10:53 <G-Mobile> UmbralRaptor the distribution of a system, a set of sets

10:54 <G-Mobile> and really subsets

10:55 <G-Mobile> the way I've been dealing with these is to solve for dominant patterns, which is easy for the fighter group, and the caster group, and you can find that pattern in the other four, but it's hard to pick out the subpatterns from the whole

10:56 <G-Mobile> so the archetype of the caster group is long range, squishy, little healing, high DPS, lots of magic

10:57 <G-Mobile> if you bring the large values down, you get the dominant rogue archetype

10:57 <G-Mobile> if you rotate rogue 2 steps clockwise, you get a major cleric subtype

10:58 <G-Mobile> paladin is much more generic potato, and that exists as a subtype in cleric, suggesting potato may be the divine group

10:59 <G-Mobile> barbarian and fighter are close in, low magic, armor tanking cannons with a small amount of healing, mainly self healing, a mutated version of this is present as rogue and ranger subtypes

10:59 <G-Mobile> ranger also has a bit of the divine group, but it's mostly discordant

11:03 <G-Mobile> it's not the distribution of individual numbers that I particularly care about, but some kind of analysis sets

11:11 <G-Mobile> I've got a visualization sitting at home where I've taken the five average values from each class and used them as the quaternion parameters and an offset on a bunch of objects in blender

11:14 <G-Mobile> range of course is the offset. it might be more revealing if I piped in all 17 results, but that's 17 * 5 * 9 things to enter and I'm too lazy to learn python to import this poorly shapen spreadsheet

16:06 <Greys> this but D&D and he's the villain https://youtu.be/BX6hv9a7QBc

16:11 <G-Mobile> "look, we cant send the army. He's just some guy! But we do have to prevent him from doing this."