Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
But here is hoping you find a breakthrough on water calculation.
Can't wait to try out the water when the update comes.
I want a secret setting which would adjust that. If you put it in the C# part of the code is fine ;)
https://twitter.com/Vuthric/status/1512640001472962565
Probably not. :)
Hehe, well, tbh I didn't actually test it :D But based on the the current flow behaviour and performance implications, it's safe to assume that this would be a performance killer :D
Maybe it's not a bad idea to have a hidden option for that ^^
That's a good idea (irrespective of water), I will put that on our to-do list :)
Although the water system in RW is CPU bound (since all flow calculations happen on the CPU). The water system definitely benefits from many CPU cores.
Most likely yes: Placing a regular water block somewhere will not work in this case (it will just flow away), However, we will probably implement a special "infinite" water block (which acts as an infinite source of water), which would be suitable for a water fall. We still have to run some tests to get a better idea of the overall performance implications of such a feature ^^
Unfortunately this is indeed not the type of fluid simulation we will see in RW =/ This is definitely a very impressive fluid simulation the Epic dev is showing there, but you have to take it with a grain of salt: This kind of water simulation isn't possible on a larger scale, e.g. having a lake or even an ocean with this type of water. It's great for small scale water simulations though (filling a room with water, for example, or maybe a swimming pool with dynamic water).
This is basically a GPU based particle simulation: There are tens or hundreds of thousands of individual water "particles" (like small blobs) simulated on the GPU (which is good for parallelizable tasks). A special type of renderer is then responsible to visually "merge" these individual blobs (so they look like a continuous body of water).
This has a few limitations though: Even a powerful graphics card will struggle if there are too many water particles (a powerful modern GPU will probably be able to handle a few millions of particles, but that's still not enough to fill a lake, for example).
Another issue are persistent worlds: If you want to store the current state of the water, you need to get the data back to the CPU. This will be a bit problematic if there are really millions of water particles out there - this results in way too much data. In multiplayer, this also needs to be synced across all clients. This isn't necessarily impossible to do, but it opens up a whole new can of worms ^^
tl;dr Flowing water with this level of detail is only suitable for small-scale scenarios nowadays (small river / waterfall, ponds, swimming pool etc). It will take ~ 10-20 years until we can see something like that on a large scale (e.g. as a full water solution in a game like Rising World) ;)
Also yeah i think we should have a setting that controls water updates per second, based on the GIF we saw, it looks like water updates 20 times per second which is like 20 FPS?
People with weaker CPUs could lower the number, water physics will act more janky but it would be less taxing, also the setting is server side, if you join a server where the water update is 10, then it only updates 10 times per second, even if a player's water update is say 60 (would be 60 FPS) it would still be 10 times per second since the server controls it. But if a player's water update was 5, then for them it would only be 5 times per second, server still has it at 10, but players with a lower number than the server will take priority for performance reasons.
Well observed ;) The water on the gif is updated approx. every 50 ms, resulting in 20 updates per second. This value is already configurable, and indeed, reducing the update interval definitely takes some load from the CPU.
Exactly: In multiplayer, water flow is calculated on the server, so only the server setting is relevant in that case.
Currently this is unsupported (the player cannot force the game to update the water less frequently than the server), but I agree that it would be quite helpful if the client could at least lower the update frequency on his end... it's a bit tricky to implement though, because water updates are basically like regular world edits, and we have to make sure the world is still properly synced if the client skips certain world edit packets. But it's on our to-do list :)
It looks much better than I expected, when we first talked about water in another old thread.
Great job!
Our water system is written in a way that it can basically handle various liquids with different viscosity - so oil or even lava could flow slower than regular water ^^