Some of the characters you can find in Apex Legends.

A case study in User Experience / User Interface success and failures through a video game: Apex Legends

Note: this post will be updated as time goes. It’s not a snapshot, it’s a WIP.

Note²: this is a VERY long article, going in-depth and telling many details about the game to help understand its design.

User Experience (UX), which is sometimes linked to User Interface (UI) and Design, when it comes to digital experiences, is hell of a thing. For long it was just about the “physical, non digital” world. So claimed the guy who brought this term to life, Donald Normal. We have a long story of ergonomics, satisfaction, utility, efficiency and interfaces. But since the 70’s and the start of the neverending era of industrialized software, we’ve had a serious boost in software production, creativity, and a market need to dinstinguish the products we’re working on. Since I spend a lot of my time examining websites UX (and slamming OpQuast / WCAG best practices into people’s faces, basically), I thought I’d jump off to another type of product for a while.

And along with that new world of possibilities, we’ve had a batch of new people to fight. In Apex Legends, a Respawn Entertainment game that appeared in winter 2019 as one of the most played games of the year (so claims its editor, with around 70 million players worldwide 6 months after it was launched), you need to fight other teams in a Battle Royale style game (Multiplayer Online Battle Arena -MOBA-), trying to survive however you can until you’re the last one alive.
In software design, you need to fight different kinds of ennemies until your product finally rests in peace: the money-driven world, project managers, shareholders, cost-killers, ROI-afficionados, and most of the time laziness, and many (many) other bosses, with all the meanings that this word can take.

As usual, when opening criticts to a piece of software, as a developer myself, I’d like to remind everyone reading this article that this is never about blaming anyone, especially not developers, but to see where success and failure hit. I’ll try to focus on failures, though, for they lead to better explanation. What’s interesting in this is to spot things that were in plain sight for many, but detected by few.

N.B.: This could at worst be a critic against game designers, but it won’t even be such a thing. After allowing myself over 2000 hours of gaming (and a few hundreds of euros on its Gacha-style store to basically open useless but sexy skins and finishers for weapons and characters, the game is using the now classic “free to play, pay to skin” model), I can say more than many that I do enjoy this game very, very much (I even hold a fun and useless world record statistic). But as for many high-budget, top-franchise games, the ones that you usually end up playing for a loooooong time, I’ve started encoutering a pretty large amount of UX failures and successes. Many of them are usually fixed and presentely as silent bugs (they’re not even in software update notes, for instance). The game developers did a pretty good thing setting up a Trello instance for the game, which, I admit, is quite a nice idea for a major video game license, by the way. First point for developers in UX, here!

So let me give you a journey about many things I stumbled upon in this game, that I’ll try to explain as best as I can. From both a user and a UX expert eye (which can sometimes be hard to differ when you feel a little too much involved). Let’s just sit back, relax, and talk a little.

The Champion Edition trailer (quite recent) sums up a little what Apex Legends is about.

The journey.

A brilliant game design

I won’t talk about this for a long time, but could do this for hours. Apex Legends was mostly successful for one very particular reason: it sums up and gathered almost all the game design failures and successes of its predecessors, like BattleGrounds (PlayerUnkown / Krafton), now simply renamed PUBG, or Fortnite (People Can Fly / Epic Games).

The most acclaimed feature in Apex Legends is by far its ping system. For that, I won’t go into details, but you can read this article about the UX success of the ping system in Apex Legends.

The maps have a better design, the game is a little more elitist regarding hardware, and despite better FPS (frames per second, the rendering unit from graphics boards), it still raises the bar a little higher. The results are pretty good. They keep less updates of the game, with scarse need to wait for immense downloads. They focused on characters, they introduced a better reactive automated dialogue system (much like Left 4 Dead did 12 years ago). Players follow a single, randomly picked and transferrable choice of direction while jumping from the sky. They improved the movements a lot. They added specific abilities (like you’d find in OverWatch -Iron Galaxy / Blizzard- for instance). The inventory management is totally click/press-optimizedand works like a charm… The list is so long I won’t bother sitting on it: the game developers clearly studied user feedback very well and found a great way to innovate out of it.

The mess of the core concepts

I’ll start with the basics. And go straight to the point.
It’s a common thing that many players never fully understand a game they’re playing. Especially when it comes to FPS (First-Person Shooters), which are not famous for having the most clever and brilliant communities. But as you could expect players of a large MMORPG (Massively Multiplayer Online Role-Playing Game), for instance, to never fully remember the thousands of statistics, formulas, places, numbers of those games, you usually expect FPS players to understand what their favourite FPS games are about. And let me tell you: you are absolutely wrong. Apex Legends is not exception. After venturing for thousands of hours into this game, and disussing with hundreds of players, even the most hardcore ones I could find (ranked Master — top 2%), I can assure you: players don’t understand how Apex works.

