Baked in Accessibility: How Features Were Approached in Borderlands 3

[Music] hi everyone I'm Andrew bear and I was lead UI programmer on Borderlands 3 I left gearbox software at the beginning of the year but they've graciously allowed me to still get this talk today I'll be talking about how we approached creating accessibility features or as we saw it any feature that lowers the barrier of entry for our games so more people could enjoy them of course this includes things like colorblind settings and subtitles but it also includes features like difficulty settings these are some of the accessibility features we added to Borderlands 3 and these are the ones I'll be talking about today these features were chosen randomly but rather have specific stories I'd like to tell in order to help you all learn what worked and what didn't for us I'll also be weaving in some organizational stories for those of you in leadership or production rules so that's what I'm bringing to the table but I'm making two assumptions first I'm assuming y'all are devs which seems like a pretty safe bet and secondly I assume that if you're here you already know that accessibility

is important if you're a designer you want a better experience for your players if you're a business person you want to accommodate a larger audience if you're a producer you want the best bang for your buck in terms of features and if you're a programmer like me you just want to work on the really cool feature that everybody's going to use now if you're not convinced why accessibility is important don't take my word for it we were successful in Borderlands 3 and this talk is only a case study about that successful approach for accessibility features there are plenty of brilliant people out there who are talking and writing about why accessibility should be at the forefront of dust minds I strongly encourage just listening to advocates and outlets for a while these are my favorite sources and I referenced them heavily or spoke with them when creating features and creating this talk Karen Steve's research at EA is second to none and Ian Hamilton and sherry Thompson are wonderful advocates additionally the AbleGamers charity does incredible work helping gamers including creating custom control rigs and finally

the can i play that organization has fantastic articles in review including borderlands 3 is perfect mobility score and seven out of ten deaf scores now my team and I are just UI does but that's no scuse us tubs still have to make an effort even when we're under the gun to create a much-anticipated triple a sequel this is exactly where we found ourselves on board unless three to give you a better idea of what I mean our story begins the wave long back forgotten age of 2017 it was a whole new world and everybody was relearning the engine from code to design - localization - audio for UI in particular we spent a month or so using Yui for su mg and realized it just wasn't a good fit for us we had a couple of reasons for this but the largest was on our UI designers were struggling with the transition from Adobe's tools to UMG and we just couldn't do what we needed to in the schedule we were on so on June 1st 2017 you I went back to the devil we knew Scaleform this CL was basically our

starting missile for UI over the next two years and one month which is nine months less time between Borderlands and water board lines - we would build Borderlands 3 UI from the ground up since all these changes meant we had no starting code or content unlike when moving from BL want to be ill - we basically had a partial style guide in a dream we still created over sixty hundreds and 33 menus that gold and for comparison Borderlands 1 importance - had about 30 to 40 hundred it's depending on how you count and about a dozen menus Borderlands 3 had the most complicated skill tree and inventory yet and we still kept split-screen two-player top bottom and four player co-op a few bullet points on a slide doesn't really do it justice so to pull just a few examples first we have the humble crosshair board is 3 is a shooter and we needed to actually create a deceptively complicated crosshair I'm showing 6 here

but we only have one or weapon type with a few variants mat to match the manufacturer for example that second one from the top and for the left is the alternate version of the TD or shotgun crosshair then we have the action skill widget on the right is the traditional action skill which is showing the Beastmasters raccattack ability and on the left is the Beastmaster Spirent ability both of these have charge and cooldowns and error states the player tries to use it when they can we also have an experience bar and we have an item card so you can see all the stats for your loot and Borderlands 3 is a co-op game so we have buddy widgets so you can see your friends stats and pet health and shield bars and a boss bar and overhead nameplates and a 3d mini-map and an inventory system not to mention a skill system a big map and my personal favorite the options menu there's quite a few bits I didn't mention but when you add in the meetings the iterations and the bugs on top of

the implementation you start feeling like this tired Gordon UI we created isn't perfect but it's a small miracle we pulled it off and it's all thanks to brilliant and dedicated UI team at gearbox everyone thinks their team is the best but I know from experience that I worked with some outstanding individuals I don't say all of this just to brag but because it wasn't just the UI team working hard everybody was firing on all cylinders to get this game to the state it did so when I approached our creative director about starting an accessibility strike team it was no surprise that he said only if it didn't derail all other work that since we had already tried to fit more game in less time we would need to be clever in order to keep our efforts in the margins and not to track from the main game I would submit to Spain dismayed by his response if I'm being honest but I understood why he reached the conclusion and we pressed on so that's where I want to focus today at

