On KSP2 and Special Relativity

[LitaAlto]

I'm going to try to be as careful as I can with my phrasing here, because I really don't want to be misconstrued. (I tried asking this on one of the KSP groups on Facebook and it spiraled out of control pretty fast.)

KSP2 developers stated that there will be no FTL drives in the game. I'm perfectly happy with this, and to be honest, would have been let down if they did add warp drives, wormholes, or other handwavium.

Also, implementing full Special Relativity is way impractical for a game, and besides, the math gets hairy enough that I can't imagine a developer spending time trying to add proper SR into a game.

However, it does occur to me that a developer could easily add the speed of light as an upper speed limit for ships, and also model acceleration so that, as your velocity is a larger and larger fraction of the speed of light, it becomes harder to accelerate faster.

For stock KSP1, the parts are nowhere near powerful enough to reach sizable fractions of the speed of light with a reasonable fuel supply. However, they can in theory exceed the speed of light with the infinite fuel cheat, and with enough patience. Also, glitches can and have sent craft well exceeding the speed of light, as Danny2462 and others can attest.

For KSP2, presuming that the Project Orion/Project Daedalus parts are similar to their real-world counterparts, they'd be able to reach a top speed in the neighborhood of 0.1c, but I can't hold myself to that presumption. It's entirely possible that these parts will perform more powerfully than their real-world equivalents, to reduce the travel time between stars.

So having said that:

• What is the likelihood that KSP2 will have an upper speed limit?
• Do we even want an upper speed limit, as a community? (There are arguments for and against. I'd favor the realism of it taking a long while to reach other stars, even with advanced sublight drives, but I know some will insist on a mod that permits FTL travel, or think the concern is irrelevant.)

I welcome your well-considered and respectful thoughts.

Thanks.

NOTE: I've updated the title and the post below to reflect that I was saying "General Relativity", but as @chaos_forge pointed out, Special Relativity is where the cosmic speed limit was first set. As SR is a special case of GR, my statement wasn't entirely inaccurate, but it was imprecise.

Edited August 22, 2019 by LitaAlto
Updated for clarity, with a note in post explaining why.

[DunaManiac]

By General relativity, do you mean N-body physics?

 

[chaos_forge]

I think the reason people might be getting confused that you're talking about special relativity, not general relativity. Special relativity says that the speed of light is constant and nothing can go faster than it. General relativity says gravity isn't a force but rather the bending of spacetime.

As far as implementing SR goes, relativistic effects only really start becoming noticeable at above .5c IIRC. So I don't think adding relativistic effects is necessary, but I also wouldn't complain if they do decide to add them.

[4472TJ]

I think considering that this is a real world physics simulation that Special relativity should be taken into some account. As we know Special relatively is very complex and modeling it accurately will be very hard so a simpler more watered down model could be used, and you've got to make space for those really Kerbal moments where you go faster than the speed of light for no reason! 

Edited August 22, 2019 by 4472TJ

[chaos_forge]

One more thing:

32 minutes ago, LitaAlto said:

It's entirely possible that these parts will perform more powerfully than their real-world equivalents, to reduce the travel time between stars.

Generally, KSP reduces travel times by making things smaller and closer together, not by making parts more powerful. Engines in KSP actually tend to be less powerful than their real-life counterparts (apart from the ion drive, but that's because you can't accelerate during time warp). So if anything, I'd expect the max achievable speed to be lower than .1c, not higher.

 

[LitaAlto]

18 minutes ago, chaos_forge said:

I think the reason people might be getting confused that you're talking about special relativity, not general relativity. Special relativity says that the speed of light is constant and nothing can go faster than it. General relativity says gravity isn't a force but rather the bending of spacetime.

As far as implementing SR goes, relativistic effects only really start becoming noticeable at above .5c IIRC. So I don't think adding relativistic effects is necessary, but I also wouldn't complain if they do decide to add them.

