A lot of snow has fallen since the last Project Aphelion playtest vlog – for a very good reason. Namely, I encountered a bit of a teeny-tiny flaw in the Faction algorithm that turns otherwise sensible organizations into bloodthirsty maniacs on the warpath against all that lives. Opsie. Here, let me break this down.
Where Did It Go Wrong?
As you read in the previous entries of the playtest (you can always catch up if you haven’t), the Faction generation system is easy, accessible, and provides a lot of detail in a short time. The bug in question sits on the priority table, deciding what the Faction will do. We already saw our protagonists immediately throwing down their Assets against each other in the first Interval, as their priorities called for Resources or Assets others had.
I was initially happy the algorithm creates events and drama from the first moment of running because that’s what it’s for, but when I started running the next iterations, I realized the error of my ways. See, the first Cycle had Factions throwing down over Resources they covet using Projects they had available. More often than not, those Projects targeted other Factions or spoiled their efforts in another way – lowering the Familiarity. In short, they started hating on one another in Cycle 2, even the initially cooperative ones. They quickly started losing Assets and calling for Hirelings, and creating more Factions, which also fell into the vicious circle.
In Month Three of the game, I had double digits Factions called to life precisely because of the growing animosity. Factions started losing Assets and Resources and reacted to it by hiring Contractors and improving their own security measures, quickly shifting towards Warfare. Before I knew it, the Sector died in an all-out war. It was then I realized I might have made a teeny-tiny mistake in the math. C’est la vie, no?
I had to locate the problem’s source, and meanwhile, Anna provided me with tons of crucial feedback from her playtest (check it out too). I figured I might as well start editing the text and eradicate issues like that.
It took me a while, but I found the problem. It actually wasn’t a big issue with the algorithm itself, only how priorities worked. As it was, Factions were interested in following their behaviors and traits and running Projects based on Assets they had available. It sounds functional, but I forgot about one thing – they had no overarching goal.
All Factions want to grow, get better, expand, and improve as the players do – but the algorithm ruling their behavior didn’t reflect that. I now added additional weights into the gears, focusing their attention on acquiring new Assets and improving their rank. I did some calculations, and it seems to be working – Factions commit to being selfish as they should. See, a Faction can only have five Assets for each rating point but starts with far fewer. By putting weight into scoring more Assets and Resources to get them, I shifted their attention from coveting the neighbor’s horse to making their own horse fatter. The game, as a result, became much more stable. The first few Cycles won’t be as exciting now, but it should overall be a massive improvement.
Things To Do
As I’m writing these words, I still have to apply a similar solution to Faction growth itself, which shouldn’t take long. I’ll finish editing the text and come back to you with the next blog; see how long it’ll be before the whole thing explodes again. Boy, aren’t dev blogs fun?