With over 30 million accounts cross platform, Dauntless routinely provides no shortage of challenges for the game’s team at Phoenix Labs. Community lead Danielle took the opportunity to put some questions to members of the Dauntless team - including producer Teresa, technical designer Anthony, lead environment artist Eddie, and product lead Jordan - about the unique traits of operating a live service game, multiplayer design learnings, and current development influences.
The team released a major update late May. What’s something you or the team is especially proud of, that might not be readily obvious?
Eddie, lead environment artist
Radiant Escalation’s night atmosphere. While it might look like a simple swap of existing “night” atmospheres, there was lot of detail work done with post-processing volumes to make sure that it felt like night time as a whole, all while still ensuring arenas where Behemoths show up are properly lit up so gameplay isn’t negatively affected.We’ve learned a ton about this from previous patches. That knowledge now allows us to do really purposeful reveals of points of interest and have lots of “mood” contrasts throughout our islands.
Additionally, we set scoping expectations for this Escalation extremely well, and that’s what guided our work. I think this is the best we’ve done so far in terms of hitting our targets on time and having time to go above and beyond.
Jordan, product lead
I’m really proud of the way each discipline really owns their space, but is also very comfortable giving each other feedback. It’s a remarkable team that is open and comfortable talking about what’s working and what isn’t.
That’s a really healthy attitude that just results in a better experience for our players overall. With Alyra and Radiant Escalation, we had daily team playtests and iterated until it all felt right.
What kind of differences are there between creating a live service game like Dauntless versus (for example) a more traditional single player game?
Anthony, technical designer
A live service game is iterative at its core. Decisions have to be made a lot earlier than they would be in a “boxed” game, to ensure that players get content at a consistent rate and that we’re using our time wisely.
The feedback you get on a release happens a lot quicker and is more direct than it is on a boxed product, so you have the ability to make adjustments sooner rather than later.
Feedback for sure. I remember developing for boxed games could sometimes feel like working inside a box with no idea if I was making the best choices.
I love how quick and steadily I get feedback from the community which I can apply right away. This is great for me because I have a lot of pride in what we do here, but it’s also great for the community because they can see that we care about being responded to quickly.
^ more on feedback, I can literally go ask players if something I’m working on is something they’d like, or if an idea for an improvement is something that would work for them. That’s so great.
Supporting a live game aligns with how we develop –being agile. It’s also a balance of planning long-term and short-term development, and delivering engaging core features and content on a regular basis, and listening and being empathetic of players' needs, and engaging players and incorporating their feedback.
Our patch release cadence allows us to regularly showcase and update the game while being responsive. We find this consistency of delivering content while engaging with players helps build and foster trust.
Live service games are exactly that: a service. It’s easy to fall into a trap of wanting to always change core systems to improve this or reimagine that. But: players are here in the game expecting a service. We have to serve them.
To use a clunky metaphor: If you went to a restaurant, and instead of serving you food, the staff just kept changing the chairs, polishing the plates, and fixing the tables – all of which are good things that make the restaurant more comfortable– you’d eventually leave hungry. The biggest challenge for a live service game is to find out what the food is. We need to really understand what inspires players to log in and play, and then balance that with all other improvements.
On the other hand, we also have to always be thinking about the shared space of our game world. Every encounter can have one to six players in it at any given time. Usually that number changes during the fight. Each player has different goals and different ways of approaching content. They’re at different power levels and have different amounts of time they can spend in the game. The shared space needs to serve all of them.
We also have to balance our brains between the patch that’s live, the patch we’re preparing to release, the patch we’re working on, the patch we’re planning, and the patches beyond. And each individual on the team is focused on different parts of the game at different times. So I’m sure it’s true everywhere, but particularly in a live service game, organization and communication is KEY. We would be nowhere without producers like Torri, Josh, Teresa, and Tevin on our team.
How do you balance player expectations with design goals?
Making sure that we understand what we’re setting out to do, setting the right expectations early, and following through on them.