That's a fair point, although one could say GR is a generalization of SR that accounts for spacetime curvature, while SR presumes flat spacetime. But that wasn't really the point of confusion. It had more to do with people presuming I wanted FTL, when I thought I had made clear that I didn't.

23 minutes ago, DunaManiac said:

By General relativity, do you mean N-body physics?

 

No. N-body physics has to do with the collective gravitational influences of more than two bodies in a given system.

And as @chaos_forge pointed out, I mispoke. I meant Special Relativity, since SR is where the speed of light was first set as a hard limit. That hard limit holds true for General Relativity and I tend to refer to GR more than SR, but all the same, my main question has to do with whether it's possible that KSP2 will prevent FTL travel by any means at all, even accidentally through glitches, or if that loophole will still exist in some sense.

[HebaruSan]

I'd love to see a nod to SR, but I doubt that they'll want to get into time dilation.

There's always the "interstellar dust drag" speed limit instead.

[LitaAlto]

17 minutes ago, chaos_forge said:

One more thing:

Generally, KSP reduces travel times by making things smaller and closer together, not by making parts more powerful. Engines in KSP actually tend to be less powerful than their real-life counterparts (apart from the ion drive, but that's because you can't accelerate during time warp). So if anything, I'd expect the max achievable speed to be lower than .1c, not higher.

 

I'm not sure I see how having stars and planets closer together would reduce the maximum practical speed to be lower than 0.1c. Could you elaborate?

[Spaceception]

SR seems like a pretty niche thing to get hung up about. And it only really starts to become noticeable near the speed of light (which is what would really matter in the game).

As far as I'm aware, the only interstellar drives that will be in the game is Orion and Daedalus. Both are limited by 10-15% of the speed of light if it's modeled anywhere close to real life, so I feel SR would be irrelevant to gameplay.

As for things like antimatter (which I suspect will only be in mods), you could get to 50-90% or higher depending on how much fuel you have, but how would SR enrich the experience? How would adding those calculations and everything in the game make it better? (especially if the only drives capable were in mods, not stock. Why would the devs bother?) And why would you add it for odd situations for when the game bugs out? I honestly can't see a reason why we'd need it.

 

 

It really comes down to fun gameplay, and accuracy (and the learning curves associated). Where would things be more worthwhile?

Where KSP is now, I think is mostly good. You could take a page from realism overhaul, and make the aerodynamics better, make reentry better, make reaction wheels not as op, so you need RCS more, and so forth. And that could add to gameplay, but things like SR would be irrelevant most of the time when you aren't traveling between systems with sublight drives, and (another example) N-body would just make the game harder to learn, and (from what I've seen) be more tedious to play.

Edited August 22, 2019 by Spaceception
some corrections

[MechBFP]

I was kinda hoping for FTL drives so that I could send a message back to Jeb telling him to send an expendable Kerbal instead.

[LitaAlto]

3 minutes ago, Spaceception said:

 

It really comes down to fun gameplay, and accuracy (and the learning curves associated). Where would things be more worthwhile?

For me it has more to do with setting the realistic expectation that traveling between stars takes a lot more time than it would between planets.

I agree that, for gameplay purposes, the developers may decide this is too much of a complication to add in.

I'm not sure we can presume that Orion/Daedalus derived parts will only reach speeds in the neighborhood of 0.1c, however. The devs could decide that speeds that would be relativistic in the real world is the more practical approach, and that could occur even if the galaxy is scaled down similar to the way the original solar system is.

2 minutes ago, MechBFP said:

I was kinda hoping for FTL drives so that I could send a message back to Jeb telling him to send an expendable Kerbal instead.

I'm sure a modder will add some sort of FTL drive, and I wouldn't be surprised if one of the first mods for KSP2 is a variant of the Alcubierre drive.

[HebaruSan]

4 minutes ago, Spaceception said:

but how would GSR enrich the experience? How would adding those calculations and everything in the game make it better?

