Cellosseum
Cellosseum (internally known as Project Nova) was my third project with WolverineSoft Studio. It shipped in 2023 on PC on itch.io (August, 2023) and Steam (September, 2023).
The game itself is a case study of the game Nova Drift. Nova Drift is a twin-stick shooter with tank controls where the goal is to survive an unlimited number of progressively difficult waves. The player continuously earn both points towards a high score as well as new upgrades & modifications for their ship. The upgrade pool in the game is randomly generated each run which serves to provide high replay value as well as a large variety of playstyles.
I was the senior game designer on the project with a team of over 20 developers where my role included both systems design and studio leadership.
My personal design work included things like player leveling as well as most of the game’s upgrades (and documenting those designs in Confluence).
Other aspects of my role as the senior designer included frequent communication with studio leadership in regards to the design department as well as leading other designers towards a cohesive vision for the game.
Ship Upgrades
One of the core aspects that we were trying to accomplish as a design team was replayability. A lot of the design work that I personally handled on the project involved creating upgrades which allowed the player to customize their playstyle.
Unfortunately, the team didn’t have the bandwidth to create the dozens of upgrades that we saw in the original case study, but I was still able to design ~18 different upgrades for the game. While WolverineSoft is a fairly large team as whole, we generally ship games in just 3 short months, which ends up limiting the scope of what we’re capable of executing on.
In designing upgrades, I was looking for inspiration from various places beyond the case study (like Cuphead, for example). I wanted to be able to give the player a sense of empowerment as well as the ability to choose a risk vs reward playstyle where skilled players were able to gain benefits when they put themselves in danger more. This lead me to a variety of upgrades which sought to allow players to have a different experience each time they play the game for quite awhile.
An example of this risk vs. reward style gameplay is an upgrade that I designed that was originally called Enrage. This upgrade’s effects were as follows:
- When the player’s shield is broken, damage is doubled and Fire Rate is increased by 50%
- -15% maximum shields
- Shield only begins cooldown when it breaks
- 20% slower shield cooldown (the time it takes for your shield to begin recharging)
This particular upgrade basically incentivized the player to put themself in danger for the substantial benefit of having a burst of power when they’re closing in towards death (which ends the run). Not only is it something that allows the player to survive a risky situation, but it’s also a lot of fun to play in an “enraged” state. These kinds of thoughts ended up driving a lot of the upgrades I created for the game.
Other risk vs. reward options included a couple of upgrades that boosted the player’s damage and health permanently when their health went below a certain threshold. While this kind of buff requires attention to prevent making the game unbalanced and making the player effectively invulnerable, I felt like they fit the risk vs. reward style that I was seeking with them fairly well.
One of the things that Nova Drift did well was that it gave the player a sense of empowerment as they gained more and more upgrades. So, I designed some upgrade to fit in the lane of making the player feel powerful as well. This included things like Explosive Rounds (projectiles create explosions when they hit enemies), Singularity Rounds (explosions are basically upgraded to black holes that suck enemies towards their center and do damage over time), and Splinter Projectiles (projectiles split into smaller, weaker, homing projectiles when they hit enemies).
Adding upgrades to the weapons like this helped create more variety for the game and gave the player more choices for how they wanted to build their character. It helped to support both pillars of replayability and player empowerment.
Stat Upgrades
Another key aspect of the game systems that I worked on was the stat upgrades. In an effort to extend the limited number of upgrades that we were able to create for the game’s scope, I came up with the idea that every other time the player leveled up, we could give them pure stat boosts, rather than behavior modification upgrades. This not only gave us more mileage from limited elements, but also served as another vehicle for allowing the player to customize their playstyle even more, which added replayability to the game.
Since the stat upgrades were something that came along with the regular ship upgrades, I ended up bundling certain stat categories together to give the player multiple benefits at once. This included things like a Projectile Upgrade (+10% Projectile Damage, +5% Projectile Size, +5% Projectile Velocity) and a Shield Upgrade (Shield Points +15%, Shield Cooldown -10%, Shield Recharge Rate +15%).
