Launching online games is not an easy task, and doing so during a pandemic throws bigger, nasty curveballs into the process. So when Proletariat took Spellbreak from early access to release in September, it was dealing not just with the pitfalls of finalizing the launch status of an online game but also managing how to do so with a distributed team, each working at home with different machines, different internet connections, and different distractions (some cuter than others).
Spellbreak is an intriguing battle royale. You equip gauntlets that throw magic (think fireballs and such), and it has far more verticality than Fortnite or PlayerUnknown’s Battlegrounds. But with the recent Spellstorm update, Proletariat went beyond new content and rewards by adding story-based quests. This is one of several reasons why executive producer Cardell Kerr labels Spellbreak a “battle royale RPG.” We’ll explore the others in this edited transcript.
Time for an update
GamesBeat: This is the first update since official launch. How is it going to change or refine the game?
Cardell Kerr: Technically, we’ve had a couple of patches beforehand, so we’ve added other features. We’re no strangers to live content. Our studio is made heavily from the DNA of Zynga and MMOs. In that sense, we knew we were in for a large amount of work in general. This is our first big update, though. The patches fixed a lot of issues in the month after we launched, but this is adding the full chapter system. This is how we want to convey story and content about the evolving world of Spellbreak to our players.
GamesBeat: How many players do you have now?
Kerr: We have enough, but we always want to have more. The only figure we’ve released so far is we have more than 5 million.
GamesBeat: How has player behavior changed from launch to now? Did you have a big rush, and then players dropped off?
Kerr: I’d definitely say that — when we launched, we didn’t do very much marketing at all. That was not our goal. We just wanted to see if we could stick the landing as far as being a simultaneous platform launch title. There aren’t too many examples of studios that have been successful doing that. If you asked me eight months before September whether we would be one of those studios, I would have laughed. When we first launched, we did not plan for the influx of players we got. That said, we were obviously surprised and delighted about the number of players we got. But since then we’ve been just trying to fix the problems we had. We’re trying to fix up our game in general. We’re trying to remove a lot of gaps in features that we knew we wanted for launch, but we didn’t have time to build. That’s what the chapter system itself is. Our plan is to go much wider with this update than we did for initial launch.
About player behavior, I’d say when we launched, we really had no good idea or model for how we thought players would play in a battle royale game across platforms. I say that — it sounds weird with the amount of time we spent in alpha and beta, but adding a platform changes how players play at a fundamental level. The fact that you can now group up with your friend who plays on PS4 means that a player who plays maybe once a week might go to playing every day, because now they can play with their friend. Before that, maybe they just played once and didn’t come back. I’d say that over the course of the last few weeks, months, some players are playing every day. They play for a long period of time. They play a very high skill game. We also have a lot of players who come and play a few days a week. They don’t leave. They don’t stop. They just always play a few days a week. That’s a really good sign from my perspective, because it means that the game is interesting enough for them to not just play a couple times and drop it entirely.
Then you have a lot of players who play a bunch at the beginning, and then come back for patches, but they’re mostly only interested in content. With all that said, it’s allowed us to be able to understand, well, these are the types of attributes our players have, which has allowed us to be more surgical about the types of features we want to build in the future. For this first update, we wanted to finish the game that we launched. The chapter system is effectively that, because we always wanted story to be a big part of the game. Part of being a battle royale RPG is you need to have a story to go with it. RPGs without stories typically don’t exist. This is more of a completed vision as opposed to iterating on the player behavior we’ve seen so far.
GamesBeat: What is the story you’re trying to convey with chapters?
Kerr: We’re going to dive into — how did the world get to be the way it is? There’s a bunch of backstory before you can tell people where it’s going to go. We ultimately want to look at the fracture in this case. Why does the fracture exist at all? What’s happening with the fracture? I assume you’re familiar with the fracture in the game. Obviously it’s fairly unique in BR to have an area you can fall into where you instantly die. It’s pretty controversial, putting it in in the first place. Not a lot of people were happy about dying regardless of their level of health or armor. But we kept it in because narratively, we knew what we wanted to do with the fracture, and it was important that people respected or feared it. Trying to explain more of where it came from and where it’s going is most of the focus of chapter one.
