Game Design Deep Dive: Dead Star 's capital ship invasions

Game Design Deep Dive is an ongoing Gamasutra series with the goal of shedding light on specific design features or mechanics within a video game, in order to show how seemingly simple, fundamental design decisions aren't really that simple at all.

Check out earlier installments on the the action-based RPG battlesin Undertale , situational awareness and player frustrationin GRIP, and the traffic systemsof Cities: Skylines.

Also, dig into our ever-growing Deep Dive archivefor developer-minded features on everything from rocket jumping in Rocket League to the one-hit kills in Titan Souls.

Who: Tom Ivey

My name’s Tom Ivey and I’m the Lead Designer on Dead Star at Armature Studio. Before coming to Armature, I worked at Retro Studios for 9 years as a designer on various franchises, such as Metroid Prime, Donkey Kong Country Returns , and Mario Kart . My first job in the games industry was way back in the long, long ago, when I worked on Ultima Online at Origin Systems, Inc.

What: Recreating Episode “33” of Battlestar Galactica

Dead Star is a competitive, team-based multiplayer game set in the far reaches of outer space. While the moment-to-moment action is built around the fast-paced combat of a twin stick shooter, we stack strategic elements from modern genres on top of that central core.

In our main game mode, called Conquest, two teams battle for control of a region of space: dog-fighting in debris fields, mining asteroids for precious ore, and capturing and upgrading outposts to establish their supremacy over the region.

In addition to this primary mode, players can earn loot drops which grant them access to our version of a “raid” mode.  In this mode, called “Escape Run”, a squad of players escorts a massive Capital Ship as it warps directly into a series of live player matches, disrupting the normal match mode as a temporary third team. The Capital Ship crew has to defend their ship for five minutes before they warp into the next player match, earning a rare reward for each successful jump they complete before they escape the region or the Capital Ship is destroyed.

Why?:

When we first started talking about Dead Star internally, we were really interested in creating a game that had a heavily asymmetrical element to the competing teams. Central to this idea was an episode of the TV show Battlestar Galactica titled “33." In the episode, the Galactica warps through space with the Cylon fleet hot on its tail. Every time the Galactica comes out of warp, the crew launches fighters and prepares for the Cylon attack, hoping against hope that they’ve finally escaped the Cylon threat.

In the original game pitch, this “small crew against overwhelming odds” concept was the core of the entire game. A small team would fly escort ships around a giant Capital Ship, hidden in a randomized location within a massive, procedurally generated region of space. The other team, called the Hunters, was comprised of three to four times as many players as the Capital Ship team. The Hunters were tasked with heading out from their home base, locating the Capital Ship, and destroying it before it powered up its warp drive and escaped the region.

As a match progressed, the Hunters would capture outposts, giving them new spawn locations, increasing the range of their scanners, and generally making the region of space more dangerous for the Capital Ship crew. The Capital Ship crew, on the other hand, tried to hide the presence of the Capital Ship by taking out sensor outposts, harvesting ore to upgrade the Ship’s defenses, or using a limited “skip drive” to teleport the Capital Ship to new locations within the region at set intervals.

Before we created the full Capital Ship experience, we were playing matches every day in a hybridized test mode we called Conquest. In this mode, we generated a smaller, randomized region of space dotted by outposts that could be captured and controlled. We were testing and balancing ships, implementing key features in the upgrade system and other aspects of the moment-to-moment combat. As soon as we got the first battles online, we were hearing people cheer and call each other out across the office as they pulled off an amazing move, or finished a match just one kill ahead of the other team. The ships were fun and balancing out well. The combat was really hitting the variety of notes we had hoped for. The act of building up outposts and fighting for control of the match in a tightly packed region was a real thrill.

That was the good news.