It would fit the theme of "learn by doing" that we all enjoyed so much for orbital mechanics. Rather than reading textbook descriptions of SR or even sci fi stories, we would directly see how fast we could go, how long supplies last, or how much time passes back home (DIY twins paradox!). We'd experience how those considerations affect mission planning. Again I doubt the devs will do it, but I would definitely find it enriching if they did.

Somewhat relatedly, I'm excited to play around with the boundary between interplanetary and interstellar travel. What additional considerations are there when departing for bodies that don't orbit your sun? Should we sun-dive for the Oberth effect? What kinds of orbits do you end up in when capturing from interstellar space? (Yes, I know there are mods for some of this.)

[chaos_forge]

17 minutes ago, LitaAlto said:

That's a fair point, although one could say GR is a generalization of SR that accounts for spacetime curvature, while SR presumes flat spacetime.

GR is a generalization of SR, but the case of a ship traveling through interstellar space is perfectly encapsulated by SR, so there's no need to make things more complicated. Quantum mechanics is a generalization of Newtonian mechanics, but that doesn't mean I start bringing up wave functions every time I wanna talk about parabolic motion.

9 minutes ago, LitaAlto said:

I'm not sure I see how having stars and planets closer together would reduce the maximum practical speed to be lower than 0.1c. Could you elaborate?

It's not that having them closer together would reduce the speed, it's that engines in KSP tend to be less powerful than their IRL counterparts, because things are closer together. So if anything, I'd expect the engine in KSP to be less powerful than the Daedalus engine, not more powerful.

9 minutes ago, LitaAlto said:

For me it has more to do with setting the realistic expectation that traveling between stars takes a lot more time than it would between planets.

Would that expectation not already be established by the simple fact that stars are way father away from each other than planets are?

[Delay]

3 minutes ago, HebaruSan said:

Should we sun-dive for the Oberth effect?

Sun diving would be horrendously inefficient in this case, right? You only need to double Kerbin orbital velocity to leave Kerbol orbit, rather than subtracted nearly all and then adding a lot...

[Spaceception]

1 minute ago, LitaAlto said:

For me it has more to do with setting the realistic expectation that traveling between stars takes a lot more time than it would between planets.

I agree that, for gameplay purposes, the developers may decide this is too much of a complication to add in.

I'm not sure we can presume that Orion/Daedalus derived parts will only reach speeds in the neighborhood of 0.1c, however. The devs could decide that speeds that would be relativistic in the real world is the more practical approach, and that could occur even if the galaxy is scaled down similar to the way the original solar system is.

Well, not really I would say. We have timewarp, so we don't need to wait several years real-time to get somewhere. So why would that matter? Yes, waiting for timewarp between stars would take longer than waiting in interplanetary space (unless they add enhanced timewarp for interstellar space), but other than that, I don't feel that expectation. just boredom unless I have another mission going.

The only way for "distance" to really matter in gameplay between stars (and outer planets), is if they added some sort of comms delay. That might be interesting, but it would make it harder to perform longer missions. So then you'd want to send a crew mission to some places first, so you can remotely control it from there, and delay wouldn't be long. Like long-term planning. Or have a crewed mission follow a probe, so it can scout the system without too much delay before they arrive, something like that.

I think we can. And like @chaos_forge said, they may even be less powerful. Because if it wasn't, it would just make Orion way too OP, with its high thrust/high isp build.

[HebaruSan]

7 minutes ago, Delay said:

Sun diving would be horrendously inefficient in this case, right?

That's what I look forward to finding out! It's a thing in A World Out of Time, so there may be scenarios where it makes sense.

7 minutes ago, Delay said:

You only need to double Kerbin orbital velocity to leave Kerbol orbit, rather than subtracted nearly all and then adding a lot...

But mere solar escape velocity would be quite slow for interstellar travel. You'd be depending a lot on acceleration after you leave, and a big kick from a sun-dive could help there.

