Foreword
I’ve been wrestling with whether or not I should write anything about Dread Hunger for the last eight months. Post mortems can be incredibly illuminating but there is a very real risk of painting people in a bad light or misrepresenting situations that occurred during a game’s development. With that said, please bear in mind that these thoughts represent one person’s perspective on the development of Dread Hunger, and are not a complete picture of everything that happened in the last 3 (ish) years.
Games are incredibly complex and difficult to build, and things that seem obvious in hindsight were almost certainly not so clear at the time. If you’re a developer or even just someone who has dabbled a little in game development I trust you understand what I’m talking about. If not, I’d encourage you to give it a try some time. It can be one of the most rewarding and fulfilling experiences of your life, or one of the most punishing and bleak.
In our case, it was both.
A Game for a Pandemic
Dread Hunger began development almost four months before Covid 19 gripped the world, but the vast majority of the game’s development took place during lockdowns, in an environment of uncertainty and fear. Inevitably, the game took on many of these characteristics.
The goal was to build an experience where players could sail through a frozen world, immersing themselves in the grim realities of a 19th century Arctic expedition. As Creative Director, my job was to maximise the player’s engagement with that scenario and insert them as vividly as possible into the role of an arctic explorer. To that end, I chose “First Person Shooter” as the rough template for the game. It was a decision that would have larger consequences later, but at the time it was done simply to push that feeling of immersion in the world and stick the player directly into a pair of snow dusted boots.
Playtesting began in March of 2020 with a simple prototype that included a ship, a handful of weapons, and some very basic AI enemies. Our earliest tests didn’t include any kind of social deduction or betrayal ; they were purely cooperative experiences. We wanted to see if it was fun to simply sail a ship, gather coal, and forage for food and supplies. If we could nail that basic setup, our bet was that social deduction would be the needed “spice” to elevate it into something special.
One side effect of people being stuck at home in the early days of the pandemic was that everyone was glued to their PCs and working on the game. Between March and December of 2020 we developed the game in record time, building systems and iterating at a pace that still makes my head spin. We went from having a bunch of default UE4 mannequins walking around in a box to sailing a giant ship around an arctic world teeming with rabbits, wolves, and craftable items — all in around ten months.
The Calm Before the Storm
One of the biggest dangers of developing a game in isolation is that you start losing track of who you’re actually building it for. We had a functional, decently fun prototype at this point, but we really had no idea what gamers would think of it. Building something you think is cool is great, but building something everyone else thinks is cool is how you stay in business.
With that in mind, we decided to start a round of curated beta testing. We had builds on Steam and were making use of Steam’s new playtesting feature which allowed us to bring in bite sized groups of people to try the game out and give us feedback on it. We opened up a Discord server and started engaging with what was rapidly becoming a budding community around our game. It was really cool to interact with people and get their feedback on what features they thought would be the most fun.
Looking back, this was essentially the “Honeymoon” phase of the game’s development. Ideas were in rich supply, the fans were excited, and we still hadn’t spent enough time or money to force us to focus on the realities of running the game as a business. We could just pretend we were gamers, and build a fun game!
Trial By Fire
It was Q1 2021 and at this point the game’s development had stretched on over a year. Playtesting had been successful in helping us fine tune the game’s concept but we were rapidly hitting a point where the costs were getting too high and we needed to start recouping our money.
We made the decision to enter Early Access as soon as possible, targeting an April launch. Some fans were upset by our proposed price point, that their ideas hadn’t been considered, or felt they hadn’t had long enough to play the game before it went public but we pushed on and launched anyway. We didn’t really have any other choice. The launch itself went fairly well, with a modest burst of sales and some generally positive feedback from players, but two weeks into Early Access we started getting some disturbing reports of “griefing”.
Players were killing their own teammates regardless of whether they were Thralls or not. Worse, we were hearing reports of racism, sexual harassment, and all sorts of other “colorful” behavior over comms. I don’t think it ever occurred to us just how much time and energy we would have to dump into moderating player behavior. We certainly expected that there would be a few bad apples here or there, but the sheer volume of reports and the types of things we were seeing people say and do kind of sent us into a panic.
I could blame the Pandemic and say that people were feeling cooped up, antisocial, and angry. I could blame the type of game we made and say that it fed these negative behaviors, or I could just point the finger back at us and say we were too naïve. Regardless of what you want to believe, our plans for the content roadmap had to be put on hold as we went full force on trying to build moderation systems to give us insight into what players were doing in the game, and how to curb toxic behavior. The mood shifted from “how do we build a fun game” to “how do we stop people from being shitty in our game” which dramatically impacted the mental well being of the team, and started putting us at odds with our community.
