Sebastian Schroeder

I guess it's more like a content side question than like a tech limitation. Genesis never really limited us in terms of what we can do. It's more like what we do on the content side and making it realistic. Obviously, we've made a really nice progress already towards getting closer to what Earth and nature has to offer. It's always our primary target and goal just because it's familiar and it's a really great reference point. And now we're really looking forward to doing more alien things and adding things that make it feel less familiar and just different, but without going into the uncanny valley kind of thing. So it's a balancing act, but we have a nice baseline now and now we just go from that and make it more and better and alien.

Do you think we'll eventually get to something that's just completely unrecognizable, something far more, I'm thinking like Felucia in Star Wars or something where it's all a bunch of magentas and neon blues and stuff like that, just a completely alien world? Is that something we're working towards?

It's definitely, we've tried this in the past. Actually, Hurston was meant to be like blue trees in the beginning, then we iterated on it and it ended up being more or less like Africa now. But yeah, we'll definitely try that again and again until we get it right, but it has to feel good and feel believable and make for a nice experience to be in the place. But it's definitely on the roadmap to possibly be a thing.

Thankfully, the list for what we need to do for clouds is finally starting to get shorter instead of longer. That's nice. We have the fundamental framework falling into place. But there's obviously still quite a few more things we want to try and do, like improving to shaping, shading, performance as well and make them all more stable. As far as the tiling issues are concerned, that's more of a content thing as opposed to a tech thing, as in we have plans to make tools that will definitely improve that and just produce better-looking cloud formations with less of those issues showing up. Obviously, it would be nice if we could throw unlimited resources at everything, but we are still making a game and we have similar limitations to what we can render on the screen as all other games. It's just that our scale is a bit worse or better, depending on who you ask. That makes it a bit more difficult, but we definitely have plans and know how to tackle those issues a bit better in the future. It's just iterating and iterating and learning how to best make the data to drive the system in the best possible way.

So it's kind of a bit of a toss up really. We use the term vertical slice specifically, and that's something I'll kind of dive into a bit. We've kind of tried to do a bit of everything so that we can see the whole picture and then kind of tweak our design from there and start finishing out each bit. And that means that yes, we could build a planet, but there's no way we could ship it. It's not performant enough. It's not varied enough. It doesn't check all of the boxes that we want. There'd be all sorts of problems with the tech that is working for the difficult bits, but not working for the bits that mean that you can have 30 in a solar system and then 30 in another solar system and so on. One of the reasons we've approached it like this is because there's kind of such a fundamental change in the tech. Like the Genesis and the Planet Tech improvements are not just kind of changes to the output of the existing system. It is a fundamental kind of reshuffle and redistribution of work, which is largely because V4 kind of addressed the needs of planets, although the Planet Tech that we have kind of addressed the needs of planets as they were at 3.0. But nowadays, there's a lot more requirements from that tech, whether it be kind of server meshing things, base building things, harvestables, persistence. So we've approached it from this vertical slice, which means, yes, we can build a planet, but it's not the standards we want. But we're now at that point where we take that. We can say, OK, well, here's what worked. Here's what we need to redesign. Here is what we know is great. We've solved some of the big tech problems. Now we're transitioning to more of a kind of milestone based approach where we're saying, OK, well, now we're going to finish out GPU spawning and now we're going to finish out the orbital look and so on. And we can start kind of taking chunks of tech and finishing it. And I think one of the really important things as well is that we're not aiming to release this as one big floodgates. And suddenly everything is out and boom, you've got Genesis. It's going to be features coming online, type bit by bit. And one of the one of the early bits of tech is actually already in the 4.0 streams that are going out to EPTU right now. The planetary rewrite of how we store and subdivide our planet into patches and so on is already out there in the world and working. And that's kind of one of the benefits of having done this as a vertical slice. It means that we're starting to reap the benefits already. In terms of greybox, Seb, you can probably explain better what you meant by greybox in your sections.

So in our process, we usually have whitebox, greybox, and then final and polish. For vegetation, it doesn't really work that well. Whiteboxes don't really represent vegetation assets super well. So we just make a very rough version of it. Also, we were asked fairly late into the process of making the videos to add those labels. And we might have been a bit sloppy. So the swarms were definitely further along than the jungle, for example. Jungle definitely very much what we consider whitebox. Swarms and redwoods a bit further along, with the swarms being the furthest. And yeah, it's all coming together. And we are working through them. And a lot of what you've seen in whitebox is further along now and might be close to polished and final. And then we just need all systems there, every other system to make it into new planets and then release the thing maybe one day. No, I don't know.

