Will space end?

[SunlitZelkova]

In stock KSP 1, there is no interstellar travel. So even if you do a Jool flyby and burn at periapsis during it, no matter how high your velocity is you will never leave the Kerbolar system, and remain in a very hyperbolic solar orbit.

In stock KSP 2, there will be interstellar travel. You leave the Kerbolar system, fly through interstellar space, and then intercept a star (or maybe planet).

But what if you don't intercept (either deliberately or accidently)? Does your spacecraft keep flying endlessly onwards to no where? Or will it hit an "edge" and be destroyed or turn around?

[PlutoISaPlanet]

I like to think the ship will just explode

[linuxgurugamer]

Eventually you will run into some sort of math limitation

[Dientus]

14 minutes ago, linuxgurugamer said:

Eventually you will run into some sort of math limitation

Depending on how its coded, I imagine that would be a real possibility. But I was thinking that the devs wouldn't just leave it at that... Game crashing at an overflow or division by zero error... that is poor design among other things.

 

Since the devs will have a decent idea where this may occur, I would imagine they would just have the ship disappear with all hands considered killed well before this point. I doubt they would do it 'asteroids style' and you reappear on the extreme other side. Having the ship considered 'lost' leaves open possibilities of future expansion of new systems as well.

 

[linuxgurugamer]

In reality, it would probable just be an endless voyage.  If done correctly, they wouldn’t get into a math overflow issue

[MechBFP]

When you are in interstellar space you are still in an orbit, just around the galactic center. I imagine the dV requirements to escape that is rather excessive lol. 
 

Edit: Milky Way escape dV is ~550,000 dV for example. 

Edited April 11, 2021 by MechBFP

[Tw1]

Glitch based acceleration and save manipulation have seen people go crazy far out even in KSP1, resulting in glitchiness, so I suspect the same will be the case here.

[Dfthu]

Just keep going forever like KSP 1.

 

[swjr-swis]

2 hours ago, Dientus said:

I doubt they would do it 'asteroids style' and you reappear on the extreme other side.

Why not? Coding such a thing could practically be 'natural', and mathematically/physically... well, we're still not sure of the nature of our universe. It may indeed wrap in on itself, just well beyond our observation horizon.

I'm kinda partial to the Asteroids solution.

[K^2]

4 hours ago, Dientus said:

Depending on how its coded, I imagine that would be a real possibility. But I was thinking that the devs wouldn't just leave it at that... Game crashing at an overflow or division by zero error... that is poor design among other things.

That's not the kind of problems you start getting. Coordinates are almost always stored in floating point format. We can separately discuss finer points of single vs double precision arithmetic as well as the use of origin relocation in games to avoid the worst of the problems, but the gist of it is that the further you get from (0, 0, 0), the worse the math precision gets. If handled improperly, you start getting camera and various object jitters and physics grows unstable. Eventually, Kraken would get you. Of course, even KSP is large enough that this had to be addressed early on, most likely via relocation. With relocation in place, it's possible to entirely avoid physics problem. Camera behavior is still subject to implementation. But eventually, you'll get to the point where you simply can't move origin anymore at given velocity, and the craft will freeze in place, still showing the velocity, but unable to actually make progress.

You'll have to get mighty far before that happens, though, and by then, how would you even know? You'd be too far out to notice any change in your position relative to anything else. So I think it's in the category of problems that aren't.

[kerbiloid]

Once you get enough far, your craft will destroy the universe by math error.
Let's just hope this won't happen to Voyagers.

Edited April 11, 2021 by kerbiloid

[Dientus]

2 hours ago, K^2 said:

But eventually, you'll get to the point where you simply can't move origin anymore at given velocity, and the craft will freeze in place, still showing the velocity, but unable to actually make progress.

When I read this the first thing that came into my mind was an event horizon!

