I Can’t Stop Thinking About Throwing Fireballs
# I Can’t Stop Thinking About Throwing Fireballs
I’ve been obsessed with this one idea lately: the difference between getting lost with your mind and getting lost with your body. We were talking VR, rhythm, and spell‑slung tower defense with Robin, and the moment he described the toss of a virtual fireball, I felt that little dev‑brain click. Throwing isn’t a button. It’s a feeling. It’s timing, release, and the tiny fingertip drag you don’t notice until a headset turns it into a binary flip. In VR you don’t just think the motion, you live in it. That’s why the whole conversation stuck with me.
We also wandered into the messier stuff. Reddit’s hot‑and‑cold feedback loop. YouTube’s interest‑based roulette wheel. The creator economy’s gravitational pull on game discovery. And sitting right in the middle of it all is AI. In business, it’s a tool. In creative spaces, it can be a scarlet letter. I’m not a purist. I’m a pragmatist. Tools don’t make taste, but they can buy you time.
## What we talked about
Robin came in through a Reddit post, which already tells you the current reality of indie discovery: you go to where conversations happen, even if half the room is throwing tomatoes. We dug into his project, Rhythmic, a VR rhythm‑meets‑tower‑defense game built around motion and music. He’s using AI for the soundtrack and a mentor‑style voice, plus a few tiny texture assists. Not as a gimmick, but as the scaffolding that lets a solo dev get the vision on its feet. We talked about the Steam disclaimer dance, cultural pushback, and the funny truth that “human‑made music” is now a phrase you say out loud with a straight face.
The core design is a love letter to flow state. Think Beat Saber’s arm‑swinging trance, then funnel it through a spellcasting fantasy where you draw energy and fling it at incoming enemies. He tried a 360 particle field at first, then pivoted to front‑facing lanes because modern inside‑out tracking hates your blind spots. That decision says a lot about VR design. You’re always juggling immersion with what the hardware can reliably see.
Fourth, community as compound interest. A moddable rhythm game can become a platform if you seed it with clear tools and reasons to show off. Beat Saber’s longevity wasn’t just official tracks. It was the culture of making and sharing. If Rhythmic lands a satisfying throw, a readable lane design, and a soundtrack that keeps your shoulders bouncing, modders will do the rest. That’s where I’d invest early time: documentation that doesn’t stink, a clean map format, and a path from “my first chart” to “featured in‑game.”
## My closing thoughts
I’m rooting for projects like this because they chase the feeling first. The industry is weird right now. AAA looks invincible until it isn’t. Social platforms reward novelty over nurture. Audiences trust people more than brands. In all that noise, a small team or a solo dev can still win by crafting something that feels great at human scale and then letting the right humans show it off.
My head keeps going back to that moment of release. In VR, throwing a spell is a bet that your body will accept the lie the game is telling. If the arc lands where your gut says it should, you keep playing. If it breaks, you bounce. That’s also the arc of indie marketing now. Hit the timing, find the right hands, and let gravity do the rest. Miss it, and the clip never even makes the feed.
So yes, I’m still thinking about fireballs. I’m also thinking about inboxes full of unplayed keys, a demo knocking at the door, and the modders who will eventually turn one person’s rhythm into a hundred remixes. That’s the kind of future I want to stream into existence.
Episode Info
Then we stayed on throwing. If you’ve never tuned a VR throw, it looks trivial: release, velocity, arc. In reality it’s a rabbit hole. Real hands peel off an object gradually; triggers are on or off. Push for accuracy and it can feel robotic. Push for feel and people miss. Somewhere between aim assist and natural motion sits the sweet spot. That was the heart of his current iteration loop.
On the business side, Robin went full‑time this year to give the game a real shot. The plan is simple and hard at the same time: keep scope lean, ship a demo, cultivate a mod‑friendly community, and court VR creators who love rhythm games. That last part matters. The old funnel of social posts to wishlists is weaker than it used to be. Even big followings don’t guarantee sales if your audience is mostly devs who are busy building their own stuff. Discovery has shifted to people watching their favorite creators take risks first.
## What stood out / lessons learned
First, the AI conversation is overdue for nuance. In a studio with five specialists, you can “do it all” by hand and let AI live in the ideation layer. Solo, AI can be the difference between a prototype that exists and one that never leaves a notebook. That doesn’t erase craft. It reframes where you spend it. For Rhythmic, the art isn’t a static image. It’s motion. It’s the choreography between sound and gesture. If the goal is embodied fun, the tool that gets you there faster is doing its job.
Second, VR design is humility in motion. The coolest idea loses to the sensor that can’t see behind you. Design for what the headset tracks well and you’ll keep people in flow. That’s not compromise, that’s respect for constraints. The same lesson shows up in the throw problem. You can simulate physics to the fifth decimal, but what players remember is whether their body believed the result. In a rhythm game especially, feel beats fidelity. I like the direction Robin’s taking: smooth early, depth later. Let the first ten minutes teach timing and confidence, then dial up precision for mastery.
Third, marketing reality checks. The discovery stack today looks like this: algorithmic feeds show what a user likes right now, not what they subscribed to yesterday. That means every video is a new fight to earn attention. Creator endorsements act like modern referrals. If the right streamer has fun, their audience gives you permission to exist. If they don’t, your viral ceiling lowers. For VR, the niche is even smaller. You’re fishing in a pond where many viewers don’t own a headset yet. The path forward is sharper positioning and fewer, better bets. Find the creators whose audiences already crave swinging, slashing, and sweating to a beat. Give them something delightful and streamable. Reward them in‑game. Make them part of the world.
