For the prototype, they are just stuck in a position above the surface of planets.
But there is no technical reason why they can’t orbit. It’s just to keep them in a familiar spot without having to hunt for it in the prototype and such and probably a matter of priorities, too.
I believe the plan is to make them actually orbit and such later on.
Well, first you start with the cyclone technology and insert it into a cylinder. Then, you insert it into a smaller cylinder.
Then you make the cylinder a sphere.
[dirty layman]
Isn’t it a distinction without a difference as it is in the latest (not the “dozens planets/moons” version) public demo? The planets do not orbit the star, but the planets do rotate, and gravity is modeled. So what’s the difference? Correct me if I’m wrong (please), but as it is the planetary ring, and objects therein, are basically at orbital velocity and/or effectively have perfect and infinite/perpetual station keeping.
In practice that is no different from what a “real” orbit would be, isn’t it?
[/layman]
The moons don’t orbit the gas giant, either.
It’s not that they can’t. They just wanted them always in sun light for the prototype so people don’t have to wait a long time to get daytime shots of them.
It’s really a nonissue thing right now. It’s just how it is and there isn’t a big technical problem or anything keeping them from doing it. Just have to wait until later.
I know – I was just asking if there was any practical difference between the planet being fixed but rotating and having a proper gravity well, and the planet also orbiting its star. From an object in orbit (e.g. that mining roid in the moon’s belt), what’s the difference in terms of Waisso’s question?
Ah i see.Its just that here Flaiven explains why they had to disable planet orbits since its an apparently extremely complicated problem when dealing with online networks and getting everyone synced up. Elite: Dangerous
So It seems planets orbiting the stars is off the table for now unless they get the required funding, but I hope stations/moons are going to be orbiting guaranteed. But a low orbit around a gas giant could be upwards of 50km/s depending on the mass of the gas giant. But could the servers really handle keeping say 30 people in sync with each other moving 50km/s + speed relative to each other in combat?
It’s a curious thing, that what is basically a clockwork, and so ostensibly a perfectly predictable framework, would cause trouble for MP synchronisation. Planets and moons, that is. Nothing could desync them.
Not because of funding, but because it makes organizing filming for streamers waaay better so they don’t have to wait for the planets to align(literally). It’s entirely possible and doesn’t require anything extra as far as I know it.
Potentially different tactics to use based on moons relative positions and wheher they are totally dark or not, is one reason it’s nice to have.
And yeah, apparently Flavien did say there are actual technical problems to having the planets/moons move right now.
I imagine the problem is similar to many games online where you have people on a moving platform, and you see them bouncing up and down. But… there is a solution to everything.
Nothing’s still impossible, it’s just wheher it’s supported enough to be worth the effort, I’m sure. I don’t think the game really suffers if they’re static, though. It’d be a bit buzz killing, but not game ruining.
I don’t totally understand that issue, though. Like, you have an orbit and blah blah. You’re not sending a server update for the planets position, but instead keeping times in sync to tell the body to be at such and such orbit it should be at a given time.
You send a timestamp with each packet, and you can adjust the clients delta to make sure the clients clock is running roughly the same time as the server’s, no?
edit: nevermind thought about it more and now I see the problem.
Right… Normally the deviation from the initial sync should be negligible, over the span of any given normal game session. How much deviation over a single normal game session (3 sigma or so?) would you get for the major celestial bodies (planets and moons), and would that be significant in terms of gameplay?
It just seems negligible, but maybe Flavien ruled it out because it actually isn’t.
No, I’m saying that wouldn’t happen because say the client connects, you compare timestamps and set a delta.
Then when you send some other network data for various things, you have other timestamps to compare deltas and you can see if the clients clock is running faster or slower than the servers, and adjust what they consider to be the real time that they are using for the client side movement of celestial bodies to keep them in sync with the server and thus all players get everything in the same space.
That doesn’t actually work though, nevermind.
I guess the issue comes that in the variation of time between syncing back up, even a microsecond can be thousands of kilometers that a moon is out of position I guess?