GamesBeat: You’ve been out now for a couple of months. Is that idea of instant death in a battle royale still controversial?
Kerr: I’d say that — you know, it’s so hard. The interactions you have with players tend to be a lot more spiky, is the best way to put it. You’ll read a post on a Reddit board or your Discord raging about the fact that they fell into the fracture. By the same token, you’ll see a bunch of people make a bunch of content around fracture fails, which are some really funny clips of people just falling into the fracture for weird and strange reasons. I think it will always be a part of our game in the sense that it is an outlier. And therefore there’s a learning curve there. If you come from any other battle royale, you’re going to run into the fracture at some point. It will be shocking to you, because it’s unprecedented. That’s the reason why it will always be a bit of a shock. We’re still acquiring new players. Those new players come in mostly organically. They have no idea what to expect, and therefore the first thing they tend to focus on is the differences between Spellbreak and other BRs. Some are very obvious: You’re a wizard. You’re flying in the air. You’re throwing fireballs and lightning bolts. Those are obvious differences compared to other BRs. By the same token, you’re effectively just using weapons, projectile weapons, which makes it closer to other BRs. When you’re a wizard flying through the air, though, and your rune runs out and you slowly fall downwards and end up in a hole where you die instantly, that’s something very different for most people.
A different approach to growth
GamesBeat: You said you were focusing on organic growth. Have you just been going by the promotion you’ve had in the Epic store and other methods?
Kerr: It’s a bit of a hybrid model. When I say organic, I mean you’re looking to empower people who already play your game to bring in their ever-growing circles. There’s a certain amount of just baseline, making sure people can find us on several platforms. That’s when you’re seeing us on Epic or on Xbox or PS4. You want to make sure that people can hear about Spellbreak and find it in the store relatively easily. We’re listed in there and people can easily find the free game. There’s a baseline of work that we do, that we commit to. And then there’s a layer on top of that of just upselling through social media. All of the pieces of content that people make for the game because they’re having a good time. That’s what I mean, mostly, by being organic at that stage. We haven’t done a wide-scale blitz on that front. You don’t see us doing large takeovers of Twitch or anything along those lines. I don’t even know that would be great for us as a game. We’re a game that’s unique enough and different enough that it’s important, from my perspective, to arm the players we have to go and explain the game to people and figure out amongst their friend circles whether they enjoy the game at all. Rather than bringing in a bunch of people who are not very savvy about the game, and they come in and really dislike the game because it’s not what they expected. Even though it’s not really clear what they would expect. But either way, then they leave.
Chris Cook, Amplifier Group: I can add a bit to that. The idea of using the current community as evangelists — the players we have are big fans of the game. They want to create content and spread the word. We’re making sure that the Proletariat team is supporting them and enabling them to go out and spread the word to people who want a new game to play or enjoy battle royale. Just making sure that the team internally is giving resources to players to help spread the word on that organic level. Not necessarily paid marketing, that level.
GamesBeat: Is that going to change with chapters?
Kerr: We’ve done a bunch of research on this one. The thing that will likely change with chapters is we’ll do more of what is working. It has been working for us within very small channels, where we’ve done publicity. You can expect it to go much wider once chapters come out.
GamesBeat: One last question about how you’re promoting the game. Who are the top Twitch streamers for this?
Kerr: It’s a lot of the veterans that we had play our game for a good chunk before. For a while, SypherPK and DrLupo were really on it every day. I’d say that since then it’s been more focused on — it’s itscamski or — I can get back to you. I’m sorry. I don’t remember his name right now. But these people are longstanding with our community, which makes it even more embarrassing that I can’t remember his name. They have effectively ridden that wave very much upward. They were already high skill when the game launched in September, and they ended up teaching a lot of people how to play. They’re seen as icons of the community, and to be fair, they really have been that way. They’ve taught so many people how to play in our Discord, and now that they’re cutting content that’s being seen by thousands of people. Those are the people we typically end up seeing more of. We did a Twitch Rivals event at the end of October, which was also another big one, to your point. People need to see the game to understand it. I don’t think descriptions do it justice. It’s one of those games that’s very polarizing. Once you witness it, you’re either on board or you’re not. There’s not much gray area there. Maybe to dip back to what I was saying before, it’s about empowering those content creators to get their images and clips far and wide.
