squiggle-duck

[squiggle-duck]

For sprout fight 2025 \owo/

Brain saw clover as a prompt and thought :o oo a giant clover, like an umbrella, like in animal crossing! Except not like animal crossing because the only leaf shaped umbrella it has is not a clover at all but it's been two weeks and I still can't figure out where the image in my head has actually come from BUT, if there's an umbrella then it's probably raining! If it's raining there could be puddles, maybe they could be wearing wellies which maybe fits the rain boots prompt and there could be a frog!because I think frogs and rain are a thing and I had an idea for animating it.

Also opted to use the lotus flower palette for the characters cause I'm a sucker for anything pink and green, and then I thought I could use one of the more monochrome palettes to do a background. I originally thought I'd use the honey bee palette for that but given the shade the puddles turned out in that, I opted for morning glories instead, which I think looks a little more whimsical fairy forest-ish too :P

Extra:

I was so confident the frog would be so easy to add. And then I tried to draw it. I promise I have seen a frog before xD please enjoy my many frog attempts. I started calling the one I went with meatball, got too attached , he is now my idiot son and I had to go with him.

[squiggle-duck]

Finished \owo/

Felt like I was relearning to rig from scratch when I made him but I think I can make better quicker models this way. Literally just found out sprout fight is a thing and it's given me inspo so I think I can put that to the test xD

[squiggle-duck]

A lil goober with a pet floaty orb :3

I think the orb is about 80-90% done, I just need to tweak it's X and Y movement and maybe make some idle animations. I'm not entirely sure how pet objects work in vtuber studio , the tracking is convinced I have an eye stuck to the ceiling so it's a tad hard to test things but I didn't want to go too overboard if attaching it to a character will also impact it's movement?

The actual character though: pretty much all the face/head movement is sorted. I might redo the hat, I think I cut the pieces out wrong, some of the head turns look wonky to me but I couldn't do much to move them better I just haven't figured out how else I could cut it yet xD. I also did the tail :3 very happy with it , his body and face cover most of it but I think the physics are good :3 only thing I'm stuck on, the little extra tail wag near the end of the gif, I'm trying to figure out how to make it play automatically when he smiles. For now it's like a 5 second long animation that can be manually triggered to make the tail physics go extra fast but I feel like when I've seen ppl have these extra bursts of physics it just happens automatically, the one brain cell hasn't cracked it yet but maybe by time I finish these models \owo/

[squiggle-duck]

Me latest experiment \ouo/

I wanted to try making a magical girl with fancy transformations and physics :3 also tried recording it as a MP4 this time so it's hopefully less compressed.

Yapping ↓

I'm ridiculously happy with the little side hair doot :3 Only real regrets with this one:

I forgot to do anything in the art with the boots when the body Y is down so her legs just awkwardly slide down into them when she crouches...oops xD

I only noticed it when trying to record the movements but in a few poses the hair at the back seems disconnected from the head. The amount of times I've made this exact same mistake agdjfk give me fifty years I'll finally get the model art right the first time for this xD

One big improvement on this one though, it's the first model I've tried to optimise. It ended up only having 655 parts.. which still feels like a lot but in comparison, the chibi dragon girl, which also had a transformation effect and alternate appearance, had 1077 pieces,whilst having less toggles and less movement so I feel like I'm at least getting a little better :3

As an extra experiment, once I'd finished rigging the model as I'd planned, I tried messing around with using the deformers the way l2d intended xD when I'd seen other pixel art vtuber models before, it bugged me how the movement usually distorted the pixel look? No hate to the artists/riggers their work is lovely and expressive, it just bugged my pedantic ass that you could tell it was clearly an image being warped (you know, like an l2d model xD), rather than acting like a pixel art animation that stayed on a grid. So when I tried l2d , I almost entirely avoided any of the deformers , and made the movement more like a frame by frame animation, but I've been trying to rig some non pixel art models and I like the extra liveliness and flexibility you can get by using the program as your supposed to, so I wanted to see if could combine the two without hating it. It's very subtle, you might need a magnifying glass xD but I was able to add a bit more tilt to the model and fix issues like the boots. I couldn't push it very far without breaking it as I hadn't set up the art with it in mind and I didn't want to spend hours and hours testing it, but my current hypothesis I want to test on my next model:

