. Your skills should also overlap a little bit even if you don’t do it, to eliminate some of the friction of handing them your work. Try to prevent issues from happening when you hand off content to the next team. If they have to make adjustments, that adds time and money. As a lead in particular, you need to be aware of how your work effects others so you can keep scope in mind when making decisions.
• Q (Jimmy): What is your ideation process for designing manufacturers?
  • Jimmy: For the existing manufacturers like Maliwan and Torgue, there’s lots of aesthetics and history already established and we build on that. For new ones, we work heavily with narrative so they can contextualize with rest of game. Concept artists are the people who “pen-to-paper” ideate. They sketch, figure out aesthetics and theme, and iterate. Eventually we start making proxy meshes, start prototyping, rig and animate. It usually takes around a week from proxy to prototype. And we always keep fun in mind.
• Q (Jimmy) Does your team have concept artists dedicated just to making guns?
  • Jimmy: Yeah, there’s one guy who just drew guns for two years. He could also outsource some of the work. At peak gun-making time, there were probably two or three guys just drawing guns. Like I mentioned earlier, people are starting to go to their cubbies in the industry and doing what they do best.
• Q (Jimmy) Regarding spectacle creep (the focus on making numbers bigger, like improving the “millions of guns” figure); is this a problem or a good thing?
  • Jimmy: Uncontrolled spectacle creep can be a detriment to game perception. Thankfully, Borderlands has built a brand identity around the “millions/billions of guns” concept. Overselling and underdelivering is a historical problem in entertainment, so it is something to be aware of, but spectacle can be used to your advantage as long as your game lives up.

These lectures were part of the Game Developer’s eXchange (GDX) Masterclass Lecture series. Keep an eye out for more GDX events in the future!