PERSONAL

7

2,378

Week Off's Over

4 years ago - Edited 4 years ago1,477 words

I took the last week and a half or so off working on Atonal Dreams, and did a bunch of other creative stuff instead, like modelling a human figure I hope to give dynamic, customisable features to! It's a big improvement over my last attempt at a dynamic 3D model back in 2017. I'm getting back to Atonal Dreams today, though.

It felt strange not writing a weekly update post over the weekend, but it's not as if I had anything to say! I've also been really busy; I've been more consistently occupied with making things these past few days than I have for the past few months with Divine/Atonal Dreams work. I think a big part of it was because I knew my time was limited, that I had to stop what I was doing and get back to Atonal Dreams this week, so I had to make my time count.

I've been wondering whether to talk about what I've been doing, but one of the main things about it was that I wanted to make something that wouldn't be tied to others' judgement or making money. Something I could really get lost in and enjoy without wondering whether people other than me might like it or what they think should be done differently. It wasn't completely worthlessly selfish, though, as I feel I refined some skills and learned some new tricks that I'll be using in other projects that are for public consumption. I also liked the thought of just indulging some creative impulse without the thought that I'd have to stick with it to completion. Perhaps doing that is the best way to cope with these sparks of inspiration, rather than giving up my current project to work on a new one as I did so many times in the past.

I do have an urge to share despite the risk of being shot down though. So I'll do so vaguely: one of the things I've been doing is exploring a game idea that came to me which would be heavily dependent on procedural generation, which a whole lot of indie developers build their projects around as it allows for experiences you can 'just play' without the need to plot out a narrative. What I have in mind would be the kind of experience that'd be different every time, but which would be more in line with the kinds of things I make and like than just some exploration of randomly-generated dungeon architecture or something.

I'd love to write out a full pitch like I used to do with new game ideas back in the Flash days, but it's not as if I'm going to be working on it if I have Atonal Dreams in the works. Perhaps I'll just play with it a bit in my free time in the background, and if it seems feasible, maybe I could make it after Atonal Dreams. It's structured such that it could be developed fairly quickly, but I think that about everything and the reality always turns out quite different.



I've been focusing on improving my 3D modelling and animation skills, as they're the things I most recently had to do - and still do need to do - on Atonal Dreams, and I was getting somewhat frustrated by some rough patches in my knowledge and technique. In particular, I wanted to create a dynamic human model, which could have its appearance altered by genetic factors such that every individual looked different, allowing for randomly-generated people. You know, like the kind of character creators a ton of games have these days. They're nothing new at this point!

I made some in 2D back when I made Yden a few years ago (and then again in two subsequent versions with improvements each time), but I've only really experimented with a 3D one once, when I was just starting to familiarise myself with Blender and Unity. So it was fairly terrible and I didn't finish it. I just looked it up, and apparently I made it in 2017, and it looks like this:



I suppose terrible isn't a fair assessment of it, but it's fairly clear (to me at least) that I was still getting familiar with 3D. I think I'd only been using Blender for a few months at most at this point (though I was drawing on years of anatomical studies I'd done for drawing so I at least knew how human bodies worked). It was rigged using some plugin called Rigify, which I loathe these days (it creates so much clutter; now I prefer to make my own rigs and to keep everything minimal, neat, and tidy).

Here's the person I made over the weekend:





I'd say it's a bit of a step up! With modelling, it's important to keep good topology - that is, the flow of the rectangles, or 'quads' - and the colours here mark loops of quads to make it clear where consideration has been given to this. I used a lot of images like this when learning, and it's interesting knowing I've got to the point where I can make one of my own without it being an imitation of a reference.

Here's an animated turntable from a deformation test from before I added hair:



The point of the silly pose is to make sure the limbs all deform as they should when the skeleton underneath is moved, which is something you set up by assigning numbers to individual vertices. It's fiddly work and requires a lot of correction. (Vertex weights can typically be auto-assigned, though I assign all of mine manually by typing in numbers. This is one of the reasons I like models with few vertices!)

I was going to post this on Twitter - since I already posted the head, which I started with - but wrote out a tweet three times before ultimately deciding not to. Largely it's because I was wondering how non-artists perceive something like this. Do their thoughts turn to sex? Would they notice the dark edge loops around the crotch and assume nudity? Does the sticking-out-tongue produce an uncomfortable emotional response? Would they assume I'm a pervert in some way?!?

It made me wonder what it must be like being in a field which routinely deals with corpses, autopsies, etc. If you've literally sliced open bodies and taken out their entrails, would you ever be turned on by nude photos? Or do things like that fundamentally shift your perceptions about bodies? I feel like I've certainly had to see bodies differently in order to be able to create depictions of them like this, but obviously most people don't have that experience.

Anyway, I wonder too much about what goes on in other minds, and worry too much about being judged!

I thought the (obviously stylised) proportions of this model seemed okay, though it was annoying importing it into Unity and getting a different impression. I think this looks weirdly emaciated, especially around the shoulders:



Though I wonder if the lack of hair is largely responsible for that? Lighting also seemed to make a big difference:





I started adding some genetic variation:



Yes.

Here's what I had at the end of last night, after spending a couple of days on this:



There's variation in the eyes, but there are a lot of small people here so it's probably not easy to see. No variation in other facial features yet though, as I ran out of time and energy. It's something I'd like to return to as a hobby between work stuff.

It's annoying that I never feel entirely satisfied with the model, though. I keep wanting to tweak bits, but in order to implement some types of genetic variation, I'll need to apply the mirror modifier and use shape keys, which I'm reluctant to do... Technical stuff.

These would be used in the game idea I've had but not written about, but even if I never get around to that, I feel like I learned a lot just making this! It's also nice to check every so often how far I've come in terms of skills. 2017 seems like forever ago, but objectively it was surprisingly recent.

I've spent longer than I intended to writing this, so I should get to work! I can't say I'm brimming with enthusiasm - writing this has probably been procrastination, honestly - but I barely slept last night for factors outside my control that I won't bother rambling about, so I'm tired, and other excuses... Eh, I probably just need to open the game up and play around with it a bit to renew my interest in it!