 

We can make reasonable guesses on how KSP2 could possibly act based on other games using a coordinate system in a virtual 3 dimensional space, and I do understand what you are saying. My thought was before any of this happened, regardless of outcome, that they would simply consider craft and kerbal lost. Of course, @swjr-swis could be right and they code it like a pocket universe based on Hawkins theories as well. Code-wise that's easy enough to do as well, or maybe they actually model it with a SOI of a black hole in the center and they let it go like KSP where you never leave the galaxy. They can even do something entirely different and unexpected.

 

So I was curious, which way would you guess the devs handle this?

[K^2]

11 minutes ago, Dientus said:

My thought was before any of this happened, regardless of outcome, that they would simply consider craft and kerbal lost.

Well, the reason I don't like the idea of placing arbitrary boundary on distance is because it might make it difficult for a mod to place something waaay out in the distance.

I think a reasonable solution would be to do origin relocation in two jumps. First places coordinate origin at the nearest star. Second places the simulation/rendering origin at the craft. This basically means that distances can get a tiny bit wonky really, really far between stars, but you won't have a point of reference to notice. And you won't get krakeened anyways, since all your physics is relative to craft's center of mass. And once you get closer to another star, you'll switch to that one as your reference and everything's fine again. That would allow modders to place things way, way out there. And yeah, maybe if the craft is far enough from any star and is on trajectory that's taking it away from all of them, then maybe it should be considered lost. I just wouldn't base it exclusively on distance from arbitrary zero point.

[Dientus]

15 minutes ago, K^2 said:

I think a reasonable solution would be to do origin relocation in two jumps. First places coordinate origin at the nearest star. Second places the simulation/rendering origin at the craft

That, my good sir, would be an excellent idea! So the point of 'lost' is based on the outer most edge of the outer most star, so to speak. It seems doing it this way could very well create a potentially unlimited universe for the lil guys and gals to explore with minimal Kracken attacks. 

 

[Single stage to ocean]

Then what if all the galaxies orbit a star that is impossible to escape, and calculations are done like in KSP 1?

[SpaceFace545]

12 hours ago, Tw1 said:

Glitch based acceleration and save manipulation have seen people go crazy far out even in KSP1, resulting in glitchiness, so I suspect the same will be the case here.

I'd be plenty mad if the devs made kraken drives impossible

[MechBFP]

5 minutes ago, SpaceFace545 said:

I'd be plenty mad if the devs made kraken drives impossible

Expect to get mad then. The developers aren't going to purposefully program bugs. Ridiculous to even expect it.

[PlutoISaPlanet]

Kracken Drive would be a bad, funny, but bad.

[Terminal Velocity]

Maybye you could keep going until you reach the supermassive black hole at the center of the galaxy and well, you know what would happen 

[B.Riches2500]

21 hours ago, SunlitZelkova said:

In stock KSP 1, there is no interstellar travel. So even if you do a Jool flyby and burn at periapsis during it, no matter how high your velocity is you will never leave the Kerbolar system, and remain in a very hyperbolic solar orbit.

In stock KSP 2, there will be interstellar travel. You leave the Kerbolar system, fly through interstellar space, and then intercept a star (or maybe planet).

But what if you don't intercept (either deliberately or accidently)? Does your spacecraft keep flying endlessly onwards to no where? Or will it hit an "edge" and be destroyed or turn around?

I think it would be cool if you could do a Voyager type thing and fire something out of the system, but due to its speed and distance between systems, it probably wont go anywhere and will just be left drifting in space.

[Wubslin]

On 4/10/2021 at 7:28 PM, linuxgurugamer said:

Eventually you will run into some sort of math limitation

In KSP 1, sure.

It has something to do with quaternions and double length floating point limitations and wizard magic if I recall correctly? Anyways, KSP 2 won't have that problem because there's going to be a grid system. Representing extreme distances becomes doable because you're never really that far from a boundary in the grid.

Anyways for the orbital thing, there might be stellar SOI and the heliopause would be like the boundary where outside you'd be orbiting the galactic core...? Maybe there'd be fancy N-body but I doubt it. And what of relative stellar motion? Things in space outside of interplanetary scales are so alien to me.