The bad news is that when we expanded to the larger scope required for the Capital Ship hunt, the concept devolved into a tiresome slog. We quickly identified two primary issues, both related to the “scale” of the game:

Tension often devolved into boredom – Trying to provide a large enough area of space to make the hunt difficult meant that individual players might not encounter another player for a large period of time. Playing multiple matches where so much of the player’s time was spent building up their army, only to have a match end with a single quick and violent skirmish seemed like a letdown. We realized we had been designing the hunt like a Real Time Strategy game, where the player is making multiple types of decisions at once: mining resources, building their bases and units, sending out probing attacks, and dealing with direct skirmishes. But our game was focused at the single-unit level, and at that scale a player wasn’t doing nearly enough in the moment to moment of a match to make the game fun. Certain ships didn’t feel like valid choices - We had built Dead Star ’s ships around the idea of providing many different styles of play, from fast-flying scouts to lumbering frigates bristling with weapons. With a giant playfield to explore, we had a problem of physical scale: frigate and raider class ships took too long to engage in combat, and our relatively quick Time-To-Kill meant they were making long trips too often. Where we originally saw frigates as utility ships that players would switch to when the need arose, it was quickly apparent that many players would want to play frigates as their sole ship class.  Unfortunately, during the “hunt” for the Capital Ship, that choice just wasn’t fun.

Our early playtests showed us a strong core game in the ship-to-ship combat and outpost capturing, but we weren’t getting to that core gameplay quickly enough. As we pushed forward, it seemed like our early design concept – large scale, long buildup, hide and seek -- was conflicting with what the playtests were telling us was the actual fun part of the game. We started to question exactly what type of game it was that we wanted to make.

Why? Part II:

A couple of months into development we stood at a crossroads: we had a really fun game with the systems we already had online, but it wasn’t yet integrating the “giant capital ship versus marauding horde” idea that had formed the original core of the design.

It was around this time that a few of us sat down and took stock of the current game in comparison to the original pitch. We talked about what had first inspired us from that Battlestar Galactica episode.  It wasn’t necessarily the hunt and the long buildup. It was the short, sharp tension right after the ship came out of warp, when the future was uncertain. It was the scattered panic as the scanners filled with incoming enemy ships and the Galactica’s pilots launched their fighters. It was the last breathless moment as a volley of nukes flew in, about to blow the Galactica to pieces.

Would the warp drive recharge in time?

Would the Galactica make it out alive?

And how could we recreate that feeling for our players?

The answer came to us in a flash: instead of having the primary game mode centered on the long strategic buildup and hunt, we’d combine the Capital Ship concept with the Conquest test mode we were already playing. Instead of having the Capital Ship crew as one of the primary teams in the main match, they would warp in as a dynamic 3rd team, interrupting a match in progress and providing the red and blue teams with a new shared goal. Teams could continue their fight for control of the region, or make a tenuous alliance and work together to destroy the Capital Ship before it could recharge its engines and warp out of the match.

It was this idea that really got the ball rolling for us.  It took the “asymmetrical teams” concept of the original pitch, but incorporated it into a novel premise: players effectively running “raids” into our Conquest match mode, disrupting the base experience in a really fun way.

The first step in integrating this new concept was finalizing the details of a mode we had previously only been using for testing. We focused on three primary features for our new Conquest mode that kept the feeling of the original pitch, while fixing the issues that had arisen in playtesting:

