Tethia Asunder was both the largest team I’d ever worked on and the only one so far in which I was not the sole writer. Though I took the lead for much of the story, Tethia Asunder was very much a cowritten project. Because the project did not have a hierarchy, with one person guiding the story’s vision, most story decisions we made were discussed in the group until we reached consensus to avoid a “too many cooks” scenario.

For the most part, this went smoothly; our combined excitement about the project made us eager to collaborate, and nobody on our team was prone to defensiveness or hostility towards the others. Sometimes, though, it was difficult to prioritize, especially when elements needed to be cut for scope or edited down. For instance, the NPC Zul was initially a very minor character. He was optional to interact with and basically existed to provide flavor and depth to the story, like Fauwke or Tubi. However, one of our three writers became very attached to Zul and wanted him to become a larger part of the narrative, including an exploration of his deeper backstory with Tahiri and a more extensive conversation. This necessitated shifting the dynamics of the castle NPCs to center Zul more, replacing Alzenius (a royal advisor), who was previously the key NPC in the scene. In the end, centering Zul worked better than our initial plan with Alzenius and facilitated more natural exposition. But it exemplified the risks of a flat hierarchy of writers: we tend to get very attached, and it’s hard to make sacrifices. (Multiple of my NPCs, including a stoned slug and a morose musician, were cut for scope. Ah well, that’s how it goes sometimes!)

Sauronia is a place very unlike Earth, and we are three writers who love worldbuilding, as our 40-page world bible attests. The story in the game, however, could only be told in dialogue and art; we did not use in-game documents, external narration, or other sources of lore to communicate details about the world. This challenge came on top of the usual issues with dialogue writing: making it sound natural, avoiding “as you and I both know” monologues, creating distinctive character voices, and so forth.

For this reason, information management — how smoothly we communicated key concepts like Ram and Sill (deific moons), the Moongifted (the magic users in our setting), etc. — was a heavy focus when we sent the script to beta readers. The forms we gave them asked them to specifically describe what they thought these concepts were, and based on how well their responses matched what we intended to convey, we were able to tweak and refine our exposition to ensure the right message came across and that they were understanding the story.

By the end of our editing process, there was very little confusion in our readership. Success!

The endless runner genre of game — Subway Surfers, Temple Run, and so forth — usually has incredibly minimal story, just enough to justify why you’re running. As such, we had very little template to work from for Tethia Asunder, and it was also difficult to describe how the runner gameplay and the narrative scenes fit together. In all honesty, I’m not sure they do. The game feels more like a visual novel with random runner sections interspersed than it does a cohesive product. Perhaps part of this is due to the rough-and-ready appearance of the gameplay portions compared to the very polished 2D narrative ones, but I feel the transition from one to the other could be smoother and if I were to repeat this project, I’d keep in closer communication with the technical team to find a better way to do so. In hindsight, much of the narrative work was siloed from the programming, and we could’ve used more cross-team meetings to more deeply understand how our parts fit together rather than simply being two rather disparate halves.

Of course, some of this is a problem of scope, as so many things in games are. If we had time to create more 3D assets or render them more, it would likely feel cohesive. But adding story to an endless runner was always going to be a tall order. I also wish we’d found a way to make the unique memorization mechanic of our gameplay relevant to the story; we could’ve done more thematically with the idea that Tahiri’s weapon is his mind or memory.

The saying goes that a picture is worth a thousand words, but it certainly feels as if a word can be worth a thousand pictures, as well — that is, it’s easy for writers to underestimate how many assets their story will necessitate. Our 2D artist, Cato Emerson, was also one of the three writers and helped to give us a reality check, but even he sometimes overestimated the amount of work it would take to, say, complete a cutscene as we’d originally planned it. This led to a thorough schooling in trimming and reducing the art-relevant aspects of the script down to their bare necessities. For instance, six-panel long storyboards would become two or three; characters who’d initially had multiple expressions made do with just one. Staying aware of how much a story will burden the art team — and what aspects are especially scope-ruining — was an important lesson that I will take to future projects.

Some of the scope issues also appeared after it was too late to change the story, namely the rendering of certain 3D assets. Mercius in the finished game is black; he and the other monsters were meant to be orange, and they are referenced as orange in the script. At the time, we had no idea that this would be out of scope, because the scope shrunk due to complications finding 3D artists and creating/implementing renders. (It wasn’t until very last minute that the 3D animations during the running and fighting sequences worked, due to a longstanding bug in the code.) This is unfortunate; maybe someday we will release an update with correct renders.