the intersection of no time no dedicated resources and an already mountain of work to do elsewhere I'll be telling six stories that I think best summarize our approach first with our backs against the walls our best move was to waste time on slack in the beginning I decided we'd meet once from the outset and to outline our goals make sure people could put faces name therefore it was necessary to find a place we could serve as our home base without requiring anyone to physically move we created a simple chaplain room anybody in the company including our sister studio in Canada could this gave a voice to everyone allowing any designer programmer tester whatever to raise concerns ask questions or even offer suggestions it also provided a place where we could post articles these articles we posted and more importantly the resulting conversations from them was by far the best thing we could have done for the team we were able to see

what worked for others and define a standard for ourselves we were forced to constantly hone and strengthen our beliefs and designs also by constantly pinging the chat room with new articles to read it kept accessibility in the foreground it wasn't just an afterthought to us it was what we talked about day in and day out to articles in particular really stick out the first one talks about Domino's Pizza and Raoul's a court case headed to the Supreme Court at the time it's a fascinating read and I strongly recommend folks to get out the other article was from Ubisoft and talked about the usage metrics other accessibility features the data showed just how widespread the usage was and how it demanded the industry start pouring real resources into their development these conversations kept people informed interest coupled out with everyone at gearbox being able and encouraged for a dissipate it paid massive dividends as you'll hear in the other stories this is where we started not with a green light from our creator vector not with a kickoff meeting but in

these discussions without this home-based return - without it constantly stoking the flames accessibility wouldn't have been the back of everyone's mind as they went about their work since we had that open line of communication we received a few requests for example one day from during pre-production while we were still figuring out what Borderlands 3 would be and how we would build it a member of our QA staff posted in our chat room that she encountered motion sickness easily and was concerned that the proposed head bob system especially while moving in the recently pitch mac iron bear would be a problem recognizing the potential year our animation team explored possible solutions as I mentioned earlier we were still in pre-pro and there was just enough time for our animation team to experiment with a few ideas before proposing a solution a player modifiable slider and that would alter the magnitude of the head Bob with the ability to turn it completely off it sounds so simple and

straightforward but with all the gears moving around us with all the chaos of development we weren't confident from the get-go after a day or so of experimentation it only took the environment airing myself about thirty minutes to implement the solution had we not started talking about accessibility what we did with as many people included as possible we probably would have ended up with some tact on afterthought of a solution if we could find time for it at all now we have a neat little multiplier without reacting anyway the takeaway here really is to start talking about accessibility as early as possible with every person possible you don't have to make concrete plans but talk early and often and talking about accessibility with everyone can lead to some unseen benefits for example one day while working on the item card we realized that it would be difficult for some folks to identify different rarity types at a distance early on we decided not to spawn the item card at large distances

just like we did with BL - but experimented with a handful ideas and to help identify the gear these ideas included displaying a small icon while looking at the item like we do in the ship game when the player is a bit closer showing this icon at a larger distance really got in the way of normal play especially combat and it didn't end up being something that we actually went with we also thought about displaying an additional icon on the mini-map like how legendaries are displayed in the shipping game but this was scrapped over clutter concerns after a while we realized we had hit a brick wall and we asked folks to think about it as we turned our focus elsewhere a few days later our absolutely world-class effects director shows up with this on the side he had taken it upon himself to create loot beams at scale and intensity and size with rarity level the jiff is a bit artifact but even if you completely disregard the colors in this jiff you can see the differences in the heights

in the bloom effect at the base of beam and how many wispy particles are they did from the lower portion of the beam common all the way on the left has the shortest most bland effect uncommon directly to its right is slightly taller and has a larger base rarer is taller still has a slightly larger base and the occasional wispy particle epic has the second tallest base and a constant handful of wispy particles and the legendary effect is larger than the gun itself obscenely tall a torrent of particles at the bottom and even has dedicated star icon on the mini-map you could see in the top right we were totally floored by this approach we hadn't considered looking at our humble loot beams it took a unique perspective to rethink what information beams can convey and unique skill to articulate that concept so the takeaway here is to empower everyone to experiments not every idea will make it all the way but if you encourage everyone to have ownership they'll stun you with their

ingenuity and unique approaches honestly I hope us devs begin to treat accessibility the same way we treat performance or memory it's everyone's responsibility to make the game accessible not one dev or even a group of devs off in a corner somewhere a few days after these loot beam changes with him this lesson was further cemented when one of our audio designers on the team heard about what we did and create a unique sound effects for the different rarities dropping it was a well-received change and our players have this almost Pavlovian response to when they hear on legendary drop in the world it's incredible and it's all things to our devs feeling like they were empowered to experiment on the other hand empowering isn't a silver bullets for this next door we thought we had the right idea but we fell short unlike movies or TV just displaying what characters a really misses critical audio cues our players needed inspired by minecraft closed caption system that you can see here we really wanted to display additional text

prompts to alert players just specific events in our world that they needed to react to we also concern we were also concerned that closed captions triggered too frequently or a large number of them would take up too much space and much needed screen real estate and overall lead to a negative experience so we opted for captions that would only message comment information that was needed immediately that wasn't already visually message somehow in the final version this meant that we displayed closed captions for barrels igniting or exploding within the range of the player as well as when base embosses begin certain windup animations as you can see here in the center and the right images respectively it also meant for example that we didn't caption grenades landing the other player since we had a dedicated widget for them as seen on the far left this wasn't received very well essentially and I