Obviously, for CitizenCon, we made the biomes look their most intense, most dense. We want to see what does it need to look like at its most densest, I guess. But obviously, with the new system and everything having ideal conditions and less ideal conditions, you will have varying distances. It could be over multiple kilometers and just have the forest or have something turn into a forest over a large distance, or it could be over some meters or get clearance in the forest. The system will support all of that. We've just only shown the most dense and the most purest version of each biome at CitizenCon. So that's why they look very dense and not very friendly for landing and everything, but it's going to be very unique and varied everywhere on the planet, so you might have very different experiences depending on where you are.

We did see some areas where there were paths carved through the jungle and the forest and stuff, so it is possible to create paths and stuff through this stuff.

Technically, yeah, but they will also just happen naturally based on what data we put into the system and how fertile the soil is in certain areas, and maybe that was water washing it away and then you have a path cleared out from certain species of vegetation that don't want to grow there, or there's more, I'm not going to say more rocks there, but there could be more rocks that don't have enough vegetation to grow, but stuff like that. And it's all going to be more systemic and also more physically plausible. We won't have rocks everywhere anymore. They will be where they would be deposited based on erosion of hydraulic or thermal nature.

Yeah, I think it's basically the same answer on the last question as well. It needs to be. We need to make it all more plausible and the data will be and then it'll just happen.

We kind of need for the weather, we need the data to be more physically correct so that the weather can match. We can't have it snowy in somewhere that's 20 degrees Celsius. So it comes hand in hand. The weather will be based on physically based stuff. But then Seb has to go and make sure the planets are following them physically based rules. With V5 it'll be automatic with the new planets. But when it comes to the older planets, we'll have to touch them up to make sure they do, but they are plausible. Maybe some bits get more snowy or less snowy as that happens to make sure it all makes sense physically.

All right. Conceptually, obviously we've looked at a lot of papers and weather simulations and how clouds form and move on Earth. And conceptually, they will follow the general planet's rotation from orbit and from a larger picture point of view. They will be deviating from this direction, though, just so you get what you get in nature. Obviously, we might not be able to do the full thing, like running a full weather simulation on the servers might be problematic for performance, to say the least. So we're trying to make it as efficient as possible but also look the best. So they will be able to deviate from the primary direction they are moving in. How far we can push this, we don't fully know yet. We're still exploring what we can do with the tech, how we can improve it a little bit further, and what we can do with the data to make it do more with the same amount, ideally. But yeah, we definitely plan to have varying weather patterns moving all over the planet in believable ways based on the data that the planet is providing.

I mean, ArcCorp does use the planetech. I mean, there's a lot of special cases on ArcCorp. It shaded differently. We kind of abused the system a little bit to get the results we want on ArcCorp, to say the least. It probably will be a similar story. We're not planning on spending a year making bespoke ArcCorp tech. But there is parts of the new system that we're hoping will allow improvements on ArcCorp. Obviously, things about the amount of light that will gather underneath a tree and whether a fern can grow isn't really relevant. But there are many other parts that are relevant with the density, with the way we position things. We have a grid pattern for where we're going to put things, but then we randomize it. We could kind of just remove that randomization and get an actual grid pattern more naturally. And yeah, a lot of these rule sets, they're built for nature. But some of them, again, they're more powerful this time. So we might be able to abuse them again to get something nice out of ArcCorp. But we're not focusing on that at the moment. We're making sure that we can deliver every other planet. And we're hopeful that we will then be able to do something nice with ArcCorp at the end as a result. Because all planets must eventually use this tech. There is no option long term. They all must use it. It's just a question of how much we have to change to make it suitable for ArcCorp at the end. Which we won't fully know yet until we give it a bash.

It'll definitely get weather though.

That's definitely in the cards for sure. I don't know if we want to talk about the whole entity thing. But conceptually, yeah. We'll definitely be able to have them. Right now they're tied to the biome they're in. But in the future we can be more specific and spawn them on cliff sides. Or where the rocks fall down at the base of a cliff. And then you will find them there and stuff like that. That's definitely possible, I think.