This is a common fact in the video game industry. There are needs to stick to previous beacons for players. Users need those beacons. Users need landmarks, things they can hook on. Whatever is new takes time to advent. What people once called “the lizard brain” still is getting solved by biologists. Therefore, lots of players tend to diverge from the original and intended path that the game conceptors designed.

N.B.: Most of the time over the reason of what I call the “fun over competitive but actually not fun” way, in which people can’t really position themselves as invested players or simply players passing by, and end up defending themselves with a time-related approach (the time invested segregating those two ensembles of player profiles), but this is not the topic today, and this slides away from analysis to personal thoughts.

As a consequence, players tend to diverge to a different way of experiencing the game. This divergence in usage is very often summed up as this image that you can find on the web here and there, which is debatable and criticizable, for it does not really reflect what UX means, but still debunks a little part of the problematic:

The common path metaphor to visually explain userland experience versus designed experience.

So, let’s dive into it a little. To sum up, Apex Legends has five main domains:

  • Simple game modes: classic, events and item training boot camp
  • Seasonal, limited rails of rewards: called Battle Passes, that cost around 7 EUR / USD, that require you to play daily to unlock bonuses per Battle Pass level for a limited period of time, or pay more to unlock those levels.
  • A competitive-oriented mode: Ranked games
  • Community management: friends and clubs
  • The store items and decorations, where you essentially buy a few items directly, and Gacha-like “Apex Packs”, that contain unlocked skins, or ressources to unlock those skins.

In Apex Legends, your account holds value through:

  • Unlocked characters you can play with (legends)
  • Unlocked visual and audio effects: loading backgrounds and music, skins and visual addons for weapons, finishers, taunting speech lines, and decorative ordnance (skins, charms and holosprays)
  • Currency. Three of them: Apex Coins (that hold the business financial conversion), Legend Tokens (to recycle daily quests and unlock legends), and Crafting Material (to unlock what the previous bullet point in this list holds)
  • Statistics. Per-character, per-account. That includes badges to showcase those stats: character statistic trackers and character/account badges. A user can select up to three trackers and badges for each character, and they’re shown all along the games to other players.
  • Ranked points and current rank in Ranked games
  • Your account experience level (a number reflecting in-game experience, with a displayed decoration badge)

That is basically what one needs to know when starting to play the game. Which, as you can see, is already a lot to know, but this is only the beginning. To solve this kind of problem (or to check that they don’t exist), it’s a best common practice to start usability tests.

Experience rewards

When you play Apex Legends, you basically fight in an arena until the last team (3 or less) of players is alive. That gives you points, that are strictly converted into experience. But here’s the first design failure: experience is scarcely rewarded. At first, it was barely rewarded with Apex Packs (the Gacha items to crack and hope) for the first 25 levels only. Then, an update after a few months boosted that initiative up to 200 levels. The reward was never understood by people. They didn’t know why they got those packs, nor where they came from. The reason? It was only explained in patch notes. It was simply not materialized in an educational way through the game. The game doesn’t have a playing tutorial, it only has an item tutorial, where you basically learn how to interact with things found in Apex Legends.

This is an interesting case study of game design: developers assumed that the game basics were simple, and that whatever was left out would fall under the “individual experience or perception”. Which was wrong here: perception took over part of the game mechanics.

Experience summary of games

When you play a game, whatever the game you play, you accumulate the following once it’s done:

  • Experience points
  • Battle Pass progress (if you bought one of those)

And most players (actually over 90% of the ones I asked for this article in my friends list) know how these work. The reason? Complexity and the lack of incentive. Let’s see.

First, let’s see what a match XP result breakdown look like:

A typical winner XP breakdown summary of a match
  • As you can see, what you get is points, converted into experience. You get them by just staying alive, and staying alive is the most rewarding thing you can do. Most people don’t know that.
  • You keep gaining points every second you’re alive, but also every second one of your teammates is still alive. Most people don’t know that.
  • It start once you get the countdown on the dropship to land from the sky onto the map. Most people don’t know that (precisely, at least).
  • You get 3 points for every second your squad is alive. Most people don’t know that.
  • You had double XP by the time this screenshot was taken, for a week. Most people didn’t know that.
  • You make around 70% of your points for just staying alive and surviving up to a certain position. Most people don’t know that.
  • You barely get XP for the most insane and hardest game actions, like reviving teammates, respawning them, dealing damage to ennemis, killing them, being the kill leader or killing the kill leader. Most of the points actually go to grinding, but most people don’t know that.
  • You get 1/4 point for each damage point done, but most people don’t know that. You get 500 points for your first kill each day, but most people don’t know that.
  • You get extra points for being the top 5/3/1 then, now 5/1 top squad, same again.

