Another game that has amazing clouds, where you move around them horizontally and vertically, is the recently released Ace Combat 8. The game itself is UE5 but clouds seem to be completely custom made, and the in-game clouds look amazing, especially noticeable as you're flying around/through them in the game. Examples: https://x.com/PROJECT_ACES/status/2042437413260443803 - https://www.youtube.com/watch?v=BG7DLLdQf1k
The game is even more amazing in VR where you really get a better spatial understanding of the entire sky, environment and the clouds.
Super pretty indeed, but that AAA look is way out of my league :) Probably took several rendering engineers several months. And it wouldn't fit my art style anyway.
I'd love to play the game when it's ready. I don't play much of anything but was searching for open-world sailing/exploration games very recently on Steam.
Sounds like exactly what I was looking for... Strange coincidence.
There are a couple of ships you can sail with a simple economic engine - you start small and work up. It has some simple sail handling - its not enough to just turn right/left and manage sails up/down. Multi-sail ships are an interesting challenge but manageable.
It's pretty slow-paced, but like I said, I had a lot of fun with it for ~20 hours.
Oh wow, thanks for this. I'm working on a game with some similarities to this (at least in the sense of "includes ocean travel"). Just this week I was working on a sky shader in Godot, and have yet to implement clouds. That whole series of articles looks relevant to my interests... ocean/wave rendering is also on my to-do list.
Game itself sounds interesting and I look forward to playing.
> "On Earth, exploration often meant gruesome exploitation and violence, whose effects can be felt even today... However, this game is not about any of that; it shows how things could have gone, if people had behaved more kindly towards each other."
I love this concept. The world I'm building is almost the inverse of this, taking the reckless real-world exploitation of Earth to its logical conclusions and challenging the player to find a better way of life in the ruins. But despite the different approaches, I think we're both trying to imagine how this world could be better, instead of giving in to bleak, helpless doomerism. We need more fiction like this.
Yes, and my approach does capture that to some extent. Or rather, Godot's ambient lighting approach does. The (ir?)radiance cube map(s?) cover all directions, including the ground. Right now I'm using a uniform gray from below — [albedo 0.39](https://en.wikipedia.org/wiki/Albedo_0.39), but I just learned that this value is outdated. For slightly more realism, I could colour this according to the terrain underneath. Future work!
Of course, there's a chicken-and-egg problem here that isn't captured: the very presence of the clouds also affects how much light is reflected back onto them. That would be a bit harder to achieve.
That's where I learned about the transference phase function (slide 24), which includes two spikes on the return angle that correspond to the relatively unknown Glory rainbow: https://en.wikipedia.org/wiki/Glory_(optical_phenomenon)
I don't know if this method is what is used in video games today.
Thanks for the pointer! It seems like my approach is similar to, but much more simplified than, what is described in the slides (the thesis itself is not loading for me right now). I'm using a 4D function (LUT) rather than 5D. The correspondences are: size/thickness (of a sphere, not a slab), lighting angle, view angle. I additionally use the surface normal, they use viewpoint depth and scattering order.
The observation of different behaviours for different scattering orders is interesting. I could try adding a cheap approximate first-order term to the shader to get more pronounced silver linings.
My LUT is calculated using the Henyey-Greenstein phase function, so without the backward spikes. Since this is all computed offline, I could easily add the more accurate Mie function.
An interesting read, given nowadays no one seem to care anymore how things are actually done in detail.
Doing a pretty sky in my game? Just delegate the entire process to AI.
If it doesn’t look good just keep prompting your way to a pretty sky. Learn nothing while you do it, and when you need to do a simulated 3D sky again, have zero reusable knowledge, after all you can just point your codex at your successful previous attempt.
20 comments
[ 2.2 ms ] story [ 21.3 ms ] threadThe game is even more amazing in VR where you really get a better spatial understanding of the entire sky, environment and the clouds.
Sounds like exactly what I was looking for... Strange coincidence.
https://store.steampowered.com/app/1764530/Sailwind/
There are a couple of ships you can sail with a simple economic engine - you start small and work up. It has some simple sail handling - its not enough to just turn right/left and manage sails up/down. Multi-sail ships are an interesting challenge but manageable.
It's pretty slow-paced, but like I said, I had a lot of fun with it for ~20 hours.
I particularly enjoyed the post about cloud, and I also watched Ten Skies lately... Strange coincidence.
Game itself sounds interesting and I look forward to playing.
> "On Earth, exploration often meant gruesome exploitation and violence, whose effects can be felt even today... However, this game is not about any of that; it shows how things could have gone, if people had behaved more kindly towards each other."
I love this concept. The world I'm building is almost the inverse of this, taking the reckless real-world exploitation of Earth to its logical conclusions and challenging the player to find a better way of life in the ruins. But despite the different approaches, I think we're both trying to imagine how this world could be better, instead of giving in to bleak, helpless doomerism. We need more fiction like this.
Of course, there's a chicken-and-egg problem here that isn't captured: the very presence of the clouds also affects how much light is reflected back onto them. That would be a bit harder to achieve.
The fun part for me is that you managed to get 2 comments out of 9 to end their reply with <end statement>... Strange coincidence.
...Stranger coincidence.
You can see screenshots of that work in his thesis defense slides https://theses.hal.science/tel-00319974/file/defense.pdf
That's where I learned about the transference phase function (slide 24), which includes two spikes on the return angle that correspond to the relatively unknown Glory rainbow: https://en.wikipedia.org/wiki/Glory_(optical_phenomenon)
I don't know if this method is what is used in video games today.
Incidentally, you may recognize his advisor Fabrice Neyret as someone very active in the Shadertoy community https://www.shadertoy.com/user/FabriceNeyret2
The observation of different behaviours for different scattering orders is interesting. I could try adding a cheap approximate first-order term to the shader to get more pronounced silver linings.
My LUT is calculated using the Henyey-Greenstein phase function, so without the backward spikes. Since this is all computed offline, I could easily add the more accurate Mie function.
What is used in games today is more raycast-based, e.g. Nubis³: https://www.guerrilla-games.com/read/nubis-cubed
And Neyret's name is well known to me, yes :)
Doing a pretty sky in my game? Just delegate the entire process to AI.
If it doesn’t look good just keep prompting your way to a pretty sky. Learn nothing while you do it, and when you need to do a simulated 3D sky again, have zero reusable knowledge, after all you can just point your codex at your successful previous attempt.