I can see another solution, though! When the server sends positional data, it can send it with an ID for the largest body to position relative to.
In a change over process, you could get some tearing of movement, though. Like say you’re flying from one moon to another along side someone, the moment it switches over to say your position relative to the other body with someone else, there could certainly be a jerking.
But this makes it so when you have ships fighting around a planet that’s moving, they’re all pretty perfectly in sync.
The orbital velocity of Earth is 30km/s or 30,000 m/s. A microsecond is 10^-6 seconds. So in 1 microsecond Earth moves only.3 m or 30 centimeters… Not quite thousands of kilometers. Not sure what you’re arguing for because I’m not necessarily following but I do know numbers!
Edit: A moons orbital velocity will always be less than the velocity of its parent otherwise it wouldn’t be orbiting the parent but the star.
Ah, and I’m not too good with thinking at that scale and how much goes on in such a time frame. 
Hm… well you can be off by a lot more than 1 microsecond when you have packets to compare clocks to a few times a second or every few seconds. A milisecond is 300 meters!
Comparing by a microsecond just seemed like a good example of “even in the best case scenario, things could be really messed up”.
Anyway, to add to that idea I had, even if you did get the celestial body sim and sim for other things that are on “rails” and procedurally positioned to a timer out of sync, if positions were given relative to the closest major thing, it seems like it’d solve those issues.
Wish I had time to throw together an example, but yeah, conceptually it’s pretty simple.
Instead of saying a ship is moving at 15,000 km/s it a vector that puts it near a moon, you say that moon has no actual speed and position but is on rails that allows the client to extrapolate its movement based on the clients clock but effectively it’s moving through space at 10km/s, you say the ship is moving 5km/s relative to that body at such and such vector.
edit: and what would happen when there are catchups and slow downs is that the closest planet/moon and all the ships whose positions/vectors that are set relative to it, including yours, would all move together in such a way that would be completely unnoticeable since it’s a small distance compared to what’s much further away.
Did you miss “as far as I know” written there?
Then with no centrifugal force the thing collapses into a ring. Thus Ring World books.
A pop to Wikipedia… “Most fictional depictions describe a solid shell of matter enclosing a star, which is considered the least plausible variant of the idea”
not looking much like a sphere.
“2013, at the Starship Century Symposium in San Diego, Dyson repeated his comments that he wished the concept had not been named after him.”
The spheres look like more trouble then they are worth. Get a sphere maker to convince me or I don’t think it’s practical.
Now your telling me that guy on the radio lied to me!!! first time ever I’m sure!
They’re not if you want to capture all the energy a star radiates. Then there is no other option. Now don’t ask me what we’d need that much energy for, but I’m sure we could think of something.
Space is probably big enough that (technologically) by the time you can build and orbit that many objects around a star, you have the means to coordinate them to not collide even in a full sphere instead of just such a ring/toroid. Any of these swarm shapes can probably get crowded enough to capture really substantial amounts of energy.
Can we folks keep this topic about screenshots pretty please.
It is getting really confusing here.
Here have a relevant topic.
If you read down a bit farther in the thread you guys are referencing, you’ll see that I describe how I solved the problem in the warp prototype, and Flavien comes back and states that he solved it in a similar way back in the ICP.
He knows how to solve the problem, but he has existing code that was built assuming stationary planets. Stationary planets were used because it makes everything vastly simpler to implement, and they have been - and still are - pressed for time. To solve the problem for orbiting planets, that existing code must be retrofitted, and it’s a bunch of work. It’s a bit like a 32-bit to 64-bit precision conversion. It gets into a lot of stuff and can be a touchy change to make. It’s definitely not something that he can tackle during the Kickstarter.
Does anyone have any new latest screen shots please so I can post them on the SC and ED forums. Some with 2/4 way dog fights going in and out of the atmosphere would be great. In fact videos of 3 dog fighting all around and below one another whilst transitioning the atmosphere would be very good. Thanks in advance.