I’d say the reason for that is that, again, XP is not rewarding. Which means the core mechanic of progression is not valuable. Therefore, players turn to other points of interest. And that is the alternative path on the visual metaphor above. Here we see how a lack of clear goal for users leads to misunderstanding, and alternate uses of an information system. Plus: the recap screen at the end of games usually follows frustration of losing or excitement of winning, and gets bypassed in a snap. The solution against that, again, should be a game training mode to teach people the basics of the game, and, essentially: why they’re here.

Battle pass mechanics

Second point, one that we shouldn’t neglect is the misunderstanding of how the core money source for this game is not understood, pretty much the same way. Namely: the Battle Passes. There were mainly three eras in Battle Passes in Apex Legends. The game uses the notion of “season” to add unlockable content for a limited time, those seasons last 3 months on the average. Those Battle Passes are a great example of how Respawn Entertainment followed the feedback of players and started fixing things while listening to and reading feedback a lot, one of the basics to better understand one’s design mistakes. And for that, video games are great, in a sense that the users give more feedback than on any other product you can have anywhere.

The first season saw a pretty simple principle: there was normal experience (see what I explained just above) and Battle Pass experience. The first one has a breakdown as explained and illustrated above. The time survived XP was copied into Battle Pass XP. And the Battle Pass XP was boosted by the fact you play with premade squads (not only friends, but anyone you invited, like, your presvious squad, which you have a window to shortly reinvite them, this a misleading wording, by the way, one more UX problem here). And then, there was a bonus applied from your current Battle Pass progress, plus additional Battle Pass XP rewarded for each legend among the 8 available back then every week, up to a certain limit (25K points, whatever).

But this was too complex to track. Players had to check individually which legends they played or not each week, and many of them already chose a main character to play, either for affection reasons, either for competitiveness reasons (the character Wraith, for example, has a long history of being overly powerful and was modified up to the current season).

So Respawn Entertainmnt made a complete rework of how the Battle Pass worked. For 5 seasons, it was made more simple: you got only one experience bar showed, and you got quests (this is a now classic marketing mechanic to ensure audience fidelity and a daily recurring comeback of your players, just like the fact Battle Passes have 110 levels and not 100, those are classic user addiction and marketing patterns: looks weird, works great). Daily, weekly and weekly recurring quests. But still, this was quite hard to track. Also, the experience required to gain one Battle Pass level was not linear: it started every new week (that happens on Tuesdays, by the way, not misleading but still a little, no real antipattern here) with 18K points and ended with infinite 54K points required after a few levels. Again, this was too misleading.

Sample progress of a Battle Pass during a season (7).

The reasons for that? Mostly human reasons. The numbers were not multiples or fractions of 10, the natural number of fingers for a human (which, as you know, generated the practical impact on our decimal counting base). This makes memorizing harder and creates useless cognitive load.
Second, XP gained had less impact at the end of weeks, so players felt frustrated finishing harder quests on week endings, which let the impression of no gain. Also, the weekly quests are unlocked until the end of the Battle Pass, and three of them give one full level, which means you usually ended up having duplicate or redundant quests, raising the probability of unlocking most of them during the last days of the season, which created a feeling of stagnation at season start.
And on top of that, the average time required to complete the Battle Pass (the main reason why those exist, actually) was set too high. Many players complained they shouldn’t spend more than 3 hours a day to simply do routine quests, which they had to. And last but not the least: the numbers were all multiples of thousands, which made everything harder to remember: 3000, 6000, 18000, 25000, 36000, 54000, all partially multipliable by 1.0–5.7 (Battle Pass progress reward boosts), which, as you can guess, made everyone drop precise tracking and simply play and check instead of getting a plan of where to go.