Initially, the percentages I chose for each category where somewhat arbitrary. After playtesting and gathering feedback, the percentages changed based on how certain stats ended up feeling far more important than others. For example, one of the stat upgrades initially offered a 15% boost to the player’s fire rate. It turned out that a fire rate increase directly correlated with DPS (damage per second) values, which pretty much made the fire rate upgrade the best overall thing a player could choose. So I ended up reducing this value to a more modest 10% increase to balance the game out a bit more.
Reflecting on the project now, I think the stat upgrades ended up working out well. The regular ship mods do some interesting and fun things, but the stat upgrades ended up leaning purely into make the player stronger overall, so they played nicely into our higher level design goal of making the player feel empowered.
XP and Player Leveling Up
Another part of the design work that I personally handled was the XP curve for the player leveling up. I love working in spreadsheets, so tinkering around with various formulas (and talking with ChatGPT) helped me tune an XP curve that allows the player to level up quickly at the beginning and to slow down as the game goes on (each level requiring around 1.25x more XP than the previous level ended up feeling pretty good). Some games use a curve which doubles the required XP for each level. I decided not to go this route because the doubling gets out of control fairly quickly by requiring too much time between levels early on and I wanted players to be able to generally level up several times early in the game (because the new upgrades feel fun) and less later on (because it helps balance the game out).
On the whole, balancing leveling systems to ensure they were fun and rewarding was an interesting challenge as a designer because the game is endless. This meant it was important to consider what could happen 30+ minutes into a run. It’s also far more difficult to predict how a game will feel strictly through a spreadsheet. So, ultimately playtesting helped to clarify that the 1.25x curve felt appropriate during development in both the early and the late game.
Steam Achievements
Another one of the cool parts that I got the opportunity to handle on Cellosseum was designing most of our Steam achievements. I really enjoy getting achievements and trophies on games that I love, so designing these felt pretty intuitive.
I considered the kinds of things that would generally be fun for players without causing them too much of a headache. This looked like designing things like score-based goals, wave based goals, shooting a certain number of projectiles, and parrying a certain number of projectiles.
Postmortem
Reflecting back on the project and playing the game now, there are a few things that I might change about the way I handled my designs.
The first change I’d make would be to actually slow down the rate at which the player levels up. The XP curve ended up increasing at a rate of 1.25x between levels, but have a bit of space from the project and playing the game now, this feels faster than intended. Due to the way the game scales, it seems to be much easier than I originally intended, which ultimately feels like it detracts from our goal of replayability. It feels like the player is getting upgrades a little bit too often and I wish this would have been a little bit more apparent to me before we launched.
In retrospect, another issue which caused the game to be a bit easier than intended was how powerful the upgrades actually were based on how often the player leveled up. Frequent leveling led to the player being overpowered pretty early in a run. One of our goals was to make the player feel powerful, but when this happens too early in a run, it breaks the difficulty curve (it’s pretty east to wreck bosses and champions with the right upgrades). Ultimately, being powerful feels great as a player but it’s a lot less satisfying when it sucks all the challenge out of the game.
At times when making a game, it’s easy get so close to a project that I tend to miss certain aspects that might be obvious to others. Looking back now, it seems like having more people playtest the game might have been a good solution to some of the changes I might have made.
Lastly, I also wish that I would have designed more achievements. It’s a minor thing, but as I look at the list I created, I feel like I could have found more creative challenges for players to engage in. I think a lot of the achievements I designed felt fairly straightforward, but I personally enjoy doing unusual challenges in games that don’t feel insanely hard or inconvenient and I wish that I would have leaned into this a little bit more to provide more entertainment and replay value to players.
Overall, the game was a lot of fun to design. I had the opportunity not only to collaborate with 4 other designers and another 20+ developers across the studio, but also to create a game in one of my favorite genres, roguelikes. Participating in the project was a huge blessing and I’m glad to have had the opportunity to ship another game with the team at WolverineSoft and to continue learning and growing as a game designer.