TBC Log 7/2/2024

I don’t know why, if its sudden influx of summer activities, or if its the government falling into a bajillion pieces (or maybe consolidating as one big dangerous piece), but I’ve been very tired this month and have struggled somewhat to stay motivated. I keep myself up and running, though, in order to keep turning those github tracking boxes green.

Its interesting, when I was doing my prototypes in college I was usually able to capture the “vibe” relatively quickly. But with TBC, a lot of work on it has been game designing towards a specific vibe instead of instantly having it. I have been showing TBC to people, and the biggest feedback now, after integrating the whisper system and the basic structure of the quest system, has been “what ties it all together?” And the true answer to “what ties it all together” is something very difficult to describe when its outside my head. I guess I’m shocked that the whisper system doesn’t give enough of a “vibe” that “ties it all together” on its own, but when I reflect on it its true, there does need to be another level of detail to help the player pick up what I’m putting down. I guess in the game in its current state, whats trying to be connected is what the animals are thinking/wanting and the player’s ability to move them around like dolls. Somewhere that on its own fits in my head. The current version also has the concept of the functionality of “player belief” increasing as the player does more tasks for various creatures.

I’ve wanted the three clans to have asymmetrical goals, outside the player’s goal to have all the creatures believe that they exist. The crocs care about believing in the player the most, the wolves care about accumulating drama, and the pigs care about completing the tech tree and clan relations. This is a little unbalanced, though. But my hope is that after TBC is released, I can make DLC for it that adds other creatures that have their own goals that end up living in the valley.

Currently I’m working on importing spreadsheets into data tables and then reading those data tables and turning them into strings to fill Whispers with. If you don’t know, Whispers are these little boxes that stay on the screen that show what all the animals are thinking either generally or about the creature that’s been selected by the player. There was no way I’d be able to figure out in a decent amount of time how to get the whispers to stay on the screen and not overlap with one another, so I’ve ended up using a plugin to make it happen. So with quests, sometimes a whisper is trying to signify that the whisper’s creature has a quest for the player. I’m still fleshing out what these quests are, but i’ve come up with some general ideas like “bump me into this other creature” or “feed me” or “make me do this action.” Its slowly happening, a design decision was made that all the creatures also have their own personality types, and that such personality types are reflected in the wording of the variofus whispers. So back to those data tables… they’ve really come in procedural clutch.

Oh! I’ve started an LLC for the games that I make, so its easier to track expenses. The current name is “Abby Yaffe LLC” but it is probably going to be changed to “Dark Meadow” in the future. Dark Meadow is this camp that my mother went to when she was a tween/teen. It was a wilderness camp where everyone lived in yurts in a swamp and they cook all their meals and do “wilderness things”. The camp no longer exists these days, but next to my locus in Vermont is a street called “Dark Meadow Drive”. I’ve been thinking about this name, it sort of fits my aesthetic and has familial weight. I can imagine the logo as the words “Dark Meadow” surrounded by like, a bunch of wild flowers. This actually harkens back to my high school art making where I’d dream of putting animated wild flowers around anything, sort of stylistically like Puella Magi Madoka Magica. I was talking with some cousins about the name and determined the vibe to be “dark femme” which I think is the coolest thing. I guess this is a big step in terms of working as a games studio.

I’ve been working with a new 3D artist this month named Pat. We met over twitter after I dm’d them saying I loved their work and wanted to work with them, and it’s worked out so far! I wanted to work with someone who could interpret my vibes but also bring their own direction to the 3D art of the game, which Pat has been doing. A big reference for them with TBC is actually Donkey Kong Country, which I would have never thought of myself but genuinely love now. Below this is from Pat's perspective.

animal kingdom report

setting out to work on the animals, my main reference point was roughly:

1. games with pre-rendered background + spritework foreground characters.

the reference point lodged in my memory is Donkey Kong Country, which makes great use of this technique.

i'll refer to it as a two-layered method, since bg and fg layers were traditionally discrete.

2. a slightly psych-tinged feel, which is something that i feel is intrinsic to the two-layered method.

to me, this is because of the strange and unavoidable delineation between the two layers.

it creates an eerie sense of placelessness (or, conversely, omnipotence) for its subjects (the foreground characters).

for instance:

in DKC, it doesn't matter where the 'light source' is in a map, organically.

the pre-rendered character will always have the same lighting, because that lighting comes from the render that produced its sprite.

so no matter the environment, the developers could only *globally* modify the color value of the sprite. we know Diddy Kong is underwater because there is a water layer drawn atop his sprite.

however:

since we are no longer beholden to the limitations which necessitated this solution, how could we go about reproducing the same 'eerie' effect?

my answer thus far has been attempting to recreate the 'style' of DKC's old CG renders.

studying their features, i have made a few observations:

1. their progenitor seems to be the clay models common in 'game manual' representations of early sprites. i'm thinking in particular of the clay models Squaresoft made for Mana and Mario RPG in particular. as such,

2. their geometry is ridiculously smooth.

this is probably because they were created using NURBS curves. i don't actually know how feasible using NURBS would be *in a game*, because i am scared of Math.

3. their textures are very simple, gratuitously bump-mapped, and were probably made using a mixture of proc-gen and photo texturing. there is little color deviation beyond specular tint.

with all of this in mind:

i have been working on models that look a bit doll-like. i think that this helps them stand out from their relatively photorealistic environment. it is at least my way of taking a stab at recreating the two-layer method in 3D.

the only model i feel is near completion currently is the pig. trying to keep things simple, i am using maps for the base color, normal (bumps), and roughness.

since the geometry of the animals is more or less sorted already, i have been spending a lot of time editing texture maps, and making sure they agree with UE's lighting system (and texture mapping behavior).

a snag i didn't expect, but which forced me to be a bit more intentional with my mapmaking, was that UE does not have an equivalent to blender's 'texture interpolation.'

a texture using "smart" interpolation in blender might appear pixelated or jagged in unreal, because its default is "linear" interpolation.

going forward, this means i have to be a bit neater with normal maps in particular, smoothing them out before importing.

learning as i go.

today, i'm working on a crocodile.

tomorrow, a wolf.