GamesBeat: Which platform is most active? Is it one of the consoles, or is it PC?
Kerr: I’d say it’s really close. Our most active right now is PC. But PS4 is a close second.
GamesBeat: Doing a live game like Spellbreak and supporting a new generation of consoles, an older generation of consoles, and PC, how challenging is that for a team to do?
Kerr: [Laughs] It’s extremely difficult. I would say that — one of the things that — everyone always talks about Fortnite. Obviously it’s a wildly successful game. Generation-defining game. The amount of work they do in order to do what we’re doing right now is amazing to me. And I don’t think anyone is aware of how hard it is to launch simultaneously on multiple platforms and then stay in sync across all those platforms. A lot of developers will launch on PC and then launch a month later on whatever their other platforms might be. Being able to support all those platforms is extremely challenging. Here’s an example I’ll use. We live in a COVID world. A lot of people can’t get access to the office. In order to emulate the platforms, you need a dev kit. QA needs test kits. Those test kits can no longer be in a communal area, because that would be a COVID risk. You need to not only coordinate the testing with people who have the equipment with them, but you also need to ensure that any issues that are platform-specific get caught through a level of rigorous testing. And every time you add another platform, it’s a multiplicative problem.
GamesBeat: Are you finding that supporting all these different platforms with COVID — like you said, with the dev kit problems and the fact that you can’t work communally in the way that you would before — has that slowed the pace of patches and updates and finding players?
Kerr: There’s a very interesting GDC talk I could give on this particular topic. In order for us to launch — everything shut down in Massachusetts back in March. We ended up launching in September. Back in March, I can tell you that our game didn’t even run on some of the consoles we launched on. We ran on PS4, but we weren’t capable of running well on Xbox at the time, and we didn’t run on Switch at all. I would say it’s kind of a mixed bag. The sad fact is that in a global pandemic, people can’t just — no one is going anywhere. I don’t know if you’ve noticed this in your experiences, but if you would normally take a vacation with your children, you’d just pack up and go somewhere. Even if that’s not necessarily somewhere you want to be, it’s still a change of scenery, a change of pace. You can share experiences with people. That is mostly off-limits right now.
What ended up happening is, there was a clear lack of efficiency when we started working remotely, but since then we’ve found enough methods of working that we’ve been able to make up for that. But there was definitely a point when people were working because there was nothing else for them to do. If anything, in some cases it was actually helpful to keep your mind off what appeared to be the growing level of anxiety within the states. No one wanted to get sick or contribute to other people getting sick. In some cases people were stuck at home with their 2-year-olds, though. Daycares were closed. Work had to change for the parents of very young children, because it wasn’t a world where you had core hours that could apply anymore. During those core hours, those people had to deal with their children. Having children myself, they’re 10 and 12 at this stage, but I remember what they were like when they were young. There’s no meeting that a 2-year-old will not disrupt in order to do what they need to do at that time. We had to find a workflow that allowed those people to have the freedom necessary to raise their families, support their families, while also being able to work. We spent the six months, five months leading up to launch figuring out that workflow and reoptimizing.
Right now, in terms of additional development, I would say we’re pretty good at it. The biggest challenge we have right now is maintaining open lines of communication. In the office you have the benefit of overhearing a conversation. You don’t have that when everything is either communicated in voice chat and Discord with limited numbers of people, or done in Slack channels that you’re not a part of, or that you don’t read until days later. That’s been our latest challenge. That’s a significant portion of what I, as a producer, have to deal with alongside the production team. Trying to find better ways to communicate and give people context for what needs to get done. That’s probably our biggest challenge right now, but we’ve gotten good enough at the actual cross-platform delivery. That process has in fact smoothed out. The biggest challenge now is, once you add new platforms, you have to figure out where they sit. An example would be — we’re launching on Steam as part of chapter one. There’s a whole separate set of things we bring online for that, a whole separate set of functionalities we have to work with there. The good news is that once you’ve built your own ID system, it’s easier to let other platforms plug into your ID system. That’s good, and we’ve already tested the ID system extensively. Otherwise we wouldn’t launch with anybody. But you still need to add another layer of exit passes, another layer of things that can break in unique ways on Steam. It’s no different than what you mentioned before about the new consoles. That’s another layer of problems that we have to deal with. We are at the point now where they’re fairly parallelizable. It’s just a question of whether we can maintain our efficiency.