Because yes, lastly, there was also a change in Battle Pass progress boosts: every player gave their whole team +10–90% XP normal experience boost instead of Battle Pass XP boost instead of just themselves, so this stacked up to three (the max game structure for a team), which ended up being… Between 0% and 270% gain (300% for the very last levels if you team was full and close to completion). But this boost only applied… To top 5 and more positions. Hard to track, eh? ;)

One antipattern that was removed, too, lies in changes of what the quests were. Most of the quests, as you can guess, were pretty simple (the game is just a survival game based on a FPS layer): go to, kill, damage, revive, assist, play for some given time, etc.
Which means the quests would alter what players would do on a normal path.

For instance, some quests required people to drop at special locations, which would totally hysterize people’s way of playing, making most players only log in to complete their daily quests, and therefore forget any tactics about the game itself. As a result, some people would just suicide, drop and leave, or act in such a weird way that their teammates would never have any way of interacting with them, or even to report them, or anything else. This example alone could illustrate how money-driven design (addiction management) can degenerate and create chimeras in gaming / user experience. The behaviour-driven experience of some affecting the experience of many others. Some quests would also require players to make kills at impossible locations, where no one would ever go, creating desperate game attempts and a never satisfied experience, filled with perpetual user frustration.

Those bad patterns of quests were later removed in the game. Better ones were then created. Instead of opinionating the quests, Respawn Entertainment made quests based on what players would normally do, reversing the way of piloting what they should do to access rewards. This, again, is a perfect (though late) example of adapting a successful model with wrong metrics.

Season 7 typical screen for the Battle Pass progress

So came season 7, and points were removed. Normal XP remained, and still boosts Battle Pass, as since season 1. But bye-bye Battle Pass points by the weird thousands. Now getting replaced with stars. Simple, countable on fingers, same mechanic, humanized values and metrics. And this works great. This received an overwhelmingly positive feedback, and players even got interested in counting their recurring experience level boost: it was 10K at launch, and lowered to 5K after only a few days only of players complains, making this Battle Pass the fastest ever given to players to rank up. Creating satisfaction. Over one year and a half after its launch, it seems Apex Legends has never been this close to stabilizing an income model that both generates money and customer satisfaction, without harrassing players.

In the end, it’s pretty simple: if users need to create whole blog posts / articles to decipher their own progress model, then you know you’re not going the right way, especially when there’s no gain from it and when you’re not especially rewarding complexity deciphering.

More UX & UI design mistakes?

Network concurrency in lobbies

Another bad design is how Apex Legends uses distributed networks and servers spawned on demand to scale players connections and elastic needs of connecting and playing. But remember: at a point in time, the game peaked at 70 000 000 players.
The game uses weird patterns, that I would call “network first, tracking second, UI last”. For example, when a player changes the game mode they desire to set up, it takes a few seconds to change, but the concurrency is not fully mastered, because it needs to have server feedback first (and you can clearly perceive that it’s not just switching, it’s also tracking). You need to wait for a server response to simply choose the queue you want to go in, not even the queueing itself. Weird? This means you can do the following scenario:

  • Get in the lobby, alone. (lobby leader)
  • Be joined by people in your friends list
  • Change game mode from Ranked to classic (named “Play Apex”) because your teammate is not good enough to sustain the Ranked mode you’d be going in (Ranked mode determines the difficulty level you’ll join based on the most ranked player in the team)
  • Start a game
  • Get in the queue to play
  • Having a game starting, including your current squad
  • Get the late response from the server for game change to Play Apex instead of Ranked
  • Enter a ranked game with an insufficiently skilled teammate who didn’t ask for anything
  • Or simply: have the UI stating that you’re queued for Ranked mode… And joining a non-Ranked game. Bad stuff.

See the frustration here? That means game clients should simply have a feedback from all the players in the lobby that they had the confirmation response sent to the clients before starting. Disappointing in load times, satisfying in quality.

Weird notifications

When you join a lobby, you actually join a lobby server, which can hold up to a certain amount of players. After which you’ll be assigned to the next one, and so on. It’s never shown, but you know it’s the only way to handle this, cause lobbies can be subject to timers, real time changes like notifications, players joining and leaving, game mode updates, next map rotation countdown, etc.

For that reason, sometimes you’ll just start the game, or end an Apex game and end up in a lobby… That doesn’t show. Instead you’ll get a weird, deceptive message saying “Server is full”, or “Server is shutting down”.

