Antoine Fauville

I'll take it from the start. Basically, what we're doing is the initial process of creating a location. First of all, we get a pitch, and we have to create a level design document that goes around it, where we do a bit of research and we detail what the location will be about. And from our side, we also have to do some research like this one, where we try to just find interesting pictures on how to create the layout. It's a bit like what artists do with the image references, but for us, we're looking at the navigation and what cool things can happen here and so on. So what I'm going to show today is: I wanted to take an underground facility and connect it to one of the caves that we have, the new caves, that one big opening cave, and then try to make a connection between those two. Like you could go down into the underground facility and then you have a section that opens up into a cave and so on. So from the modular aspect of it, it's just one section of it that we'll be creating. I won't actually connect it to a cave, but it's just to show you guys the process. We start off with some of the pictures to explore some of the ideas we could bring into this location. And I was thinking here is this open space, and we have maybe some areas where you can go from the underground facility to this thing. Quite like this thing here, where you can see some part of the cave but you still have the underground stuff, that can open up to mix this and this image. So this is what I've been looking at on the side. Then usually we would detail all that stuff into a level design document. We'll also get the mission guide that would tell us what we would precisely be doing in there, but I'll let the other guys explain a bit more about this. In here, what I'm showing you guys is an underground facility. I took this because it's a well-known place that everybody knows about, and so it's quite easy to expand from. What I'll be doing is I'll just show you guys how this place works in general. We have a hub. This is the modular aspect of a location. In landing zones, we create very bespoke layouts, but in the modular aspect, we have to think about a hub section that has multiple connectors that are very similar, on which we create these modules that can snap to it. And then we can use the procedural tool to generate different types of [layouts]. So what I'm going to do is just take out one of these, and then I will have a few sets of white box elements on my side here, and then I will start to construct something. So as you know, in the universe, we have a bunch of sets that have metrics, which is very important for us, that everything just goes with the right metrics. And all these are preset things that we can use to create our location, where we will know that this will be the right metrics, the right space for us to use.

A small note, we do have a few more pieces to build our locations than the five that he just showed there. We have quite a few. Going back to what Jared said in the beginning, I think one of the reasons why people sometimes are maybe not so aware of what level designers actually do is that, to my knowledge, depending on which company you actually work in, a level designer does slightly different things. In certain companies, they are more like scripters. In other companies, they are more block-out people. So generally, here at Cloud Imperium Games, a level designer is a person who is part of the process from the beginning to the very end. He is part of the story when, like Antoine said, we get the kickoff and the narrative. We put together the level design document, which is a fairly extensive document that contains all the information that is required for a location to be considered fully functional. These documents are usually written by the level designer, and then I review them and sign them off. And after that's been signed off, we go into a white box phase, which is what we're seeing here. Part of the level design document is also doing what Antoine did, which is gathering reference images and just getting an idea of what we want to aim for. But then we move into the white box phase, and that is the creative bit. There are, in general, two parts to level design. There's the creative bit, which is blocking out the location, building it. And then there's the slightly more boring or tedious bit, which is the markup. And that is setting up all the volumes, everything that defines atmosphere in a space and determines if it has gravity or whatever it might be. I'm sure everyone at some point went through a doorway in Star Citizen and suddenly started suffocating for some reason. And that's just some level designer who missed our room system, and he gets a big slap.

The same goes also for systems like the elevators, the trains, restricted areas, ATC, landing services. There are lots of things that come afterwards.

There's a lot of markup that we do in locations. Take, for example, the room system and the transit system. There are quite a few components to them, if you put it that way. So many times, the setup can be done completely correctly from our side, and it might break for some other reason. It could be object container streaming or whatever it might be that causes it, or it might be that there's some conflict with the layers, or there's a million things that can go wrong. So I think a lot of people who are outside of level design don't quite realize the complexity of certain things.

I guess everybody has a different background on that. For me, I think it's funny to say, but I think Minecraft really helped to get the sense of space and scale and just what makes sense, what feels cool to build and so on. But then I jumped into editors to try to build games out of these editors, and also other game engines before getting into the industry, just building games and so on. And I focused on level design; building worlds and just thinking about the spatial aspect of things is something that I like.

On my end, when I was young, I played around a bit with Warcraft 3, for example, but that was just more on the side. But when I was playing games like the Gothic or the Risen series, this is when it actually started getting me interested, and I really wanted to do this for a living. Because for level design, you actually see what you're building in the editor in the game, rather than in game design, where you do more theoretical stuff and do a lot of write-ups and things like that. So it was always more appealing to me to build something in the editor and have the freedom and try to abuse some of the assets or play around and just create a location, and also a bit of scripting and all that stuff.

Where'd you get your start, Manuel? What was the thing that made you go, I want to do this?

Well, it's just all the logic of the spatial aspect of it that made it very appealing. And when I realized that I could do that as a job, that's when I was like, oh yeah, I'm doing it.

After I knew what I wanted to do, I actually looked around at how you can get into the industry or into the job. In Germany, they actually have schools for that. At the time, it was just, I think, two. So I just applied there and I took a course in game design and level design. Once I was done, I looked for jobs in the industry and I started working for a 3D browser game and did some fantasy-type, top-down level design. In between, I also did a few years of QA, which was also quite interesting, to see how all the other departments and how testing and everything works. It's quite a different point of view. And at some point, I showed up at Andreas's door.

That kind of goes into the thing where, for us, we could just put a box. It's kind of up to us to communicate that to the artists, like, hey, that would be a storage area. Usually what I would do as well is to have each of these labeled, kind of like what I did with these boxes here, maybe with color coding, and then just explain that. Or I would have a drawing of a layout that we do in Visio or so on, where I would specify that that is a storage area. Then when the concept art comes in, they would put that in, or even just the artists would swap it out. But I thought that for the visual aspect of it, we could just build a shelf.

Even with this, right? We're using something called the Designer Solid tool. We're quite proficient with it. So sometimes it takes longer for us to open a library, find a bookshelf, and put it into a level than to just drag a box, drag three more boxes, extrude, and it's job done. It's just the way it is.

So it looks like a bookshelf at this moment because it's just a couple of things, but it could be a cupboard. It could be a cabinet. It could be a storage locker.

Basically, when we put down stuff in a location, like what Antoine did next to the stairs, where he put something that looks like a pipe, in the end that might turn into a support column. The only thing we have indicated is that we want to have something there. There is a full height of this size. Most likely it will be some kind of full cover. And then when the artists put it together, they can really use their creativity to put whatever they would like in it. If there is something specific that we want, that we really want to have this type of art piece there — we really do want to have a pipe there because when there's combat, someone shoots it and it bursts and gas comes out — then we will maybe put a color on it and say, listen, this has to be a pipe. And then that's a pipe.

I thought about expanding this into a connector about the same size as the one we have over there. And then I realized that basically we have the shaft over here, the elevator shaft, which is a problem. So in this case, for example, if I make it too big, we would have bounding box checks that would ensure that nothing would collide with one or the other, stuff like that. So maybe that new section that I have here could work over here, but maybe it wouldn't work where I'm building it right now.

That's why I mentioned that sometimes it helps if you just have a longer corridor that brings out the whole space a bit further and then allows for more generation options, because otherwise the procedural tool would tell you, hey, it would collide, so it wouldn't generate. And then it tries different combinations or different slots where it could connect this new piece to.

One of the design philosophies that I've been pushing for as well is that we really want to get away from this room, corridor, room, corridor kind of approach to building. That's not really what the world looks like. It's a series of rooms. It's a bit more structured than that, so you will never get that. But at some points in the game, you might use corridors, to what these guys have been saying, to push some content a bit further away. Because otherwise, we will run into bounding box issues with all the items that lie around it, and things will just never be able to generate. But with the procedural tool as well, we have the ability to procedurally construct corridors. So the game, when generating, can build corridors by itself.

What I wanted to say is not necessarily dependent on the procedural stuff, but what Antoine was building before: we sometimes already plan for future spaces where things could connect later on. For the initial release, it would just be cut off; there would be a locked door, and you couldn't go any further. But we already try to plan ahead a bit for how we can expand the space in the future. We already know from the level design documents what we want to build in the Tier 1, Tier 2 stuff, and we already plan out these spaces. Let's say, Area 18, the hospital: we already know where it would go, and place a facade there and something like that. So we already plan for this space and leave that for the future.