Star Citizen Live: Art of Ship Concepts

Basically, I'm one of the art directors at Cloud Imperium Games. Currently, I look after spaceship concepts, so I work with a small team, internal and external. We do a lot of the cool stuff, which then you guys buy and help fund this fabulous project. Then I also look after FPS weapons, ship weapons and ship components at the moment.

Now, you mentioned that small team of concept artists. We've had Sarah McCulloch on the team, we've had Jim Martin on the show before. Other names folks might know, Gavin Rothery, stuff like that. It's a very talented team that you're the focal point for, the head of the... what's the word I'm looking for here?

Well, I'm like a focus. They're like a bunch of wild horses, and I kind of bring them in and take direction from Chris. Also, we still have quite a lot of leeway in terms of what we get to do. We try and basically hit all the notes that Star Citizen needs to. There's a lot of functionality in our ships. It was a lot easier in those early days, I would say, because there was functionality, but it was like, let's just throw in a door, throw in a table, yeah, it looks cool. We didn't quite have the metrics. Now everything has to hit metrics, and especially we are trying to improve the process. Well, it's a continual thing, but when we deliver and hand off the concept, it's as good as it can be within the timeframe. It's never perfect, and I think the artists are always like, God damn it, why isn't this right? And I'm just like, there's a lot of moving parts.

Yeah, but I don't think I've worked on all of them. Some of them have slipped through. It's a big learning process, this, and I didn't really know a lot about spaceships when I first started, funnily enough. You think you do, and then you realize you don't know anything. And so there's been a big learning curve, and I've just been picking up from lots of people. But there were some ships that were obviously done by Chris and concept artists that he worked with. There were a few by Nate that he worked on with concept artists. Because at the time, I think there were too many at one point for just me to handle because I had Squadron as well. So we were splitting.

Just for clarity, when you say Chris, you're referring to Chris Olivia, right?

No, Chris Roberts. I guess it was because there was a lot of feedback that I would see from CR on Shotgun. So I'm not quite sure what the process was back then. I wasn't really privy to their working methodology.

Yeah. Basically, we'll take this, and what should happen is they will have rendered out ships, we'll bring it in as a separate layer, and then we'll start adding on all these effects. So there's all sorts of things here. It's a real kit that I've used over the past. So muzzle flashes, sparks for hits, details for ships. We won't need that today. Lens flares. And the funny thing with this stuff is that in the past, it's not a mistake, in the past I've done it where I've taken a concept artist's image and said, make it like this, and I will have put in all the effects and done the color grading and everything. And so it gets you to that 80, 85 percent mark really fast, but it really pisses them off because they're just like, we've done it. It's like you've given them an unwrapped present. And they're like, well, I can see what it is now. What's the benefit? So that's why I try and stick to crappy stuff like this. Seriously, so that they still have that surprise and that magic of, because they've been working on this stuff for, what, three? I mean, some of the stuff is six months we've spent on some of the large ships. I think two months is probably the fastest we've done something. Even that is two months of solid graft, eight hours a day, looking at the same thing, solving problems, solving mechanics, solving visual problems. It's like they're the baker and then suddenly you go, well, here's your cake, it's done, I've decorated it. You want to give them that payoff where they're just like, okay, yeah, this is looking cool. They get to do all that fundamental groundwork.

Because as concept ships, they have not been implemented in the game yet. They have not been built and modeled out in-game.

It depends on the artist as well. Some artists are super clean. There's no right or wrong, everybody just works differently. Somebody might build the same ship and it'll be one million polygons or tris. Another artist might build it and it'll be 200,000. But the ship team, they're spending time on creating something that works for the engine. All the bevels are correct and the shaders are correct and parallax occlusion maps and lighting. There are so many technical issues.

I'm remembering the rivets on the second generation Freelancer. In concept art, you don't have to worry about efficiency. You just have to worry about it looking good. But the second version of the Freelancer, maybe it was the third, I don't remember, but it had all the rivets along the lines and each rivet was like tens of thousands of tris because it was just created really dirty really fast. And unfortunately, when it came time to implement it in game, they took the rivet from the concept model and just said, oh good, I don't have to make a rivet. And at one point, because of that concept rivet that was in the actual hangar model, the Freelancer had a greater resource footprint than the entire Idris, because of all the duplicated rivets. It was crazy. It's a vastly different workflow creating concept ships versus in-game ships, and there's a reason they're separated.