But you never asked to join any server willingly, you didn’t ask anything, actually, the initial load screen when you launch the game simply has one button saying “Continue”. So you get a failure error message for something you never felt like doing. You’ll not really notice the server choice possibility (data center, actually), barely visible on the bottom, along with accessibility features, that are… Almost invisible. Not so good for accessibility features, right? Also, for network-saving reasons, when you go AFK for too long, you get thrown of your lobby. Which is really weird, even though it is perfectly logical for bandwidth-saving concerns, but not for a player point of view. In general, designers should avoid mentioning mechanics that are never explicitely mentioned to their users: it’s ok to expose internal problematics and technical difficulties, as long as users are aware of what they’re involved in.

Apex Legends home screen (Season 7). Data center selection and accessibilty options on the bottom.

Unlocked skins scrolling

This one is even more frustrating. When you unlock a new skin, which, if you have bought a Battle Pass, happens pretty much every day, you get a notification (a red dot) that infinitely stays until you click on that particular skin.
And the ones you unlock ALWAYS appear at the end of the list, ranked by rarity. Which means every time you unlock a new skin, you have this click depth (press-depth for consoles): Legends / Loadout > Class (legend or weapon) > Subclass for weapons only, then you start scrolling, then > one more click / press on the skin to have the notification disappear.
The red dot is simply attached to the menu button, which is kinda “itchy” to look, and boring to remove. If you look at most streamers’ screens, you’ll notice they even stopped removing those notifications at a certain point, because it was too annoying. Why? Cause there are over 1500 skins to unlock in Apex Legends, no one is willing to click on those 1500 items through their gaming time.

This could be cleared by an embedded dropdown summary with the new skins unlocked plus a “mark as read” button, duplicated by a click / press on the skin itself, leading to the appropriate menu screen.

Also, there should be a loadout / skins notifications full clearing button, anywhere near the menus, like a subpart of the menu button itself that would instantly clear all skins notifications.

Keyboard/gamepad shortcuts

What would those four interface devices have in common? How to design a shared experience through those?

This is a huge classic in video games, especially when games are designed / ported to consoles, where there is no room for custom keyboard shortcuts, since there are not enough buttons/joysticks/physical user interfaces to configure. Usually the problems stand in unified video games development and building frameworks (like GameMaker, Unity or UnrealEngine, for instance). So nowadays, while cross-platforms considerations are at stake (like responsive design, device ability design, or interface-based design would be), many -too many- developers tend to forget to design working and clear interfaces in video games.
Most game conceptors wrongly assume that players build their user experience satisfaction from the live in-game experience, only when in action (and not in the meta-actions like menus or external tools linking like Steam Overlay, the Xbox Game Bar, etc.).

For those reasons, in Apex Legends some actions…

  • Don’t have a bound shortcut
  • Have a bound shortcut, but it’s not visually reminded
  • Have a bound shortcut that’s visually reminded
  • Used to have a bound shortcut that was removed (like the end game screen continue action that was removed in season 1)

Not only does this lack of coherence, but this really stops users from a perception of sustained pace. This breaks the cognitive flow. Players have to stop, even if they don’t want to see recap screens to move joysticks or mice cursors to the appropriate buttons. While you could tell this was done to prevent accidental screen bypassing and also to give some breath to the servers, this clearly is a classic antipattern in UI. A mostly common tool to solve interface complexity is eye tracking.

Legends and weapons precedence and choice

Choosing a character to play in Apex Legends (Season 7)

One little thing in Apex that was introduced is the notion of “favorite” for character and weapons skins. Though this constitutes a great addition to the skin selection, the game developers also added the possibility to mark several items as favourite and then select a meta-item that picks up a random favourite one. Therefore, you can play with a randomly selected character skin, and apply random skins on weapons when you pick them up (that’s when the weapon skin gets applied). This automates actions users would usually manually take, which also hits and gives value to the business model of the game.

But on the other side, you can’t choose something way more tactical and less decorative: the order in which you’d like to play legends (you can pick one among the currently 15 of them), and you can’t choose in which order you’ll assign the weapons on your primary and secondary weapon slots (you can only wear two of them). This creates noise routines in the game/experience flow.

There should be a weapon priority choice list. If you take common, popular games like Counter-Strike: Global Offensive, in this game you can bind your common shortcuts and setups. Even though both games have almost completely different weapon slots mechanics, this proves it is actually possible. Enhancing player experience by automatic common tasks (like Respawn Entertainment did for automatic weapon upgrade pickups, for instance, they automatically replace and drop if a better one is picked up) is a great idea to fluidify the techniques.

Legend choice at game pre-roll

While the game automatically makes you pick up your last played legend if you don’t pick one up yourself (like, being away from your keyboard/pad for 4 seconds, the delay you have to pick one), there’s no random legend choice like there is for skin rotation. I know many friends who actually try to improve by playing different legends all the time, and some of them try to level up and improve only a handful of them (improving all of them is beyond any human ability). This is a lack of coherence in player interface, creating mind stops when doing common tasks.