GamesBeat: With this process you’ve had to adapt to, as everyone has, have there been any unexpected successes that have come out of this?
Kerr: Yes, actually. I’d say there’s been more than a few. One example would be, we often will screen share. What we found is that when we — normally it didn’t perform fast or well enough for artists to give people critical feedback. We investigated trying to do internal private Twitch streams or broadcasts, but ultimately we ended up using Discord for a lot of this work. It lets everyone witness what people are doing when they ask for feedback. It’s been great. In the past you’d have to have everyone huddle around a monitor. Once you get to about eight people, some of them are way back and can’t see. They’re somewhat checked out. When you screen share, everyone sees what you’re doing in full detail. The feedback they give you is going to be so much better. It’s way more targeted. It’s a massive success. Being able to do feedback sessions that are clear, transparent, and actionable, it’s just gotten better. That’s a pretty big one.
Another big one is, almost every company works with outsourcers to an extent, and on top of that, we have a bunch of employees that are all over the world. We have employees in Japan, on the West Coast, in Brazil, in Europe. We are semi-global. When everyone went remote, it actually made the team much more cohesive with people that weren’t in the office. It meant that everyone has to handle their communication in the same lines. That meant our ability to work with our remote employees grew to the extent that now they’re literally interchangeable with anyone else. Before you had to compartmentalize what they worked on, because inevitably there would be problems with communication. Everyone was frustrated with it. We said we would just try our best to break off things as atomically as possible, whereas now, everyone is fully integrated. Those are probably two of the easiest, biggest wins that come off the top of my head.
GamesBeat: Have you found any issues with art assets because of different screen resolutions or different machine specs? Or did you send all the machines from work to people’s homes?
Kerr: For me, I didn’t bother taking my machine home, because my home machine is strictly better. That was a no-brainer from my perspective. Other people run the gamut. Some people took their machines from work. Some people had to order new machines, because their work machine wouldn’t work with their living arrangement. To go back to people who had young children at home, they needed a laptop, and their work machine was a desktop. That transplant had to happen. Ironically, as I was saying before, asset review has gotten better because of the streaming process I was talking about before. There are only benefits, that I have found, to having varied people using different setups at home. It’s almost like you’re getting a mandatory compatibility lab test every time you run a play day or get people together to review an asset. Inevitably someone will say, hey, this thing you sent me seems busted. They’ll send images and it’ll be clear that the thing is broken on a certain monitor, at a certain refresh rate. Those issues are found much earlier now. In that sense, it’s working out fairly well. It’s a little frustrating in the moment, obviously, because you just want this to work so you can move on to the next thing. But as I often tell people, getting something to work for multiple people is more useful than getting it to work for one of them and moving on. I just finished a book called Upstream, and my new mantra internally has been, how high upstream can I go to prevent a problem? By having all these people on different PC configurations, we’ve moved the problem upstream.
Before, we had only one configuration in the office, and then from there QA might find a problem later, or maybe an outsource group would find it when we run a compatibility lab test. Then we’d not only not have the hardware to replicate it, but we don’t even know why it happened. Versus now the problem has been moved upstream. It happens to developers while they’re working on the asset. That means it gets solved while they’re working on it, which means there’s less onus at the end of the funnel to fix the mistakes from the top of the funnel.
GamesBeat: I’ve seen some complaints about empty lobbies. Have you had a dropoff in players?
