I’ve not posted here for quite some time. Not only have I been busy with university work, but I’ve also been thinking about the best way to progress Celestar. I’ve been working on various archetypes since the last post, though they don’t quite have enough polish to publish yet.


Until now, the plan was to complete everything rules and system wise, before progressing to the main narrative of the setting. As I’ve outlined before, I had good reason for doing so. However, some issues have arisen that suggest there may be a better approach.

One major component of in-development RPGs is playtesting. As rules are created and decisions are made, they are tested to see whether they add or detract from the game’s experience. You don’t really complete rules fully and only then test them. Testing is a gradual process, where you experiment with the most recent content. This allows me to catch any issues that might crop up before they’re too “baked in” to easily remove. Suppose I completed the system entirely. If I then realised I disliked the character archetype system, I’d run into trouble. Even more so if the entire system was based around that. There’s also the possibility that I’ll find underrepresented (or even not-included) parts of the setting I like, but I won’t have an easy way to slot those things in.

Through playtesting, I’ll be able to adapt to what “feels right”. More importantly, I can make sure it appeals to all kinds of players. Without this player input, I’ll struggle with sticking to the principle of appealing to a broad audience.

Ideally, therefore, it’s sensible to establish a very core “baseline” of the setting and rules, and to build out from there. We can still stick to the same principles as before (e.g., keeping things simple and accessible). It could take years to fully formulate a good setting and rules, and I’d rather use Celestar sooner than that.


So the plan for the future is as follows. I’ll start by developing the minimum required to start playtesting Celestar, things like:

  • Four character archetypes, one from each category.
  • Three races and backgrounds foreach of these archetypes, as well as weaponry and gear.
  • Having the basic outline of the setting established.
  • Other minor touch-ups or expansions where needed (provided the detail isn’t too excessive).

The first two points ensure I have at least some default “party composition”. This means that playtesting will be far more representative of a “real-game” experience (as opposed to, say, having a game where all four players are Scoundrels, or one where there’s only one player).

The third point is also critical. Though I could test much of the system with characters and rules alone, this only tests Celestar insofar of it being a rules system. When we explored the Celestar’s “target audience”, there were players who draw most of their value from the narriative and the themes of the setting. Putting too much emphasis on testing the rules shows too much favour to one narrow group of players. This is far from ideal.


I ought to address the ideas brought up in one of my earlier posts, where I said “I should design the game first, and then construct the setting around that.” This was in response to the numerous problems I had encountered when designing Extrayn. At a first glance, it looks like I’m lapsing back into the same habits that made Extrayn fail. A few lines later, however, I said this:

Of course, the setting and the game are dependent on one another – it is very difficult to do one without some idea of the other. So really, what I’m saying, is this: “The setting should be designed to fit the idea of the game I want to run, rather than designing the setting and praying the game will fit.”

I was already aware that designing the rules without delving into the setting would be difficult. Before this post, I was inadvertently ignoring the setting. I was essentially doing the complete opposite of “designing the setting and praying the game will fit“. I was designing the game first, and then praying the setting would fit. Even though Celestar is a game (and so the game aspect is important, especially in the earlier stages), I think having a slightly more even split is ideal.

Leave a Reply