Benjamin Fremiot

Well, how can we generate a lot of environments quickly? For example, the first mandate we checked was for the space station. So we want to create a lot of them. But if it takes you one month to do a space station because you assemble the different rooms manually, it will take years and years to do all the space stations for all the star systems. So what we are doing: we already have a tool, the procedural layout tool, that is a way to stitch those rooms together. You give some rules, you want the kitchen next to the bedroom, something like that, and it can generate different layouts. But the problem with that at some point was that it didn't give enough control to the user to define exactly what you wanted. And so we are designing a new way to do this procedural approach so that we can assemble those bricks more quickly and easily, and so that every artist can say, okay, I want that there and specifically that there, giving different rules, and so that we can generate a lot of them that are interesting and beautiful and fit with the fantasy of the game. And so for that we are integrating Houdini, which is a software used for procedural, into the editor, so that it can communicate between the two: the editor you normally use to create the levels, and this tool that is made for using those Lego bricks and making rules to assemble them. So it's mostly how can we do that.

One super interesting aspect to what the guys are doing is, in other companies, when you have the pressure of being on a two-year project or a three-year project, there's a lot of pressure to do tech and tools super quickly. And developing this type of idea that the guys are putting together, which essentially took a good, and correct me if I'm wrong, six to eight months of actually really thinking about it, looking at the architecture of the project and seeing how you guys can implement this, is not usually a luxury that we have on a project. So having CIG give the team the time to develop and think about a solution that can scale throughout the years was a godsend, essentially.

You've got to remember that one of the missions for the studio is to be able to create star systems. You can't create a star system by hand. The mission that Benjamin, Guillaume, and the rest of the procedural team have is basically to assist all these nice artists and level designers so that they can make more of the good stuff, in a way that keeps gameplay interesting and fun. And one of the things, you guys are a bit modest in your approach, is that it's also in the small things. Yeah, you can stitch a space station, and it's really great, but it's also in the small things. Let's say a small cable run or placing objects and stuff like that, and the ability to make that procedural so that every step is faster.

Yeah, because we think big at first with procedural, like planets and all that. But when we see our environment artists work, like if you see the books behind me, if you want to place them by hand in a level, you will have to place them one by one. You lose a lot of time, when you can say, okay, fill this shelf with books. And that's the type of tool we can do also with procedural. So we try to tackle the big stuff and also the small stuff to help them fill the world with interesting stuff.

So that Jason and Rainer don't have to spend all day placing Lego bricks.

Yeah, we can focus on the important stuff.

Yeah, placing books in a bookshelf is not always the funnest part of the day. So any procedural tools to speed things up and get into the more important stuff. And as I said earlier, the scope of this game is so massive. It's tools that we need to help us create content at a quicker pace.