Kerr: It’s fascinating, actually. We definitely have fewer players now than we had when we launched. There are lots of reasons for that. One of the biggest problems we run into — our matchmaking process is not the most robust. When we launched, it was good enough to launch with, and we thought it would be fine. It turns out that it’s a little too fractured. By virtue of that, if a person ends up in an empty lobby, it’s not because they’re not going to play. It’s because we’ve put them in a bucket, and that bucket isn’t prescriptive for what would be a good player experience. I’ve been a pretty die-hard proponent internally, for quite some time — the two things I care about never making worse are framerate and time to match. Everything else can be changed, but you never want the framerate to get worse, because you never want your game to feel sluggish or unresponsive, and you never want it to take longer to find matches. As a result, part of our push right now is to redo our matchmaking system so that people can be guaranteed to find a match quickly. It’s a very valid complaint. It’s one of those things that hits people who are opting into a restricted lobby — they only want to match with Switch players or PS4 players — or conversely playing a low population area. Say you’re playing on Switch in Brazil. There’s not a lot of Switches in Brazil, as an example. You might have challenges there. That’s where it mostly hits people right now. The whole purpose that we’re trying to work on now is to fix that.
GamesBeat: On the matchmaking side, you said it was too strict. Are you trying to keep really skilled veteran players away from the new players?
Kerr: That’s certainly one part of it. I’d say the other part of it is just — in general, we’re much more level-based than you might assume. When a person first starts, there’s an expectation for the lobbies and lobby types they’ll have. But once you get to higher levels, you’ll be in different lobbies. It’s also doubled up on by the fact that we want to make sure that ping is good for people. When we spin up different servers in different regions, that’s another constraint. You have the double dip of not only being confined by matchmaking, but also being confined by relative skill level. And then I guess the triple dip is then being confined by platform. All those things have acted together to make matches harder to find. In some cases I’ve even seen reports of people playing at peak hours and not getting matches, even though I know we have plenty of matches happening. In some cases those might be bugs, but in general, it is more a problem of player fragmentation than it is player population.
GamesBeat: How are you going to address this?
Kerr: We’re going to do a combination of things. I don’t ever want to take away a player’s choice about how they want to play. If you feel die-hard about not ever playing with a person who’s not on a console, I don’t want to take that choice away from you. Technically, based on console acceptance criteria, I don’t think that’s even possible. Reasonably speaking, Sony and Microsoft, at the very least, have maintained the ability to do that as part of their promise to their users. But I do think that — what we’re much more likely to do is find ways under the hood to say: “Hey, we notice you haven’t found a match yet. We’ll find other geos that are hopping right now so you have a better chance of getting into a match.” That means changing how we handle matchmaking now almost entirely, where it happens more in the main menu than it happens in the lobby. You would end up with more of a countdown clock there to matchmake you to the appropriate place, so we’re respectful of your time.
Throwing down the gauntlets
GamesBeat: How is the onboarding for new players? Is it hard? Do they get beat up pretty badly?
Kerr: I’d say that actually players start off OK. You’ll tend to fall off a cliff probably around six to eight hours. That’s when you’re in the, “hey, we can matchmake you with a lot of different people.” Assuming you’ve been playing matches faithfully over those six to eight hours and you queue up after you finish a match every time, it doesn’t take long for you to end up in a pool with the majority of players. That’s where it gets hard. It’s one thing to be a new player who’s playing against other new players. It’s another thing to be playing against itscamski, who’s been playing this game for years. They often will do things that you didn’t even know were possible, let alone do them in a way that’s effective in combat. That’s been a bit more difficult, certainly. But one of our main focuses has been trying to shift things away from winning and more toward achieving other things in the course of your match. In some ways that’s what the prologue system gives people a taste of, and the chapter system will give you more of a feel for it. There are very concrete things you’re doing as part of the chapter system that affect your matches. They give you a feeling of getting better at the game without necessarily having you get eight kills and cut a highlight reel on YouTube. That’s unlikely for someone who’s been playing the game for a fairly short amount of time.
GamesBeat: Do you find that using gauntlets instead of guns, the way most battle royales do, and the way your class system works — do you find that confuses players? Does it take time to adapt to? Or do they figure it out pretty quickly?