Also, the last played character is not chosen (validated and selected) by default, nor is it highlighted/reminded, though it’s the one that will be chosen by default. Players don’t have any clue what character is being selected… While the default selected character is already showing on the whole left half of the screen. This is a good example of a incomplete and incoherent design.

Technical statistics overlay overlap on ranked

This was fixed by the time I wrote this article, but when the game developers added the network/frames per second small indicator on the game HUD, they didn’t check that it would actually overlap the current ranked kills/assists. Nor did they check that it would be written to player profile, which was not the case for a few weeks. But this is fixed now.

Map daily challenges summary disappearing during special events

Map view in Apex Legends. You can see the empty space, now slightly in use.

This one is a pretty easy disappointment. There’s a space reserved while displaying the map in game, where players can see their daily quests. Those quests exist during the whole Battle Pass seasons (approximately 90 days). During special events (that last between one week and three weeks), the normal daily events (that you have 24 hours to complete)… Disappear, which makes players go back to the lobby to memorize and track their daily goals during a game… Which, of course, never happens. This is what I’d call an unpurposed disruption. Changing user habits for negative reasons is usually a clue in the presence of a UX regression, but not all of them are easy to spot and sometimes require more A/B testing.

As someone once said at IBM:

“Ease of use may be invisible, but its absence sure isn’t.”
https://twitter.com/ibmdesign/status/4423909367

No lag filtering

Well, you get the idea.

This is a major concern when building a game that involves players fighting each other. The cumulated network latency of both players can be a real frustraction. This has been progressively solved for decades now, but still doesn’t reach the human acceptable threshold. Most combat video games that involve ultra-fast reactivity go for different solutions. The most common one is to exclude anyone having excessive lagspikes. Some can be due to the infrastructure, some due to the network between players and servers, and some can be… Sometimes user-generated to get advantage.
Apex Legends uses a pattern of concurrency that tends to prevent the maximum latency spikes it can do, but there’s still a limit. To limit this, player latency is never shown, your own latency is hidden by default and you never get to see any other player’s latency. This is done mostly to prevent people from tilting and complaning about network lag (which is really actually unfair).
Clever, but you can’t trick the eye infinitely. Some players still litterally warp through the game, and that creates frustration in both parties.

This is very condemnatory, but sometimes, frustrating users because of elements they don’t have any hand on, that are totally out of their reach, is not the ideal solutions. A good pattern in video games is a disconnection, followed by a confirmation that nothing negative was added to the player profile, leaving them in a previous, stable state, not losing anything (and maybe cumulating the harmless positive items acquired during the time they could play).

Legends tokens, hover to show

This is a pretty bad update. Simple. Not the worst, not the best. There are three currencies used in Apex Legends (mentionned above).The most common one that players use in the game is Legend Tokens. The game has added more top menus items, and they simply hid one of the three currencies to gain some space. Now players have to hover the currencies to get the amount of the last, hidden one. There is clearly space for all of them. This is an example of a typical “developer-solved” (in a sense it’s not UX-solved) problem. No redesign, just cope with the existing. That’s a common bad way of orienting thoughts during a design process: throwing dust under the carpet doesn’t really clean the room, right?

News delayed display

This is a tricky one. Connect the following: players can’t start a game if a piece of news is being displayed on the screen, masking everything else, even if they clicked on “Ready” and got queued to enter a game. Then some news get displayed immediately (full screen pop-in) when players start the game. Then you have something called a blocking pop-in (a pattern that you can commonly see on websites newsletter subscription, especially since the GDPR hit the ground). This is vital to make sure people have seen the piece of information you want to show (which is a legitimate action, considering the game is free-to-play, pay-to-skin), but can lead to a lot of frustration when people actually got to see the interface, think they’re about to start a game, then… Get thrown out of queue.
It seems the news are subject to asynchronous network queries, that’s the key here: separate vital news from the other ones, so that people can’t dodge them, instead of showing them several times. Also, the pop-ins keep showing a few more times after the first one: this is a very bad antipattern.

No map training / exploration

While this might look like a weird idea, it is not. I explained this a little above. The main reason for frustration in Apex Legends comes from a simple fact: you can’t test the maps and explore them by yourself. Until the third map came (with season 7, over a year and a half after the game was out), where a discovery mode was added for 2 weeks (but removed later), there was no way to explore the maps. This is why most players a totally lost and why the learning curve of Apex Legends in substancially higher than it is for many other video games. The training maps, where you can test weapons and items, use a very small and totally different map from the ones you play on.