I think we want to treat a harvestable like any other asset. If it's a mining rock, we want to spawn it according to parameters that we would spawn a regular rock. We don't care whether it's a here's a game entity that you can mine versus here's a static piece of geo. We want to do them with the same rules. And it'd be the same because you get those revenant trees that you go in. No, revenants are bushes. Then there's the heart of the forest tree or whatever. We want to scatter them just like any other vegetation based on the same data and the same rules. So if you're asking is it going to be more realistic, then I would say yes. I think Seb also alluded to the new kind of entity management spawning-y stuff. I don't think that's going to help the specificity of the locations. But it's definitely going to help the density of these things. And I won't elaborate into that right now because the managed entity system is a very techy, boring subject that you can't cover without a TED Talk.

All right. So since we need to address the whole general V4 data versus V5 data anyway for the planets. And the clouds are just part of that process anyway. They will come out of the same toolset basically. Our aim is to have a default weather behavior for a planet. If you generate a generic planet, it will just create ideally the weather patterns for cold air moving underneath warm air. Pushing it up, causing cumulus clouds to form into cumulonimbus clouds. I hate that word. And everything in between ideally. And then also having warm air moving over cold air. Which will cause the straightest family of clouds to form. And that's our default behavior we want to do for every planet if we need to. And on top of that there will obviously be different things. And then you have a very specific need for a certain weather pattern. Or very different or alien planets with alien clouds. Whatever that means, we don't know yet. And we just try to do that as well. But the idea is that we can cover as many different cloud types as possible with the resources we have. And then vary it based on individual planet needs.

I mean, hurricanes and typhoons are the same thing. Just different names depending on where you are on the planet. It's just like very strong winds. And from orbit you have this giant circling weather front moving around. We might not be able to do it like that from orbit. We'll definitely be able to turn up the knob a little bit on what our weather looks like. Compared to what you've seen at CitizenCon. Obviously needs a lot of work from different departments to make this all a coherent experience. Tornadoes, it's not really a weather cloud thing. They move at a very different rate and the clouds will be able to move. So it would be more on VFX and graphics. I don't want to commit them to do anything that they might not be able to support anytime soon. But yeah, definitely we will have certain planets with cyclones on them. If they will move the way they do in nature, we need to see how far we can push the system. Or if they will be a little bit less. But yeah, I don't know yet. We'll see. We're still learning. We're still working with the data and the tools. Expanding them, exploring them, making better data. You can expect a lot of iteration for the weather over the next few months and maybe years.

I mean, we can finally do denser and better vegetation, so maybe we can do less rocks now. But we do love rocks. I know, but kidding aside, with the new local data, it will be, the distribution will be different. It will not be as even anymore. The big rocks will be more out of the way to some extent. And it will be more obvious where they are, and you will most likely be able to drive much further without being constantly thrown around by hitting a rock.

Yeah, it's not like we won't ever have places that do have a lot of rocks, but hopefully the most common scenario might be, I'm driving down a valley, there's a nice clear path down here. But up on the sides, there's like, you know, you can see the big piles of rocks coming down each side, but you can kind of make out a clear path. And it's not like just a random roll of the dice where they're, you know, a rock like immediately in front of you. So we're hoping it feels better, but until we type of finish a planet, drive along it at speed, you know, we'll have to see. But yeah, we're hoping this leads to much more realistic.

It is a decent amount of time investment to make clouds, especially since we're so early in the process. We don't have all the knowledge and all the tools we need or want to get whatever specific visual target. So we do have to spend a fair amount of time to develop those and figure out what we need to do to get the visuals we want. And then the system is still very much subject to change and will change dramatically over time. So we don't want to spend too much time making things that are going to have to be reworked a lot of times. So we're definitely working on rolling out more clouds to more planets. As long as the system is so much in development, we're not going to just do them for all of them. We're not going to do generic ones for all of them. We want them to feel different and portray the planet in a way that makes sense for the planet and just enhances the atmosphere or the vibe of the planet, if you know what I mean. So that's pretty much the thing. We want to do each planet justice and not just have something generic everywhere. But to do that, we need to spend our time on it and not throw away all the work we do. So it's an iterative process that takes a bit of time, if that makes sense.