We had to start becoming “the bad guys”.
We built some pretty robust moderation tools and started banning people who were exhibiting extreme antisocial or disruptive behavior. It started off with temporary bans but we quickly realized that many toxic players were repeat offenders and would often create alt accounts so that they could continue to grief even if their “mains” were banned. In some extreme cases we had players create dozens of sockpuppet accounts to ban evade and grief.
The Fall
By fall of 2021 we were all pretty exhausted.
We had spent the past 6 months battling toxicity, released 4 massive content updates, and were trying to forge a path to release, but a large segment of the community was still vocally dissatisfied. Gone were the days of positive uplifting comments about the game. Instead we had angry veterans leaving long rants on Steam and Discord about how an update had “destroyed the game” or accusing us of failing to ban griefers.
We were seeing players who had invested thousands of hours in the game suddenly turn on us over fairly minor balance changes, or make unreasonable demands for new content at a time when we were only just barely able to keep up with roadmap promises and moderation.
This was the dark side of community development.
Passion is a pendulum, and it can swing just as easily toward hate as it can to love. It can be pretty jarring watching players go from saying that Dread Hunger is their favorite game to accusing us of incompetence and saying they hate our guts.
A schism also developed in the community around this time between players who enjoyed the social deduction elements of the game, and the players who enjoyed the more action oriented aspects. Our goal had been to offer both groups something to sink their teeth into, but it was clear that we weren’t going to be able to please either group. On top of all this, our sales had slowed to a trickle. It seemed like whoever was interested in our game had already bought it, or was going to hold off until we hit 1.0.
All of this is to say that as of October 2021 Dread Hunger had, by all practical definitions, failed. The game had made back less than half of the money we had spent developing it, we had a pissed off community with ever escalating demands, and it was becoming clear that the social deduction hype Among Us spearheaded had melted away in western countries. The only thing that kept us going in the months leading up to launch was the hope that 1.0 would give the game a second chance at life.
For better (and worse), it did.
China :)
If you had asked anyone on the team which country our game would be most popular in, I don’t think a single one of us would have said China. And yet, that’s exactly what we started to see when we looked at our Steam demographics in the weeks leading up to launch. It started as a slow trickle of Chinese players and then turned into a deluge at launch. By March of 2022 we had over 100 thousand concurrent players, 99% of whom were Chinese.
Nobody saw it coming.
There is sometimes a perception that when something becomes explosively popular, it must have been planned. How could it have been? We had no money to spend on marketing and were within a hair’s breadth of just giving up on the project altogether. What happened was that an extremely influential Chinese streamer started playing Dread Hunger. It spread to other streamers through Chinese social media platforms until it was one of the most popular games in their country.
That’s it. There was no conspiracy, no secret back room deals, no pro China propaganda secretly baked into the game’s code. Just a dude who got people excited about our game, and in turn gave it a second chance at life.
We barely had a chance to let out a sigh of relief before we started facing backlash from our English speaking community. They were (understandably) confused and frustrated because the game’s server list started filling up with words they didn’t understand and their lobbies were being joined by players who spoke little to no English. It wasn’t long before we started seeing reviews like “Game has been invaded by China” and “Chinese game, stay away”. Since the game relies on voice chat for communication, this new language barrier was proving to be a deal breaker for many players. Our review scores took (yet another) beating.
Suddenly we found ourselves in the position of needing to add language filtering to the server browser, and make sure our UI properly supported Chinese characters. We also had to start translating update posts into Chinese and reaching out to the Chinese community to understand what they liked and disliked about the game. The silver lining in all of this was that the influx of Chinese players brought with it enough money to justify further development. We broke even about a month after the game took off in China and were profitable by February of 2022. As bewildering as the situation was, it was starting to look like Dread Hunger might actually work out …
China :(
In late February we started noticing a few issues with our servers. They were dying and restarting seemingly at random. It soon became clear that they were being targeted by DDOS attacks. The attacks intensified throughout March and April to the point where the amount of data being sent at our servers was on the level of what might be directed at a large bank or government website to take it offline. Our infrastructure engineer was staying up around the clock watching for attacks and attempting to preempt them, while we invested heavily in beefed up DDOS protection. Still, the attacks continued and their severity only increased.
