Categories
Blog Blog Post Uncategorized

Atavism Dev Log 8 – Tribal Warrior Model

Week 8: February 18-25

After all of my midterms passed and the dust began to settle, I finally got back to work on the weekend and spent a good long time getting this model to a satisfying place. Oh, and sorry this is two days late, but I wanted something significant to show. I’ll never post later than this on a given week.

Character Models Take Time, Learning Takes Even More

tribesmanKnife
You can see here his body proportions are a little different than the sketch, but it provided a good guide in any case. The textured, unlit view looks pretty nice as well, providing a sense of character that white-modeling doesn’t convey.

If there’s one thing I learned this weekend it’s the above. The concept was relatively simple, but the whole process of character modeling from the beginning to end took way longer than I thought it would, and I’ll kind of explain why along with other stuff I learned.

The first is that I thought I could get by without a pair of full sketches. When I started on him I had a vague idea of how he’d look but went ahead with just a human anatomy reference, despite the fact that I’d made character sketches before for previous projects. This wasted a lot of time as I made a character model that just wasn’t the right proportion for what I had been thinking of.

The second problem was that my sketches ended up being too vague. I figured I would model the body as a template and then just add clothes onto it, which is a good idea in theory, but ultimately not necessary and made me keep coming back and drawing different types of clothing for him, even since last week. His design now is barely like I originally thought it would be, besides the mask, which is the only concrete thing I drew in my initial sketch – go figure. I had to keep remodeling things as I figured out how I wanted him to look, tearing off and sewing up topology that should have just been on paper in the first place.

The third thing wasn’t so much a mistake, but something that really helped when I started using it – color. I figured since the silhouette was important and I don’t plan on texturing things anyway – color could just wait until the end. After watching some videos of other people doing low-poly character modelling, I tried assigning materials myself and I was amazed at how my perspective changed. Not only does this help me discover and refine the color palette that I’m going for, but suddenly the pants looked like pants, shapes were easier to eyeball, and even bodypart proportions seemed more intuitive to make when they were skin colored.

Scattered Reflections

heInGame
He kind of looks a bit dry in-game, but lighting and shading has yet to be done. His pelt moves!

I’ve always been afraid of rigs, but I was intent on making a good one and adding cloth physics into the mix as well. Luckily for me, both of these things were easy, and while there’s some amount of constraint work and IK left to implement, the basic human and hand rigs were quick and painless, and Unity’s cloth physics – despite knowing nothing about them – were easy to figure out and implement even when coupled with animation.

Here is a preview of the wonky cloth-physics test on the fur pelt:  https://webmshare.com/play/BjQ51

How much of a placeholder is this model? I’ll probably change or add a few details here or there but I’m satisfied with the design and general form.

What other characters are you going to make in the immediate future? After how long that took, it’s unlikely I’ll make a wildly different character model before Demo Day

How long will it take to animate this guy? Man I don’t even know but I’m aiming to have a basic set of them done by maybe March 13th. Could be faster, could be longer, don’t take my word for it, but after that I’ll have to buckle down on AI to get that done by early April.

One thought I’ve had is that I really ought to experiment with shaders to get a good sheen on this game. In the top picture of this blog he’s looking pretty slick with the unlit textures, but in-game it comes out all drab and dim. I still don’t really want to texture things fully, but just a well-made cell-shading look like Breath of the Wild could really enhance the colors and general look. I’ll have to sit down and experiment with that some week, but it probably won’t be before Demo Day, depending on my pace of work.

Next Week

I’ll be polishing up the rig a bit before jumping into the animation, just so it’s easier to do. After that I’ll plan what AI behaviors and animations are critical for the Demo Day and get to work on those, of which I’ll have a few to show next week. I plan on making a lot of the animations before importing him into Unity, just to experiment with the workflow a little, so we’ll see how much I have to show, but I can always talk about music again if the blog comes up lacking.

So yeah, this is hopefully a fairly substantial post in comparison to last week’s. I feel like I’ve learned a lot in the character-modeling process and this has been good for reflecting on it all and thinking about it for later.

Cheers.

By Steven Zwahl

I'm a Computer Science undergrad at Kansas State University.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s