Procedurally Generated Maps – It was important for us to keep our maps as generated landscapes instead of static builds. We liked the sense of adventure and strategic variance that generated maps provided. They also kept the game feeling fresh and exciting even after playing hundreds of matches. While we hand-craft the hexagonal sectors we use as building blocks to make sure they’re individually fun, we sub-randomize portions of each sector (such as the type of asteroids, nebulae, and outposts; cargo crate types and positions, etc.) to provide internal variation each match.  We then construct the region map out of a random selection of sectors, positioned and rotated in one of several map patterns that emphasize different match strategies. There’s also a chance that large scale events can occur during a match, like comet storms or pirate invasions, adding another dynamic element to the game.  The Capital Ship invasion would work seamlessly with this system, appearing as just another randomized event, in a dynamically generated sector added onto the existing map. Outpost Capture and Upgrade – Another element we wanted to keep from the original pitch was the concept of strategic landmarks that players were fighting over. We wanted those landmarks to be defensible positions that could be upgraded throughout the match (unlike capture points in many other games). Important landmarks meant that scout fighters had to compete against frigates and raiders for control of the map instead of simply flying away whenever they were in trouble. It also added a fun secondary playstyle to the game in the form of mining and gathering ore for upgrades. In order to address the issues with map “scale”, we solidified Outposts’ positions in the randomized layouts, placing an Outpost at the center of each of the hexagonal sectors that made up the map. This ensured that travel times to “points of interest” remained consistent for slower raiders and frigates, even as the battle expanded to multiple fronts. Multiple Ways To Win, On Known Time Scales - Finally, we provided new win conditions for a match. Instead of destroying the Capital Ship to win, players would fight for control of the region of space, seeking to destroy the enemy’s Home Base. While a Home Base starts the game defended by invulnerable defense shields, a team can destroy it either through a long term strategy of ticket accumulation through capturing outposts and killing enemies (providing a known time scale for the match), or a direct assault against the home base (if they can lower the base’s defenses by capturing all the adjacent outposts).  Providing both a time-limited win condition as well as a “late comeback” strategy meant that the winning team always had to stay on their toes, and the appearance of a Capital Ship in a match could help teams execute a swing in their favor.

With our new Conquest mode solidified, we started working on exactly how the Capital Ship would be integrated into the overall match experience. As we worked over the idea, we wanted to make sure it felt like a big event, but not necessarily interruptive to the main game flow. Because a Capital Ship can warp into any match (provided it’s been playing for a minimal time and another Capital Ship hasn’t jumped in), it was important that we didn’t disrupt the match for players that just wanted to play Conquest mode.

To ensure this balance between excitement at seeing a Capital Ship and frustration with it interrupting an already exciting match, we don’t force players to attack the Capital Ship. While players receive high level rewards for destroying even a single one of the Capital Ship’s turrets, the match doesn’t end if the Capital Ship is destroyed, and, more importantly, a team’s home base can still be destroyed at any moment by the enemy team, even when the Capital Ship is still inside the region.

This also provided an important tactical element for the invading Capital Ship team: there were always conflicting interests for the red and blue teams, preventing them from working in perfect unison.

In fact, you’ll often see one team take advantage of the distraction of the Capital Ship to send a strike squad deep into enemy territory, capturing Outposts that are suddenly not as well defended. Perhaps the Capital Ship is too highly upgraded, or defended by a crew that’s too experienced. If a team figures they can’t make a dent on it during this run, they’ll exploit the distraction to gain a foothold in the main match.

In this way, the Capital Ship experience became yet another element that could add a unique flavor to a particular match. Our map generation system could already add randomized events to a region, like comet storms or pirate invasions, but now we had players creating unique events as they warped a Capital Ship into a match. Sometimes you’d see crews warp in and play a conservative, turtle approach – protecting and upgrading the Capital Ship at all cost -- but we also saw experienced crews warping in and heading deep into enemy territory, taking over red and blue outposts and completely changing the landscape of the match. It’s even possible for a Capital Ship crew to destroy one of the Conquest teams’ home bases, ending the match immediately (and resulting in a free warp for the Capital Ship!).

It was exactly the feeling we had hoped for in that original pitch, played out in a completely different game concept.

Result:

In retrospect, the decision seems obvious – take the most exciting parts of the Battlestar Galactica concept and integrate them as a unique and exciting event inside an overall faster-paced base game.

As we’ve introduced Capital Ship invasions in the live game service, people have responded really well to both the experience of running the invasion as well as being in a match that’s invaded. It’s a unique thrill, and it provides a big moment in a match that can change the whole feeling of the conflict. And while we’ve struggled with finding just the right drop rate to allow players to enjoy running a Capital Ship while also making sure every match isn’t constantly invaded, there’s no doubt that we’ve found something unique with the experience as a whole.

Whether I’m on a Capital Ship crew, checking my scanners for threats as the ship warps into a new region of space, or I’m playing in a Conquest match, desperately fighting for control of the region when the massive ship warps in and opens up a new lane of attack, I still feel the thrill of watching episode “33” of Battlestar Galactica for the first time: engines primed, weapons loaded, ready for the next adventure.