Hands-on with Anime-Inspired Rogue-like Game Starlight Re:Volvers

Ever wonder what would happen if you took Sailor Moon, Kamen Rider, and Hades, threw them in a blender, and covered the result with hot neon paint? Well, wonder no longer. The answer is Starlight Re:Volver.

Recently, I was able to spend a few hours with the playtest build of the upcoming coop isometric rogue-lite Starlight Re:Volver. Made by former CAPCOM, Wayforward, and Riot developers, the first thing that will catch anyone's eye is the art style. The game uses an incredibly vibrant color palette, with hot pinks and greens used liberally in both the character designs and environments. It also has an obvious anime aesthetic that's more than skin deep. One character, Effy, has an animated transformation sequence that's an obvious homage to Sailor Moon. Meanwhile, another character, Ren, has a similar Kamen Rider-inspired one. There's a clear love for classic anime and tokusatsu shows that gives the presentation an endearing quality.

But while the visual design is a win all the way around, the gameplay (at least during this playtest) is a bit more hit and miss. To be clear, I'm not exactly a novice when it comes to isometric rogue-likes (I've already put over 140 hours into Hades 2 and the game isn't even out yet). I'm also aware that games like this are tweaked, re-balanced, and improved constantly until the final release. It's more than likely that many of the issues I have will be fixed before it comes out.

Starlight Re:Volver very much follows in the footsteps of Hades and the games like it. You go into a dungeon, defeat all the enemies in a room, and get a reward, usually in the form of three options that improve your character in one way or another. Then you choose your next room based on reward prediction, and repeat the process until you reach a boss. Win or lose, along the way, you gather certain resources that enable a form of meta progression to make subsequent runs easier.

In the case of the playtest version of the game I got to try, there was only one possible dungeon. It had a chain of 25 rooms and then a boss. Among the possible room rewards were money, shops, new equipable special attacks, a way of powering up said special attacks, general stat upgrades, HP healing, and resurrection feathers that allow you to bring back a fallen teammate—which brings us to the thing that sets Starlight Re:Volver apart from other, similar games.

Starlight Re:Volver is a four-player coop game when in the dungeons. Each player chooses one of several possible characters, three of which were available for the playtest. These characters not only look different but play differently as well. Ren has a medium-speed sword and has a quick dash as his dodge. Effy has a slow-but-powerful hammer and has a sprint for her dodge (she rides on top of her hammer). Lastly, Kira uses guns for ranged attacks while her dodge doubles as a small area-of-effect attack. Each of them also has numerous possible special attacks to unlock over the course of a run that make each character feel unique while playing.

While the core of the gameplay is solid, numerous little issues pile up. Firstly, while there was controller support, not all the buttons were properly mapped (or labeled)—and I was unable to map them manually. This means switching back to the mouse and keyboard to do certain actions, sometimes even mid-combat.

Next was an issue of information overload. Throughout a run, I had to keep track of my stats, skills, several different currencies, and dropped items. At dungeon stores, there were often a dozen or more options to sort through (and currently they have no icons to denote what type of power-up they are—permanent or temporary—or even what kind of general effect they have). Worse still, because this is a coop game, I felt as if I was holding everyone else up if I sat there and read through every item description or took a moment to review my build so far.

Also, as a coop game, you are beholden to the whims of the majority. The choice for each next room is voted on by all four players. You may want a stat increase for your build, while another player may need a new/better special attack. One of you is going to feel frustrated when they lose the vote. Repeat this for 25 rooms (which feels too long in and of itself to be honest), and things can end up more than a bit aggravating.

The final point I wanted to touch on is more an observation than a complaint. There were only servers in America for the playtest, while I am located in Japan. (It was clearly stated that other servers worldwide will be available at launch.) This allowed me to play the game as a person with a sub-optimal internet connection. (My ping was constantly over 200.) For a fast-paced roguelike, two to three milliseconds of lag might as well be an eternity. While still playable, I was having to play things very safe, dodging the moment I saw the start of an enemy attack animation. The only real issue I had was the boss, which had more than a few “bullet hell” style attacks that required precise movement and timing, though the boss itself was quite fun regardless.

I should also mention that the promotional material for this game talks about a heavy focus on the social aspect of the game—a hub world between dungeon dives where you can interact with other players in fun ways. For the playtest, the hub was a largely empty map. You could fish, craft a few upgrades, interact with a few objects, and chat with the other players, but there was little else to do beyond this, though it was clear that much more was in the works for the area.

In the end, my time with Starlight Re:Volver made me believe it is a game that will live or die based on its time in early access. With a constant stream of gameplay tweaks, UI improvements, and an attention to player feedback, this game could become something special. The style and core gameplay are there; all that's left is to build upon them and then polish, polish, polish.