This is a big one! We talk with players every single day — whether it’s on our discord, on reddit, on social media, during one of our Staff Slayer events where we play with the community, or even in-game.
We are always listening to their feedback and hearing about their hopes and expectations. For a live game like Dauntless, there are obviously things that are just out of reach, but we can try and deliver the right kind of experience.
We do a lot of work to set expectations on our roadmap or during our Developer AMAs, and have the most success when we set clear expectations and deliver exactly what we’ve promised.
Are our goals blocked in any way by present technology?
I don’t think the team is blocked by technology per se. The team and studio is very open in exploring and using different technologies.
The tools and software we use change and get updated frequently. For example, the Unreal Engine has many great updates that improve workflow and/or expands our ability to make the game work and look even better. It’s a matter of learning and evaluating which of these updates would benefit the game and the team.
Our team is resourceful, has a wide range of experience, and is always willing to help each other out. In general, the team is enthusiastic about learning different technologies that could help with development, improve the game quality, and of course add to each of our skillsets along the way.
Our imaginations are always going and we’re always coming up with cool ways to experience the Shattered Isles, but the reality is that Dauntless is already live and players are already playing. This is something we have to keep in mind.
In terms of technology, Dauntless is full cross-play on all platforms, so we have to make sure everything is working well across each platform.
Dauntless has been on consoles for three years, but available on PC in open beta for a time before that, and in development long before that. The team has done a huge amount of work upgrading our systems and tools behind the scenes to help us deliver more exciting Behemoth encounters to players, but technology moves so quickly it’s almost impossible to stay at the forefront. A lot of what we continue to learn on developing Dauntless is helping other teams at Phoenix Labs plan their games and build their technology. This is all really exciting!
Are there any particular game development/engineering/design problems the team is collectively waiting for someone to solve?
Dynamic atmospheres! We’ve been looking into this, but we aren’t able to change the atmosphere dynamically in our maps at the moment. If we ever are, we’ll be able to do things like adjust environments for time of day and/or really ramp up the differences between arenas.
Example: Having full-on swamp areas where it becomes dark and foggy and the lighting completely changes.
How does playing other games inform/benefit Dauntless’ development?
Because we are in game development, I think you can’t help but look at other video games with a different lens and try to identify elements, mechanics and nuances that make those games “work” or “not work”. This could range from seeing if the narrative draws you in (or not) to assessing if the interface is user-friendly.
At the same time, I think playing different games from various genres resets you creatively and helps to bring in different perspectives of art and development.
Almost every game, regardless of art style, platform or genre, strives to have visual reveals of cool points of interest in their worlds. This is something I’m always looking for. Every cinematic moment is a learning moment for me.
Also, playing games helps me develop better games because it helps me have fun and charge up with good game making energy!
What are some of the games the team is currently playing?
Aside from Wordle, I recently replayed a bit of The Last of Us. The HBO series is coming out, so I was inspired to revisit it again.
I’m playing Stellaris (I’m always playing Stellaris…) and Vampire The Masquerade: Swansong.
Guild Wars 2 is popping off right now. I’ve been having a lot of fun with the new expansion. They’ve done such a good job of bringing their world to life.
What advice would you give to designers who are looking to work on multiplayer games in particular?
Having a clear understanding of what your mechanics, gameplay systems, levels, incentives, etc. is vital to a healthy multiplayer game.
If something you didn’t intend to be the norm sits around for too long, it will become normalized, and it’ll be a lot harder to change.
My biggest piece of advice is: Play your game with multiple players. It seems obvious, but you can get so much good feedback and ideas from playtests with big groups of people. In Dauntless, whenever we have big releases, we’ll pull together multiple teams to playtest the new content together every day. There’s a lot that’s made clear very quickly.
Thanks to Anthony, Eddie, Teresa and Jordan for their participation!
Our longer term goal is to share deeper insights with interested parties about how and why we operate the way we do. If you have feedback, let us know.