(Don’t bother clicking this. The larger variant is at the very end of this post.)
As I’ve mentioned, I’ve been dreading the Material Step of this little project of mine. Now that I’ve dived in, my suspicions have proven correct and then some.
For those who haven’t already dug deep into these free assets, it’s obvious that who – be it one or many – ever built these material assets were in a rush, and never had (or gave) the time to give these materials a clean up pass. I’ve come across full meals of spaghetti not hooked to anything. Leftovers from a rush in iteration, I presume. Yet that is only a side dish of horrors here, as it is the methods used that sum up these textures which has caused me the greatest ache.
Conventionally, there is a diffuse map, a roughness/spec, and a normal map. Had this been done for these assets, I would have had no issue just using whatever set textures as is and make my own material to cut out the middle man. You can probably guess then that is not what was done with these assets. These assets are all Channel Packed Masks & Material Functions. Did I mention that some of the Material Functions have Material Functions with Material Functions in them? Material Func-ception! This has resulted in figuring out what effects what, and how, to be a royal pain in the butt. I can change the shader model easy, but finding where the inputs are for the opacity/mask is a hunt. And my Ax! I’m not even metioning all the hackery I’ve done to make this work at all. Material Functions are super useful (just ask @Luos_83), but when you’re trying to add and alter functionality after the fact – eesh .
I, of course, knew that when I proposed my evolve idea in the first place. That’s why my two initial characters above were all purple/pinkish/light-ish red. A proof of concept for myself to see that I could dig into the grit of these characters collection of shaders and alter them. I knew it was going to be rough, but I wanted to give it a shot all the same. This overall idea is really cool, so I wanted to give it a shot – difficulty be damned.
What I didn’t know though was how differently each character, including skins, were in method. It looks to me in the small sample size here that I’ve been working with that each character and each varent materially is its own monster. Its own hydra of make. As a result, the method that I presently have working mostly on one character works inversely as well on the other. I’ve done what I could to make it work the best I can for the moment. I’ll have to figure out if I want to use the time to try and fix this, use it as is (which I’m heavily weighing towards at the moment), or just alter the deal (and the animation) of my concept here or even the material approach entirely.
I finish this update with the most perplexing observation out of all of this; Why can’t you use instanced material variables in Sequencer?!? Why is a Material Parameter Collection the only thing the works “Out-of-the-Box” in Sequencer?!? I burned a solid hour or two in research trying to figure out if I was missing something, as this just doesn’t make sense to me at all!!???
Up next, I’m going to take my temp smoke flipbook (my first ever!) and see my options are. Once I understand that… it’s simulation time. I get to smash my head into a Blender! 
…er… something like that.
This Concludes Update 3
Bonus Update!: Boy, it would be really helpful of me if I was more apt towards blueprints than I presently am. I tried to make a blueprint, and I just couldn’t find the logic to build it out. It’s a shame, as I could probably have circumvented a lot of the trouble I’ve found through blueprints. One idea I have that I’d like to test, which again -perhaps you can see a pattern here- is WAY past my knowledgeable pay-grade, is a customized “Zone” that would be using something like “Custom Depth” or “Post Processing” to that mesh object that enters the zone. Have it apply a change materially that I could then key in sequencer. Ah well, something for me to study in the future.