If I keep the frame by frame style for specific physics things, but rely more on the deformers for movement, maybe I can satisfy my own pedantry over pixels whilst making an overall more fluid character with hopefully even less layers- which should also help make them quicker to make. \ouo/

Will see how it goes \o-o/ so long as I don't catch the plague again I'll hopefully have this or one of me other experiments done soon :3

[squiggle-duck]

I entered the rincord banner contest. Animated version is on bluesky so it's less compressed, but 300 characters is not enough to ramble my entire thought process which my brain is legally obligated to do ^^

Since I started learning l2d, every time I see something that might have been made in it, I want to try and replicate it. I see a lot of floaty animated picture things in games and music and I thought I could try making something like that \o-o/ Then this banner contest was announced, and the banner size guide doohickey said GIFs were an option so I thought I'd give it a go :3 even if I don't win, mneh \owo/ I got to practice a technique I wanted without having to think up a new character design.... Plus it's 🅱️in 🅱️en🅱️ose, I could just fill it with silly little guys which is always fun :P

Favourite bits:

I'm still figuring out how I like drawing none pixel people, and I struggle a lot with faces and for the life of me I couldn't get the angle I wanted her head to be right, till I figured I'm putting it in l2d anyway, I could just separate the head like a normal vtuber model and just rig it into the position I wanted. Think that mostly worked out, but it was fun how every time I tried zooming out, she's just there. Staring into your soul. Menacingly. Sort of. XD

Blue pikmin are my favourite pikmin, and since Nintendo always bullies them I thought the silly little man could have a silly little rest. ^^

[squiggle-duck]

I've been experimenting with live2d again \o-o/

I saw this tutorial (https://youtu.be/EZPI72EGbnY?si=NM4oB3bxwVVgwabM) for tails. I found it helpful for understanding skinning... Andddd then I wanted to try and recreate the tail movement in it in pixels by ignoring basically every step of the tutorial :D

I'd made 1 model with a tail before this, I think I prefer the new technique, though it takes a longer and I still need to iron a few things out, but for now \owo/ tah dahhh:

The old technique was made of 2 parts, the tip that could wag and the rest that was connected to body x. The tip was key framed directly on the breath parameter so it can't react to physics , but I could also connect it to mouth form so when the character smiled it wagged harder, and when they were sad it wagged less. It took maybe an hour, it was only 10 layers, I hadn't used physics much before then so it's fine but I think I could make it fancier now.

The newer technique is broken into 4 pieces, took 160 layers and a couple hours to rig and tweak the physics. It's still a little rough, though I can't tell if that's a problem with the physics settings, if I should have drawn some curves a little smoother , or if my geriatric computer was just lagging , but as a proof of concept I'm quite chuffed with it :3 I haven't yet figured out a way to make the tail wag softer/harder depending on the mouth form like in the old rigging but I'm assuming it's either a toggle or there's some physics setting I just haven't learned about yet.

On the plus side I think I can use a similar technique for animating other pixel things like long hair. I broke the tail into pieces based on where the rotation deformers would be if i was using skinning. Theoretically, it should be doable to break hair into similar sections so long stuff can move more :P already got a design ready to test that and a few other fancy techniques I've seen I just have to get the model art done \o-o/

[squiggle-duck]

Made my first vtuber model in July, I've been trying to make working pixel models ever since. This is my 4th, and I've finally, got the z axis, to freaking, work \ouo/

Also tried a tiny extra physics thing. It's hard to see on the GIFs but it's at 100% on the left and 0% on the right, but the bones and the clothes can drift a tiny bit with momentum. It's subtle but I think it helps make movements a little smoother whilst still keeping the pixels undistorted :3

Forgot to record a couple extra toggles but \o-o/ the background also got ridiculously compressed but fuck it, it's spookier that way xD

Happy Halloween!