Someone obviously wanted to wipe Dread Hunger off the internet, and it soon became clear based on the amount of data they were sending at us and the frequency of the attacks that it was a fairly large and well funded organization. DDOS attacks aren’t free to launch, and some back of the napkin calculations put the attacker’s expenditures in the tens if not hundreds of thousands of US dollars.
It was pretty terrifying to contemplate.
An angry player launching a one off attack on your servers is one thing, but a targeted, calculated assault on your game by a well funded organization is another thing entirely. When you’re an indie developer planning your game this sort of stuff doesn’t even make it on your “worst case scenarios” list. It was completely insane. What was even more frustrating than the attacks themselves was how the Chinese community seemed to react to them. Our assumption was that the players would see us as victims and give us some leniency while we tried to fortify the servers.
The opposite happened.
Almost immediately our review score dropped 10–20% with a vast number of Chinese reviews citing “weak servers” as the reason. We tried reaching out to members of the Chinese community to explain the situation, but the mood among Chinese players seemed to be “you have money, do better”. This was a pretty baffling stance, and it revealed a lot to us about how Chinese players perceive popular games. They seemed to feel that a game as successful as Dread Hunger couldn’t possibly be run by a mere handful of beleaguered developers. We were seen as a “big” game, and we were expected to respond to these problems in the way that “big” studios do. That is, we needed to throw overwhelming amounts of money and staff at the attacks to just make them “go away”.
We weren’t really in a position to do either. Our tiny team was only just barely coping with the (now constant) attacks, and slowly chipping away at a new content update. Hiring takes time and energy and we had neither. It started to feel like we were drowning under the weight of expectation, and the team morale once again took a sharp dive.
As if the DDOS attacks weren’t enough, cheating was also on the rise. In the first few months after launch the cheats we were seeing were largely boilerplate Unreal Engine speedhacks. Things you can find almost anywhere on the web. By Q2 2022 we were seeing highly advanced cheats that were custom built for Dread Hunger by obviously experienced cheaters.
A quick browse on google will show you just how popular cheating is in games, and China is no exception. In March of last year Tencent helped the Chinese authorities bust a cheating ring worth $43 million USD. They have dozens of employees, an office, and a fleet of expensive cars. These aren’t a group of script kiddies so much as a full blown cheating Mafia, and they (or groups like them) were developing some insane stuff for Dread Hunger.
As with the DDOS attacks, players were quick to blame the prevalence of cheaters on us, and once again, it was exhausting and completely unfair. We were basically the David to China’s cheating Goliath, but Chinese gamers saw things the other way around.
It was around this time that our relationship with Chinese players started to fall apart.
We couldn’t keep up with their expectations, and we were being slowly crushed to death by the combined weight of advanced cheats, DDOS attacks, and demands for new content. Add to this the continued friction between the English and Chinese speaking communities, dwindling sales, and a pretty negative review score, and you have a toxic cesspool from which nothing good can emerge.
The End
As of July 2023 all of the developers involved in Dread Hunger have moved on to other things, with Tencent investing in Digital Confectioners in September of last year to fund two new games, while my team is experimenting with some new game ideas of our own.
At this point Dread Hunger is on life support , and will continue to be until revenues drop to a point where keeping the servers running is no longer sustainable. It’ll be a sad day when that happens, but given how much time players spend playing Dread Hunger and how little recurring monetization there is, it’s more or less inevitable.
I believe our main failings as developers were three:
- We expected the game to be just a modest success.
- We didn’t plan for a language barrier
- We underestimated how toxic some players would be
No sensible developer ever goes into a launch believing that their game is going to be a massive hit. That’s how you fail big. Hitting just 1K CCUs was a big dream of ours, and we would have been blown away with a couple hundred thousand sales. We blew both figures out of the water, but it came at a great cost. Members of the team were harassed, some of us had mental and physical health problems as a result of the stress brought on from working on the game, and overall the mood toward the end of the project was one of frustration and sadness.
It didn’t end how any of us hoped.
Despite all that, Dread Hunger will always have a special place in my heart. Of all the games I’ve made it’s the one I think stands strongest in its sense of time and place. It’s an incredibly beautiful world with exciting characters, a unique historical setting, and when everyone is playing it in a fun way it can be an absolutely amazing experience.
To everyone who stayed positive and supported us through a tumultuous development, thank you.
Alex