Now, the game has a 1:50" loading time before action on the average. This is the time-to-action on the average: geting ready, queueing for a game, loading the map and game items, picking legends, presenting your squad and the winning one, waiting for a small countdown before the players can actualy start having a reactive interface (which is not the case for the seconds before that moment, you can’t even escape to the main menu, which is another huge antipattern).

For those two reasons combined, players have absolutely no way to experience the game from a learning perspective. This creates, as stated above, a very (very) slow learning curve for everyone. This is another source of frustration for users: putting them straight away face-to-face with engagement (much like asking a credit card before entering a shop would be). When players drop fast from the ship and go for a quick fight, since they’re paired with similar skills players, they essentially have a 50/50 chance to kill and to die. Which means their experience will be summed… As a majority of the logged-in time being spent in lobbies, loading a game instead of playing, destroying the possibilities of progression.

Apex Legends training map: not one of the currently playable maps.

Character voices and dialogues over silence

Can’t disable your own character voices, especially when you’re the last one alive alone. Audio dialogues between characters are mandatory, but overlap all the rest of the game audio. Setting them at 1% puts them already excessively loud. When you’re alone, your characters keep talking to themselves, and placing memos to visualize the last ring position tracking when the game becomes extremely tight pollutes the ambient audio with useless dialogues, which can hide the environment noises… Including the footsteps of the squad that will wipe you.

Ranked demotion protection

Ranked mode has a long history of being the most frustrating things in Apex Legends, when it comes to player experience. First, it doesn’t have a fluid progression. That is, levels get exponentially harder as you reach the top ones. Players get paired with and against other players with a similar tier (Bronze, Silver, Gold, Platinum, Diamond, Master, Predator, each from rank IV to III to II to I), except that since the levels are exponentially harder to reach, there are actually too few top tier players to create servers at any moment. Which means they’ll end up... On lower tier servers!
Maths show that only a fraction of the players on each game will progress towards the next major rank (entry nagative points compensated by final position and up to 5 kills per game). The tiers are separated by a HUGE difference in skills and the gap between each tier tends to be more and more a problem as times goes. Also, to prevent bad things to happen, the game developers added a protection against tier demotion for players: if you reach Platinum IV rank, you can’t lose points that would make you fall into Gold I again, which creates an amazing stagnant pool of players at the tier IV of each major tier. The vast majority of players remain at those positions.

Looking at stats of the last season, you can guess the outcome:

As of 10/21/2020, we have the following distribution among players in Split 2 who played more than 5 hours of Ranked (versus Season 5):

17.76% Bronze (16.21%)
26.23% Silver (22.83%)
36.25% Gold (33.7%)
17.75% Platinum (18.21%)
1.89% Diamond (2.51%)
0.12% Master & Apex Predator (0.2%)

Abilities reloading missing timers

Abilities reloading show a progress (loading colour) status, but not the remaining time it will take. Adding a countdown of seconds remaining would be a perfect answer to that.

Inconsistent inventory pings

This is a slight mistake, but it’s pretty simple to explain: when looking at your inventory, you can ping items for ammo, you can ping ammo slots for more ammo, you can ping empty slots for powerup, but… Can’t ping filled slots for better powerups.

Again, when users are deprived of a functionality, they create a userland path with an alternate, less desired way of fixing the solution. In this case, this makes players communicate with microhpones to explain what they need, and mics are one of the less working things, both in Apex Legends and when it comes to hardware quality.

Apex Legends inventory management. A very fast and successful interface.

Partial color blindness coverage

Players have additional colors on the main HUD, but using colorblind mode doesn’t affect them (shared colors between players). This chould change colors, there could be numbers or letters instead, or simply levels of gray: white, gray, black. Same: the infrared optics don’t adapt to colorblindness: they always show enemies in red. Not such an accessible feature, eh?

Ranked progress misleading

Ranked sublevels indicators don’t reflect the last gap between tier 1 and the next rank. They actually show a scale of equal parts of points need for the first tiers… But not the last one! Pretty much the same way that special events progress bars are NOT linear, showing the same distance between the first 200 points and the last 1000 points.

Dead ping limitations creating misleading information

Once dead, you can still communicate in the chat or with audio (your microphone), ping your death box (without any throttle, which means you can litterally tilt and ping twice per second, blanking out the game audio for the teammates still alive), you can ping the nearest respawn beacon to your team, you use the map, but you can’t use map pings to clarify what you intend (like choosing a safer respawn beacon). This is a major source of misleading information for players: they don’t know, sometimes, what they’re going to ping.

Server lag creating model-first, user-later instead of the opposite pattern

If the game server is lagging, you can’t return to the lobby at anytime like you normally would. You have to wait for in-between-servers communication. In the meantime, the whole game… Gets blocked.

Chat gets hidden as you type

That’s a pretty old annoying stuff in Apex: the chat gets hidden as you type, between game segments: picking a legend, displaying squad, displaying the champion squad, and during game before you die, after you die and when the game ends. Not cool, is it?

Player muting

In Apex Legends, you can mute your teammates, but you can’t know if they’re muting you, so you might speak to someone who’s doing so while the other player is not. An icon show that, next to the player’s name. Although some names can be pretty long (some already overflow the players’ names location containers), there’s simply room for an extra icon. This is a very old problematic in player annoyance management: letting the annoying and noisy users in doubt/misinformation (like Facebook does for page comments management, for instance, where you can isolate comments, restricting them visually to the users who wrote them and their friends) or using total transparency that can lead to even more frustrating behaviour (like assaulting the people muting you, for the sole reason that they muted you). This problematic is not yet solved by anyone in the video game industry: handling user toxicity in a shared environment is hell of a difficult thing to do.

Dead spectating inconsistency

After dying -i.e. losing the game-, you can only spectate the squad that finished your last teammate, instead of the squad that downed you. Sometimes that could be a different squad. Then, when they die, you can only do the same for the squad that did the same to theirs, and do so. Since you can’t cycle through all the remaning squads, but only that one in particular, there is a chance that you never get to spectate the squad that downed you. If one of their member is cheating, you won’t be able to report them, for the spectating stops when the last squad wins.

If you want to read more about the importance and impact of consistency through experience, I suggest that you read this article.

Map auto position: how a good idea can become a bad one

When using the map, you can zoom and drag the map to another location. When you exit the map and then open it again, the zoom level gets memorized, but the maps get reset to encompass your current character position. While this seems like a pretty good idea for new players and slower-action players like people who play with a game pad (even the analog sticks can’t outrun a mouse speed for that), this is a counterproductive design when reaching the last, precise circles on the map where you need to be ready to ping specific locations.

Wraith. The iconic and most played character in the game since start, the first one that had a storyline event.

And that’s it! I’ll try to keep this up to date as patches and seasons go through.

I think this demonstrates how a common lack of UX experts in the largest projects can lead to disappointment in user perception, alternate user paths and misleading, even in places where everything seems easy, even when your product is a fantastic one (which, again Apex Legends is).
When creating less meaningful results, you slowly lose the value of your products features, and that, is where you always need a continuous UX check of logic, consistency, feedback, A/B testing and practical userland field testing, amont many other tools to create a more hedonic approach of your design and delivery. After all, sometimes you need to be a designer and do a little Gemba Walk… Into the game communities themselves!

We are now in a world where emotions drive people, and where special care must be taken when it comes to the gaming industry: the one where people sometimes seek refuge when everything goes wrong. Those hard times, seeded by the COVID-19 situation, tend to exacerbate those emotions. And, as a great man once said:

“Cognition attempts to make sense of the world: emotion assigns value.”
― Donald A. Norman, The Design of Everyday Things.

Hope you had a great journey through this excessively long article. I had a lot of fun and pleasure writing this down and noting all of this. If this becomes popular enough, I might to the same for other video games in the future.

In the meantime, feel free to add me on Steam and come have some fun: this game is really great.To end on a more personal note: it’s been over 10 years that I didn’t have that pleasure with playing a FPS game, since I finally stopped Left 4 Dead 2 in 2019. As stated in the introduction, this has been my main and favorite game since march 2019, it drains all my time -and money-, and I’ll say it’s definitly worth it, so… Kudos to the developers! \o/

Should you want more info regarding the game itself, just go here: https://www.ea.com/games/apex-legends

--

--

Developer / Lead developer / Web Artchitect / Technical Innovation since 2007. Also, I do photography, music, playing MTG, video games, and writing, a LOT. IAD.

Love podcasts or audiobooks? Learn on the go with our new app.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
William Pinaud

William Pinaud

Developer / Lead developer / Web Artchitect / Technical Innovation since 2007. Also, I do photography, music, playing MTG, video games, and writing, a LOT. IAD.