Kerr: We were actually worried about confusing people. We did a bunch of focus groups years ago. Why don’t you just use staffs or wands? Everyone was saying, if you’ve read Harry Potter you understand wands. At the time we were very much saying, we like gauntlets. We think gauntlets are important. We didn’t want this to be a traditional Gandalf simulator. We wanted to focus on that breakers are young, youthful, athletic. Gauntlets seemed like a good way to do that. Ironically, we benefit from what Marvel has thrown out there with the Infinity Gauntlet. The Infinity Gauntlet, as a concept, what it even is, made gauntlets intuitive to people. Especially when you pick a class. You have that gauntlet when you first enter the game. I don’t think anyone has batted an eyelash about gauntlets being the source of elemental magic. That’s been decent. In terms of classes, classes are also widely used enough. People understand what they’re signing up for. There’s sometimes a bit of confusion over how to gain levels. Unlike most games where you gain levels, you don’t gain XP. You typically will just gain a level the instant you walk through a new circle. But even then, once people understand that, the system is actually pretty simple for people to wrap their heads around. The complexity comes in with understanding the interactions between your class and other classes, and understanding the abilities of other classes. That’s where things get confusing for new players. But what we’ve found is that most new players, they effectively do a dilettante run through the classes. They’ll play a class, unlock the talent within that class, and then shift to another class to play that class and unlock the talent there. Which gives the benefit of exposure to all the classes and abilities. They understand the possibility space well enough to survive versus a player that might have exponentially more time in the game than they do.
GamesBeat: Has there been a class or class combo where players say, “Do something about this, we hate it?”
Kerr: Oh my. I’d actually go so far as to say that there’s no class combo where players aren’t saying, “Do something about this, we hate it.” It’s funny because of the fact that — I actually talked about this with our design director quite a bit. There are multiple philosophies of balance. Some people’s philosophy is, “Hey, we want to make sure that there are only a few ways to play the game in a fair way,” and that’s fine. And there are other philosophies that say, “If everything is broken, nothing is broken.” I’d go so far as to say that we tend to err more on the latter side than the former. What I feel like Spellbreak does well is that it gives you a feeling of power that very few other — I’d even go so far as to say games — very few other games do, let alone other BRs. Part of that is going to be brutal effectiveness in certain cases.
Depending on your player skill, different classes are differently broken in your hands. Some people who are not necessarily the best at the game, they will often gravitate toward stoneshaper first, because they’re not that great at aiming. Individual projectiles are too hard for them to handle. They stick to a much more wide area combination of attack and defense abilities with stoneshaper, and it works well for them. Really good stoneshapers are people who use their boulders effectively too, but that may not necessarily be the case earlier. So new players will often say, “Stoneshaper is broken, it’s broken entirely, you should fix it.” What is happening is that once you get to the deep end of the pool, stoneshaper is actually viewed as a very weak class. Which isn’t altogether true either, because there are numerous people who’ve won tournaments or placed highly with stoneshaper as well. It’s mostly just knowing how to use the right tool at the right time. Right now at least, that’s the case. I’d say toxicologist, the poison class, that got a lot of heavy scrutiny initially because of how it combines with invisibility, with deception and elusiveness. It made it so that people felt like they had no fighting chance against the class. There have been some tweaks there.
But for the most part, the image we’ve always had in our heads for the toxicologist was effectively an elusive assassin, who arrives and kills you before you know he’s there. That was performing appropriately. Each of the classes has their power fantasy, and each of the classes can be viewed as extremely broken in its own right. That’s actually been good for us as a game.
GamesBeat: What’s your main? What do you like?
Kerr: I personally love lightning. I’m just a really bad shot. I love conduit, just because I love building more recharges. Burning hot in Spellbreak just feels so good. There’s very little to me that feels better than having multiple plague runes or multiple shadow steps or multiple dashes and the ability to effectively use a lot of your spells for zone control. You’re making sure that a person can’t flank you on the left while you’re fighting a person on your right. Or, conversely, you use it on yourself to make sure you can get your own effective supercharge that gives you more options than before. I love that particular aspect of it. If I were still in my youthful gaming years, I’d be a much better shot than I am now. I’ll often couple conduit with an offhand gauntlet that allows me to be a sloppier aimer. That seems to be — toxicologist, poison gauntlet. Sometimes it’s an earth gauntlet. But for the most part that’s the road I like to ride.
GamesBeat Gift Guides: