Star Citizen: Reverse the Verse LIVE - Building Lorville
An environment artist explains how Lorville was built: scaling the Hurston CBD tower, player habs and their dressing, transit trains, modular tilesets, and time-of-day lighting. They also cover navigation cues, hangars, gates, and future weather.
TL;DRLorville concepts existed from pre-production for some time, with scale and scope drawing on lore team text, and a short lore paragraph inspired Ian Leyland to want to build the location.
So it happened some time ago. It was what we call pre-production; usually that can be visual development through concept art on a very loose design brief. So we've had concepts for areas of Lorville for some time, and we've just been waiting to really get our teeth into it. Before then, obviously, we were scaling up between doing things like outposts or space stations. Grim HEX is in the game, Levski is in the game. The next one we were going to tackle was Lorville. So visual direction wise, it's been around for a while. So when we started that process, there were always a few lines from the lore team that really stood out. Having just even a paragraph, I think even before I started working at CIG, I think there was a paragraph, it must have been from Dave, just one paragraph. And just from reading that, I wanted to make that location. I can't even remember what it was now. It described a backwater landing zone with some dude who worked there, didn't like his job. But something as simple as a few lines of text can inspire so much. So from the lore side of things, we had quite a lot to go on.
We call it the HD, the Hurston showcase main building. Funny enough, one of the things we had to do when making Lorville, we had to really figure out what makes a city work. So we had to do our city planning research, and generally those areas are called CBD, Central Business District. So internally, we call it the HD CBD, a very catchy name.
TL;DRThe lore team called for a dominant presence, so the building was scaled up from a smaller version after back-and-forth with Chris, with executives living above the smog line.
It was a lot smaller. It used to be a lot smaller. So it was called out by the lore team as being this very dominant presence within the location. So it's something that, within the workers' district or for the workers within Lorville, would be this constant presence, this constant overlooking presence. So we needed something that would be dominant on the skyline, but also dominant within the landscape. And I think I covered this a little bit during the panel. I wanted something that was visible from quite some distance away, this monolithic shape that you could see, and then you see it looming over some of these old strip mines. So we knew it wanted to be big. We just didn't necessarily know at the time how big it was going to be. But then if I was going to buy a planet and have a company, my building would probably be the biggest one on the planet; you'd definitely want to see it from space. But we did play around with the scale to see what would feel right. And we scaled it up by a fair amount. But then through some backwards and forwards with Chris, we got to something that would feel really quite impressive, particularly because it was designed so that the executive level or the family level would be right at the top, above the smog line. When you're up there and you look down, it kind of feels right for the space. When you're looking down across the city from above, and then when you're down in the workers' district looking up, I think we found an interesting balance.
Yeah, well, we wanted it so that that pollution level would start to dissipate. It's not like gardens or anything up there, but it is way above the bad stuff down below.
TL;DREnvironment artists generally build internal stories about who lives in a location, which adds depth, though the MMO format limits some dressing; it ranges from large-scale ideas to small details.
Sure. I'd say almost all environment artists do it. Whatever location we build, you build internal monologues and storylines. Who lives here? How would they interact with this space? It just gives way more depth to the location, and you see it. You can play a game and you think, wow, someone's really thought about this. You can really get that depth. It's a little bit different with our game because it's an MMO. We have to be very specific and we can't do certain dressing things, but every artist working on that location would have been really thinking about things, the type of people that work there. What's really cool is when Dave comes to visit, or you have a conversation with Dave or Will or something like that, and you hear them go through it and you hear their perceptions or ideas, and ideas just really start rolling through. That's probably one of the most fun parts of being an environment artist, doing that storytelling, because it's like a creative itch that you can't find anywhere else.
Yeah, and it goes from the large scale to the small scale. You're thinking about what this world is, but you're also thinking about, oh, this guy placed a coffee mug here and left a stain, right? All of that together helps to tell the story of what this place is.
TL;DRThe habs are deliberately lived-in with personality rather than squeaky clean, and Lorville's workers' district suits a disgruntled workforce, with more upmarket, classy locations expected in the future.
It's a valid question. Ultimately, we're trying to create something that has a little bit of depth to it, nothing squeaky clean. It's a lived-in universe. So I think, going back to your 2.0 section, we wanted to give it a bit of personality, because when we made the habs, I'm like, okay, the architecture feels interesting, but I want to see a little bit of personality within there. And with the props that we had at the time, we were like, right, I think we can do something quite interesting. I have the same conversation with Todd Papy quite a lot. And then when we moved on to something specific like Lorville, we're talking about a workers' district for a very disgruntled workforce. So it's not going to be butterflies and rainbows. [There will be] locations in the future where it will be much more upmarket, much more classy. And we're kind of looking forward to doing some interesting things there.
Well, so there's a couple of different variations. There's three variants in terms of actual dressing variation. They're not drastically different. The layout of the room is certainly different. One has the bed in the middle, one has the bed at the back. One has slightly more stuff on the fridge, one's a little bit more messy with some clothes on the ground or whatever. But there's also variants in lighting. There's some that have cooler lighting. There's some that have warmer lighting. It's not this incredibly broad, oh, we have 10 different variations, you're going to spawn in a completely different place, because we only have so much time to work on these things. But you might see a couple of different places when you spawn.
TL;DRThey don't know why the toothpaste looks green, possibly a graphical effect, but the answer shifts to how pickup-and-inspect items mean they're seriously discussing dressing locations consistently so interaction doesn't break immersion.
That's interesting, because the reason I called it out in my presentation is because I had never seen it green before. I think it was just some sort of effect or something going on with the graphics that just made it green. I'd actually like to dig out the prop and see if it's actually green, because if it is, I have a couple of questions about why it's green as well.
Actually, just to briefly talk on dressing, it's actually quite an interesting topic right now, because usually what we do is we go through and we set dress the environment appropriate to the location. But as you can see with a lot of the pick-up and inspect stuff that you're starting to see now, suddenly you're looking at this static dressing within the locations and you're like, wait a minute, why can't I pick that up? Why can't I pick that up? So we're actually talking very seriously about how to dress these locations going forward. Because if I'm at this table here and I'm like, hey, I can pick up this mouse, that's great. But then there's something else on the table, something on the floor, and I can't pick it up, it breaks immersion. We have to be very consistent. Otherwise, it would be quite frustrating for the player. So going forwards, one of the most important things for me is to achieve the same level of storytelling, but where we can have that level of interaction. So going forwards, that's something that's quite important to us internally right now.
Within the visual development stage, we go quite far, because the more we do in that stage, the more you really think about the location when it actually comes time for full production. Obviously, you have to pick and choose how far you go. Ultimately, what we did inside the playable area is try and figure out those slices or those levels between the workers and the guards and the architecture, and then you've got the stuff down below. So hopefully when players are walking around, they'll just look around and start to see those cues.
So the main focus for the initial sprint was obviously on L19. One thing that made this landing zone different, you briefly touched on it, is it wasn't necessarily just a case of making a landing zone. That landing zone is sprinkled with what we call common elements. So there's things that are very specific to that location, and there's things that we can reuse much more of going down [the line]. Then within the lore, they call out many different locations. There were some interesting factory zones and things like that. And then obviously you've got the CBD, which we're pushing on right now.
Sure. It's a very large interior, so how much of that we're going to explore initially would be quite limited. But it gives us a chance to show the flip side of what people would see inside the workers' district. So we're trying out some quite new things in terms of architecture, look and feel, and obviously you've got some shops and reasons to go there. So we're hoping that together, when you walk around the workers' area and you walk around the CBD, you should get a very good idea of the narrative and the imbalance between the executives and the family and then this workforce.
Yeah, it's actually quite a deep conversation. Player navigation, many people have talked about it over many presentations, really interesting stuff. For me, ultimately, it comes down to theme park design. Theme park designers do the exact same thing. When you're in one part of the theme park, you need to look over somewhere else and think, oh, hey, that's over there, or that's over there. Otherwise, you'd just be lost. And we try to employ some similar cues inside something as small as L19. Specifically, when you're in one location, you get the vista shot, and you should get an idea of, hey, if I go that way, I'll probably get to the habs, or if I go that way, I'll probably get to the transit network, for example.
TL;DRThe speakers say seeing the physical train pass through the city sold them on Lorville's believability; Chris insisted on physical trains, which added complexity, and the train connects different areas.
That's the thing, right? You're in the editor, and you're just there, flying the camera around Lorville, and you're like, yeah, cool, that looks nice. And you're inside the transit network, whatever. And then there was this one day, it was quite early on, it was still a little bit white box, but there was this one day where we just saw this camera inside the train just going past. And you just saw this, boom, the HD building there. And at that point, I was like, okay, wow, that's quite exciting. That's when I was really sold on the believability of Lorville as a city.
Chris was very, very specific with regards to the transit: physical trains, physical trains, physical trains. And he was 100% right, because it added a lot of complexity on the teams. But when you see it, you're like, like you say, okay, I understand it now. So it was something that Chris was very, very, very specific with. And you see it in game and you ride on it, it all makes sense.
I think it's also great in the sense of it being a connective element between so many things, not just in the literal sense that it helps you get from one area to another. You talked a lot about the recessed bar that had the windows where you could see the train go through. I think the idea that you could be in multiple places of the city but see the same train go through helps connect the disparate areas. Am I on base here?
No, no, you're right. One thing that was interesting, I don't think many of us had ever done a physical train before, or a physical transit network before. And what was interesting, I think it was the demo playthrough a few hours before the event, as we were taking off and flying off, we just looked down and there was this train just shuttling to the spaceport, and all of us there were just, that's pretty cool.
TL;DRThey aim for designs that are the most advanced yet still relatable, and keep the current look utilitarian to leave room for later art styles; some transit ideas are being tried internally.
Very, very good question. So I studied industrial design, and part of that training to be an industrial designer is to design products and spaces understanding what came before. With trying to design something 900 years in the future, there's this concept called 'most advanced yet relatable.' So Jared, I could design you a train that looks like a space whale, butterflies kind of come out of it when it flies, and I say, that's a train 900 years in the future. It would be too advanced for you to relate to it. It would be so far gone. They use similar techniques in other sci-fi genres and games and things like that. There's only so far you can actually go. So that's one part of it. The second part is we're trying to describe some of the older parts of our universe. The whole timeline of Star Citizen has been mapped out. And throughout that time period, we define different art styles. So what we're seeing right now is all pretty much utilitarianism. And if we go too advanced too early, we haven't really got any kind of wiggle room. So what we're showing is like the oldest thing in the universe. So for us, having quite a relatable train design right now, it felt right for Lorville, felt right for the time period. But that's not to say in the future... we've already got some interesting ideas for transit networks that we're working on internally.
TL;DRArtists rely on an internal reference library built by exposing themselves to real architecture and design, such as the London Underground and Paris metro, so research on a location is critical.
Also, another good question. As artists, we rely massively on our internal reference library. An example would be if I asked you, Jared, to draw a castle on a post-it note, and then I asked someone who'd had more of a classical art training to draw a castle, I'm saying the two would look very different, because they're calling from their internal reference libraries. So basically, what we need to do is constantly explore or expose ourselves to new things, be it architecture or design. So when we're talking about the transit areas, I remember going through the London Underground about six months ago, and I was just taking pictures of walls, ceilings. I was getting some very strange looks from the security. You can actually buy reference photography packs of cables, which we would look at and go, that's a great cable. That's a great cable. Actually doing research on a location is critical. You see other game studios or film production studios, when they do pre-production, they go to the location. They feel what that environment is, or they experience it: how it smells, how it looks, how it sounds, because you need to expose yourself to try and translate that digitally. Sometimes it's slightly harder for us, because we can't go to the moon just yet. So wherever we can, we try and base it on real-world locations or designs that have already been figured out. So just to bring it back to the transit, the London Underground: next time you're on the London Underground, just look up. Look up at the cables.
That's one of the things I did, definitely. Because when I was dressing and making the transit station, I had just gone to Paris and I got to see the Paris metro system. I took a lot of inspiration from there, and there was a lot of grandiosity to the architecture. I would look up and I'd see all these pipes, and everybody would give you looks because you're just there like, oh man, look at how that pipe works. And I like how all these things are working with one another, and you can see the history of this place and how things were made. And those are the things that you're constantly trying to develop as an artist. It's just like Ian said, this visual library, which just comes over time. You start to remember how environments that you've been to looked, and you try to pull from this library in your head and try to construct something believable. That's something I definitely did whilst doing the transit: going to train stations and taking pictures of really random things. And you try to do it in a bit of a subtle way. But yeah, it's quite important to understand what the environment that you're making is.
When I'm traveling with my family, I have two photo albums. One with the family in, one without the family in. [Referring to Jared's castle drawing:] So what you're thinking about there is battlement design, construction with those blocks, drawbridge design. But when you're thinking about that, you're probably recalling back to a film or a castle, but if you've never been to a castle, you would struggle. So that reference library is so crucial.
TL;DRModular tilesets are essential but a complicated relationship; locations aim for roughly 70% modular and 30% unique, using dressing, lighting and new pieces to avoid repetition and keep variety across locations.
It's a bit of both. It's a complicated relationship. You have this set that will help you. We're making a modular game. We're trying to make something that can expand. And you're not going to get that, as I touched on in the panel, by making custom pieces for everything. Otherwise, you'll be here for centuries. So you've got to break the environment down in a modular way, but you also have to create enough variance in these pieces so that you can try to break up that constant... I don't know what the right word is here. But you also use the dressing and the lighting to make those different areas feel different. Though you might be looking at the same wall here and there, the dressing and the lighting are going to help you understand these are different places. I'm sure you have much more to develop on that.
Sure. Ultimately, yeah, Luan's right, we are a modular game. Unless you have lots of tilesets... within a tileset, there's many different components. Until you've really fleshed those out, you're going to see repetition. So because our engine is really good and it's all real time, we can build out locations fairly quickly, straight to metric, and get to that. But then when you start to look at it and you see that repetition, that's when you start combining: okay, this guy's going to be unique, let's do something unique for there. On a location, we run to the rule of 70-30 or something: 70% of that location is going to be modular, 30% is going to be unique, because it doesn't matter how good your modular set is, you're going to see that fatigue. You're going to see that repetition. Probably more important for me is even just outside of a singular location. If I had a two to three hour play experience within Star Citizen, I've gone to this location, I've gone to this location, I've gone to this location, it's very important I'm not going from one modular location to another. We have to find ways to break that up. Some may be more modular than others. Some may be a bit more custom than others. It just depends on what that location is. So it's an absolutely essential part of our pipeline, but we have to try and squeeze in as much time as we can to fit in those unique areas. Otherwise, if the player's saying, oh, that's that same one, that's that same one, it breaks immersion. That's not really what we want.
And that happens at the start of an environment as well, because when you lay out the architecture the first time around, it is all the same wall piece. And the more you look at that, the more you just want to start breaking that up. But that's when you start seeing certain areas. For example, in the transit, when I laid everything out, everything was two and four meter pieces and it looked pretty hideous. And then I saw, this is a chance for me to do something interesting with the wall here. So then I took that and I tried to find different tilesets or different wall pieces that might make that area feel a little bit nicer. And I would even go as far as to create... if there's a 12 meter area and it's all broken up with these lines, and you want to create something more interesting and more bespoke and you have some time, you certainly go in and you try to create a new piece that not only could be used here, but could be used by someone else later down the line elsewhere. But these tilesets are ever-evolving. You always add a little bit here, a little bit there. We create a new location and suddenly we have a few more pieces, or we go back and we touch up a couple of other pieces. It's all about this variation that you just get with time.
And then going back to just the art styles within the game, we've got some great art styles that we really want to get into, but the investment of time into a new building set is quite a lot. So we have to be very measured when we do that process.
TL;DRBecause the sun is fixed, they physically moved the planet to validate times of day, shaped Lorville so workers' areas get little sun, and are discussing having the city change from day to night.
Yeah. With Lorville right now, we had to create or find a way to give it the actual... So, rewind. Traditionally, you have a standard time of day slider. You can change that curve, so you can validate specifically what that cycle is going to be. We can't do that. Our sun is a fixed thing. So we just moved the planet around to get it right. I very specifically wanted sunset to be over the spaceport, looking from the habs. So we had to physically move the planet to validate that time of day, and then we're like, right, okay, how's it looking at two o'clock, how's it looking at one o'clock? So we do these internal time-lapses. I had a video I was going to show at CitizenCon. The presentation was already too long, so I took it out. But it just shows... and usually we put some music to it, because the right time of day will add so much to it. If the sun's at the wrong angle, like you say, it would wash out the scene. It doesn't create the feel. One of the things we very specifically wanted to do with Lorville: we were walking around it, and if the sun was in a workers' area, it just didn't feel right. It's like, why doesn't this feel right? Because if I'm in a sunny location, it has a certain feeling to it. So I was like, let's raise those walls. Let's push that landing zone down. So the feeling is that the sun never... or maybe in some places, but the sun just can't touch. So if you're a worker in L19, you're rarely going to get the sun on your face. And I thought that was going to be quite interesting. So straight from white box, it's something that we're trying to consider, and going forwards, how that city or how that location changes as the time of day changes. When you look at a cityscape as that sun goes down, you see those time-lapses, the city kind of wakes up. As that sun goes down, lights start to come on, magic starts happening. This is something that we're talking about internally: I want, when that sun goes down, the city to kind of change. So imagine I'm a player and I come to Lorville for the first time, and it was night. If I was a player and I came in the daytime, if Luan and I had those conversations, I want that experience to be quite different. And then that feeds into the AI. All sorts of really interesting things start to happen when you start to think about the possibilities of what we can do.
I wanted it to go down. I just wanted that very specific vista. And previously, I'd say, when I was lighting it, I'd go, hey, man, sun, there, boom, boom, done. But then I was talking with Chris, lead lighting artist, and I'm like, Chris, I want that sun there. And he was like, I don't think the sun's going to go there. So then we had to really figure it out. So thinking about day-night cycles on a planetary scale is quite complex. We can't necessarily validate it for an outpost or a modular location. But when we're trying to build something like a major landing zone, I think it's very important to really understand what it feels like at those different times of day, because we're trying to say to the player, hey, spend some time here.
Sunset. It's quite interesting in our game, like Ian was saying. Usually, traditionally, you have a time of day slider, and you can see things at different times. In ours, if you want to see it at a different time of day, you grab the sun and you move it down, which is quite insane when you think about it. But sometimes I'll be working on something in Lorville, and I'll just zoom out, and I'm just looking at the city, and I go, down a bit. And then I lower the field of view, and it's, okay, here we are. Then I put some music on, and then you start to feel the city, and you start to really get inspired, and everything looks better at sunset, really.
I just want it to rain in Area18. That's my biggest dream, because if you think about deep metropolis, neon lights, rain, sunset, I'd be happy that day.
But specifically talking about weather, it's a much larger conversation, and it's not like, hey, just put a rain entity, or make it rain, press the make-it-rain button. It's a much larger conversation, because if I'm approaching Hurston and I see a weather volume and I fly down into there, the weather effects need to be contextual to that location. So it's something that we are looking into, but it's not a small undertaking. But when we do it, we'll do it right.
TL;DRThe hangars were built to a size spec around fitting the largest ship, with the work focused on making them run very efficiently so the ship gets the lighting budget, and performance is still being actively studied.
So these hangars had to run. So when we were developing, figuring out the complexity of the landing zone, we're like, okay, we're going to need a spaceport, we're going to need hangars, figuring out exactly what we need. With the hangars, we design, create a specification: small, medium, large, extra large. And we were thinking, right, how big is the extra large? And I'm like, right, it needs to fit a Reclaimer in. I'm like, right, okay, that's going to be a big hangar. So we obviously did some visual design, and at the end of the day, we had a box to play with. We got a box from levels and, all right, what can we do to make this box interesting and make it run? So a lot of the work that we put into it, not just artistically, was getting something that we could run very efficiently. Without going into the details: just how we authored the meshes, even how we authored the master material and the arrangement of the IDs, so when the engine tried to render it, all the IDs were optimal. So for example, if the painted metal was ID 16, if it was ID 5, we'd get a little bit of a perf saving. So everything about it, we had to make sure that we could hit our modular goals to hit the different hangar sizes. It had to run, and it had to run very fast. So that was critical, because the star of the show is the ship. So we need to make sure that all of that surrounding is focused on the product, shall we say. So most of the lighting budget went on making the ship look good. We tried to make the space feel detailed, to get that scale, but it needs to be cheap so that we can have more budget for the ship. And then what happens is, because it's not necessarily just a closed box, when that door opens, we've got to draw whatever's on the outside of it. And when what's on the outside is Lorville, it just needs to be even more efficient.
That's one of the biggest things when you're making, say, the LODs for the hangar or anything like that. One of the things we did, we would put it under stress test and we'd go all the way out to, say, as far as I think a ship would come in from quantum. And you open those doors and then you have to balance the performance. You have to go, okay, how far away am I seeing this from? How expensive, or rather how cheap, can I make the mesh? What can I do to really save on performance here? I think we've done a pretty good job with the hangars in terms of performance.
We're not getting shouted at. This time. No, to be honest with you, performance is something that we are actively drilling into every day. Understanding exactly how the GPU processes the content we give it is a very deep conversation. And then when you make a modular game, it's very different, because you're giving it loads of individual chunks. So then finding the most efficient way to author those assets, we're still trying to understand the best way to feed that.
TL;DRThe wall and gates were a deliberate balance between looks and playability under a firm deadline, described as a first step into making a city that could be improved, with ideas for the future.
The gates and the wall caused, or just required, a lot of thinking. And could we do more or improve that in the future? Sure. If you look at a city, a city just doesn't end. A city translates into something else. We had a very specific deadline to hit with Lorville, so we had to make some technical and creative decisions to validate that critical path. Because if we took away that wall and just had the city, it would just get so complicated to figure out, okay, where can the player go? Where can't he go? So we had to find that right balance between something that looked good but also played good, because we tried some designs where it was very complex, and then I was talking with Todd about it and he's like, dude, find your way to that gate. Because it was like me being on the outskirts of London and saying, find this area. It's an impossibility. So I would consider it one of our first steps into making a city. We've got plenty of ideas or things that we want to try in the future. Watch this space.
TL;DRBoth say they are very excited to see players experience Lorville, as the community's screenshots and videos inspire them and it's a passion project they put everything into.
Screenshots. Screenshots, YouTube videos, Twitch streams. Go on Reddit and you see... sometimes I'm taking a couple of screenshots around, and then I'll go on Reddit, and there's just this one guy that takes screenshots, and he took the screenshot of the hangar, and I was like, that's one of the best screenshots I've ever seen. The screenshots the community comes up with, it's quite inspiring. You see that, and you see everything come alive, and everybody discussing the things that are happening in that screenshot, or everybody gets quite excited, and that's what gets me really excited about 3.3.5. It's just seeing people taking screenshots around Lorville.
I think internally, there's very specific people that we follow. We can't mention any names, but there's one guy who plays it, and he plays the game how I see it in my head, in cinematic mode. I'm watching up at night. I'm just watching. I'm like, all right, that's cool. The amount of time we've spent, the blood, sweat and tears that have gone into Hurston and Lorville and the moons, I think I can speak for everyone, we're just really looking forward to seeing how it's received and how people play it. Because we put all of our creative energy, we put everything into this project. It's a passion project. So yeah, when we put it out, we get that nervous twitch and we're like, oh, no, but it's always great seeing it, because that for me is one of the coolest parts of working on Star Citizen: we don't develop behind closed doors for years and years and then a publisher puts it out. We work on something, and before too long we put it out, and you're watching someone playing it and you're like, hey man, you found it. You found that spot.
It's like the first few [weeks] when 3.0 went out and these time-lapses started coming out of Yela and Cellin and whatever, and it just completely made you like, yeah, this is why we're doing this. This is what's so exciting to us, and it's quite nice that it's exciting for other people too.