Randomness in Shmup Game Design — Part 3 of 8
Why I Don't Randomize IS2's Core Weapon
The Interstellar Sentinel 2 demo is out. You can play it right now on Steam: [https://store.steampowered.com/app/3486180/Interstellar_Sentinel_2/](https://store.steampowered.com/app/3486180/Interstellar_Sentinel_2/)
I want to use this post to explain one of the core design decisions behind the combat system — a decision that is easy to miss if you're just playing casually, but that shapes almost everything about how the game feels at higher skill levels.
## The Mechanic: Why You Don't Just Hold the Button
IS2's combat system is built around a mechanic that gives players a reason to *not* hold the fire button — or to switch between firing modes actively rather than defaulting to auto-fire and zoning out.
This kind of mechanic is fairly rare. Most shmup design takes fire as a given: hold button, shoot. The interesting decisions are about positioning and routing around bullets, not about the act of firing itself.
IS2 asks players to think about when and how they fire. There is active engagement in the offensive action, not just the defensive positioning.
The reason I built it this way connects directly to everything I have been writing about in this series.
## What This Mechanic Does for Skill Expression
When your firing behavior is something players can get better at — not just the positioning, not just the routing, but the act of when and how to shoot — you are expanding the skill expression space significantly.
A player who has mastered IS2's combat system is doing something qualitatively different from a player who has not. They are not just surviving better. They are playing a different game. The inputs they are making are more precise, more deliberate, more efficient. And that difference is something *they can feel*.
That is what mastery feels like from the inside: a growing sense that you are doing the same thing with less waste. That you understand something the game is offering and you are meeting it.
This only works if the system is consistent. If the core weapon behavior were random — if the timing windows or the firing mode interactions behaved unpredictably — players could not build that model. They would have no stable target to get better at.
So: no randomness in the core weapon system. The behavior is predictable. The reward for learning it is real.
## Building the Skill Floor
Here is the other side of this: the system needs to be accessible enough that players will engage with it long enough to discover the depth.
There is a version of IS2's mechanic that is tuned so tightly that a new player picks it up, gets punished immediately for not already understanding it, and gives up before they learn anything. That would be bad design. The mechanic needs a forgiving entry point.
The skill floor I was aiming for: a new player can survive by mostly holding the button. They will not be optimal. They will not be discovering the depth. But they will not hit a wall that stops the experience cold.
As they pay attention — as they notice that switching behavior feels good in certain moments, or that conserving fire does something useful — they start climbing naturally. The system has depth waiting for them.
Broad entry, deep ceiling. That is the design target.
## Try It
If you want to feel the mechanic directly, the demo is the fastest way to understand what I am talking about.
[Play the IS2 demo on Steam here](https://store.steampowered.com/app/3486180/Interstellar_Sentinel_2/)
I would genuinely love feedback. If you find yourself naturally starting to engage with the firing mechanic, tell me when that clicked for you. If it feels opaque, tell me that too. That feedback directly shapes how I tune things for launch.
---
*Previous: Post 2 — The GunVein Secret: How Light Randomness Elevates High-Skill Play*
*Next in series: Post 4 — The Skill Expression Stack: What Players Are Actually Learning in Your Game*