100% agree in retrospect the reviewer is calling out that we simply didn't caption everything we should have that we missed environmental hazards or additional attacks from boss and non boss enemies and they also called out that we should have captioning motion or sound effects in our subtitles that by limiting ourselves to combat captions our dialogue lost some of its punch when being read in the end nothing beats testing with your target audience personally not being able to find the time in our schedule to make more user testing happen was our number one flaw we should have figured out a way to focus test these features more thoroughly so that when we shipped the game we started off on the right foot that said sometimes we don't have to look very far to make sure that we're on that right though fresh off of battleborn we wanted to provide expansive color options to accommodate players as you can see here on battleborn we gave players the ability to recolored not just parts that

you I like the crosshair and the nameplates and health bars but also the detailing of the minions the centuries and even the build notes when it came to Borderlands 3 we started wanting to implement in his quiver an idea of color swaps it's about this time that I learned that one of our UI designers is colorblind I had never known and it was none of my business it just simply hadn't come up until now he's a fantastic artist and he really took the time to dive in deep and figure out what color information we needed to have alternate messaging for what he found was that rarity was almost always in plain text and when it wasn't that plain text for the rarities name was always quick to obtain for example you can navigate to an item and see its rarity quickly in all of our item menus the same is true for the elemental damage types which always have unique icon icons and are always spelled out in plain text he also figured out that the crosshair was the only widget that

relied solely on color to convey whether the player is aiming at her friendly an enemy or neither when we realized this we just couldn't justify the amount of time we need in our Rd crammed scheduled for a full color wheel or even for a color picker for just the one widget and opted for a set number of baiting colors a player to choose from he and some other devs figured out what would be the best preset color options and with his approval we knew we had a solid list without spending unnecessary time making something that would be overkill for our specific aids the takeaway here is simple challenge yourself to think elegantly the large deluxe version of a feature is nice but is it the best use of your time can that time be freed up and used elsewhere I can't say for certain just how much time we saved going this route let alone what it was used for but I'm certain everyone here will agree with me that our time is precious and needs to be saved where it can be is so we can absorb the nasty

surprises to somehow always rare there has the worst possible time speaking of less than great surprises even with buy-in from the team some devs with their minds on deadlines and with the very best of intentions won't be easily convinced that accessibility should be a higher priority as I'm sure anyone here can attest you these devs have to make hard choices on what receives their attention at what doesn't every day and it's sometimes difficult for folks to remember what a wide-ranging impact accessibility features can have even with our in the margin strategy we still had a few instances of this but one particular example sticks out and early milestones of and a designer had learned about the head Bob scaling feature I mentioned previously he only learned about the feature itself not about how we approached it or how little time we actually spent on it he assumed that we

had spent a lot of time and resources on the future and vocalized his dismay about our approach by learning of the feature from the results and not from the process he was concerned we were shirking our other responsibilities he posted in our slack channel and before I could even type out a response the conversation quickly became heated on the one hand this is a spark of high drama aimed at a powder keg of anxious - trying to make deadlines but on the other hand I was really glad he felt confident enough to say something in such a public way it maybe wasn't the best phrasing but I'm sure he wasn't the only one who came to such a conclusion and it gave us a chance to either confirm that we were on the right track or have an outside observer call so something we missed in the end taking the time to explain the process one on one in our case over a beer after work turned a powder keg of a situation into something that reinforced our confidence it also turned a vocal skeptic into a vocal opponent of our approach so the takeaway here is

nothing surprising make room for dissenting opinions and approach them and those with them with respect rather than drama so those are my stories as a quick recap and if you want to take a picture or a screenshot of this slide now is definitely the time chatrooms keep accessibility in the foreground talk early often and with everyone in empower experimentation test often think elegantly and leave room for dissension I'm going to pause here for people to take screenshots ok here are some of the articles that I use either for this talk or when discussing features and approaches with the team in addition to my twitter handle and those that i frequently communicate about accessibility options i'm going to pause here but other than that thank you for listening to my talk