TL;DRClipping is prevented by building clothing with enough polygon density to match the underlying base body, and they are looking into larger cloth simulation and, hopefully, wrinkle maps.
So what we do, since everything that we make is built off, like Josh was saying, a base character that might blend in between 50 other characters, which in this example could be 50 different t-shirts. So in order to prevent that, you really have to take the base male that we use for our game and match whatever assets or clothing that we have on top of them as close as possible one-to-one with what's underneath it, because the male body will always be there or the female body will always be underneath the armor or the clothing or whatever. So you really have to build or create enough polygons or density in your assets to prevent anything from clipping. Because usually when that happens, it's because you don't have enough topology or polygons in your clothing to support what's underneath it. So to really prevent that with modern games now, you can add so many polygons. It's really just a matter of supporting whatever's on top of your base model to follow the underlying structure of your model to prevent that from happening.
And then on top of that, we have a basic way to do it now, but we're looking into a much larger cloth sim. We want to try to expand on that further so it gets into a much more immersive look. We've also been looking into, hoping, crossing our fingers for wrinkle maps. There's a lot of games where you'll see when somebody lifts up their shirt or they twist, the wrinkles in their shirt change completely. Whereas the old gen of games would have the same wrinkles, but when they would bend, it would just twist into this weird shape. So we're looking into getting that as well to push the immersion a little further.
TL;DRClothing doesn't blow apart like ships; wear and dirt are hand-painted masks that blend in gradually as an item is worn, differing by material such as cloth, leather or metal.
So right now our clothes don't blow apart like they would if you shoot a ship, which would be pretty cool. But the way we do wear and dirt for clothing or armor is kind of more of an artistic approach. When I say that I mean we have to hand paint or generate these black and white maps that kind of blend in as the character is wearing their clothing. So when you first start, that map will be turned off completely, so it'll be completely white. And then as you wear it, whatever marks we made with wear and dirt or painted in, it'll blend in slowly over the course of time, ranging from a white where there's no dirt to a black where it's just full dirt, full wear. So the way it wears is we look at what we're making, if it's leather or cloth. We paint that stuff in: how would this actually look if it was fully worn? So we have to go in there and actually hand paint these masks that dictate the level of wear that they would have. So it's kind of a zero to one kind of effect.
Yeah, they're pretty high. If you're in-game looking at any of these heads, it's going to be really hard to tell how low they are, just because they're pretty dense.
And the biggest difference really isn't a visual difference. It's an animation difference. So they have less animations or less blend shapes, but the models themselves should stand up really well next to each other.
Yeah, I've actually been in the process of getting a lot of hair in engine, so we can start to do the tweaks we need to do and get it to where it needs to be. So that process is starting now. It's coming.
TL;DRThe main challenges are learning the pipeline and tech, keeping all character artists' work looking like one universe, and balancing massive open-world customization with the close-up quality needed for Squadron 42.
Making just characters in general? For this game? Well, coming on, learning the pipeline, the tech of where to start and where you finish, and what makes up our characters, since there's the high poly, which as a character artist is kind of a general thing that everyone really knows how to do, but how to implement your clothing that you make, and what the tech that we use can make it different. How do we switch up materials? How do we do color variance? Stuff like that. The challenges are really just learning the tools and getting, I would say, a similarity between all the character artists, and all coming out looking like the characters all belong in the same universe is probably the hardest trick to do, really.
Yeah, the tools might differ slightly between if you use Max or use Maya, but those tools do the same thing. So it is the same process, but the tools can change out. It doesn't matter as long as you're getting the result that's necessary. That's important.
I think the biggest thing, just to go back to the question, the biggest challenge that I see is balancing making a huge open world game where you want to have massive customization, and then doing Squadron as well. Where at Naughty Dog, I would imagine the amount of character heads and stuff was a lot, but the amount of ones that were really going to be close up and in cinematic, close-up ready, was probably a handful, maybe 10, maybe 20. Whereas we have 120, 130, 150, 200, like how many heads are we going to have, right? Nobody knows. We're just going to keep adding stuff to it. But to start with, an average game has this many. We've got a plan though. Definitely. We've got a plan. It's just that's the challenge of making a game that's pushing it both on the tech and the gameplay.
TL;DRA shirt took about five days for the show filming and could take about that long unrushed, though reviews, other tasks and the materials library affect the time.
Well, when we did that shirt for the filming of the ATV show, that one was about, I think, five days. I think I was told on Monday, okay, this needs to be done. And you also have to record yourself while you're doing it, which is kind of intimidating. It's kind of like writing in my diary and people are looking at what I'm writing. It's like, oh god, I better not mess up. But yeah, for a shirt you could spend more days depending on how much you want to polish your asset, because for a t-shirt a lot of the detail comes from the high poly sculpt and everything else is just kind of material, which you can always change up on the fly, especially with our system. But if you're not rushing it, maybe five days to do a t-shirt from start to finish.
Yeah. And it always kind of takes longer, too, because everything you do has to get reviewed. So if no one's reviewing your stuff and you're just going from beginning to end, it would be a lot faster. But since we have to work as a team and get stuff to work well, it takes a little longer.
The other thing is there will always be things that come up while you're creating an asset, where it's, oh, it'll take five days, but for Thursday, I had to do the other thing, and Wednesday, this popped up. There's always little things that require attention.
TL;DRThe speakers name favorites: Shane's first jumpsuit, unreleased work they can't discuss, and past projects like Groot and the Avengers Iron Man suit, with no specifics on upcoming Star Citizen work.
Well, I'm working on something pretty cool right now that's not out yet, but I think that's going to be my new favorite thing. But I've been doing a lot of t-shirts and stuff like that. The first thing I did, which is a jumpsuit that a lot of the characters wear, which I've been seeing in a lot of our videos, was probably my favorite thing, because it's the first thing I've created here. Also something I've actually seen in game. So that's pretty cool. Every time I make something, I'll be like, oh, this is my new favorite thing.
I am currently working on my first asset, and I can't talk about it. I'm enjoying it. It's been fun working with our concept artist, Jeremiah. At Naughty Dog it was probably any time, because I led the multiplayer team there for characters, and we had a few opportunities just to make up our own stuff, and that was probably my favorite things to work on, just because it's, all right, here's a loose idea, and just sculpt and make something and then carry it through the whole production and it ends up in game. Stuff like that is really fun.
Favorite thing here is I was just doing some concept art for some alien races, which I cannot talk about. I think it's going really well. It is really cool. Can I even say what the races are? What races do we know?
My favorite thing aside from that would probably be, it's a toss-up between Groot for Guardians 1 and Guardians 2, because I consider them kind of the same thing, but they're not. And then I think the first Iron Man suit I did on Avengers was my favorite. It's still got a special place right here.
That was the Mark 7. When he gets tossed out of the building and he flies down and it goes around him in a circle. The first one that really goes on to him without him having to step into a suit-up sequence.
I've been doing a lot of work with the concept art company One Pixel Brush for the past few years. I've been around in LA for the past few years. Before that, I worked as a matte painter for a little while at The Mill, was in VFX, some other freelance work. But yeah, sunny SoCal, I've been here for a little while and I like it quite a bit.