[LitaAlto]

15 minutes ago, chaos_forge said:

It's not that having them closer together would reduce the speed, it's that engines in KSP tend to be less powerful than their IRL counterparts, because things are closer together. So if anything, I'd expect the engine in KSP to be less powerful than the Daedalus engine, not more powerful.

...

Would that expectation not already be established by the simple fact that stars are way father away from each other than planets are?

Those are fair points. I just don't think we can presume that KSP2 developers will nerf Orion/Daedalus engines by default.

Also, I half-expect someone to create a Whackjob-inspired monstrosity. Although the overall mass if the ship, including fuel, may prevent sizable fractions of the speed of light, I can't rule out someone finding a way to kludge around that.

Maybe I *am* making much ado about nothing in the end, but all the same, this is the kind of thing I think about, maybe to the point of obsession, when playing games like this. Please don't get me started about Kerbin's density.  

[magnemoe]

44 minutes ago, chaos_forge said:

I think the reason people might be getting confused that you're talking about special relativity, not general relativity. Special relativity says that the speed of light is constant and nothing can go faster than it. General relativity says gravity isn't a force but rather the bending of spacetime.

As far as implementing SR goes, relativistic effects only really start becoming noticeable at above .5c IIRC. So I don't think adding relativistic effects is necessary, but I also wouldn't complain if they do decide to add them.

Agree, orion or fusion engines will not reach relativistic speeds. Note that the orion design for interplanetary travel would not be the NASA orion who has been moded into KSP1 or battleship orion used in footfall is not interstellar design, that one would be something with an pusher plate multiple kilometer wide and using bombs of more than 100 megaton. it would have battleship orions as boats in docking bays. 

Yes both the NASA and battleship orion makes interplanetary travel an joke. Aim at target, to optimize add some planetary diameters ahead. Burn towards Eeloo and get there in 80 days. 
We will also get other engines the Mun landing in the start used an vasmir or some primitive as in near tech fusion drive on the orbital stage. 
 

[Spaceception]

25 minutes ago, HebaruSan said:

It would fit the theme of "learn by doing" that we all enjoyed so much for orbital mechanics. Rather than reading textbook descriptions of SR or even sci fi stories, we would directly see how fast we could go, how long supplies last, or how much time passes back home (DIY twins paradox!). We'd experience how those considerations affect mission planning. Again I doubt the devs will do it, but I would definitely find it enriching if they did.

Somewhat relatedly, I'm excited to play around with the boundary between interplanetary and interstellar travel. What additional considerations are there when departing for bodies that don't orbit your sun? Should we sun-dive for the Oberth effect? What kinds of orbits do you end up in when capturing from interstellar space? (Yes, I know there are mods for some of this.)

Well, okay, you have a point. But (I think) that's only if they add sublight drives, and decide to add in relativity alongside it. Which I'm not personally expecting.

Yeah, I wonder if it would be possible to model interstellar dust, so you would need to plan for an ablative shield of sorts to protect your craft, and how impacts might affect your velocity over time (each impact is like a nuke after all). Or if that would just be too intensive on the game and be more trouble than its worth...

Edited August 22, 2019 by Spaceception

[LitaAlto]

5 minutes ago, Spaceception said:

Well, not really I would say. We have timewarp, so we don't need to wait several years real-time to get somewhere. So why would that matter? Yes, waiting for timewarp between stars would take longer than waiting in interplanetary space (unless they add enhanced timewarp for interstellar space), but other than that, I don't feel that expectation. just boredom unless I have another mission going. [Emphasis added.]

That's actually part of my concern. In stock career mode, if you get caught up in a long-term mission without paying attention to the rest of the game, you can have contracts expire, transfer windows missed, and in some cases, slingshotting if you're not careful to avoid your ship's orbit passing too close to a body with an SOI. I can only imagine these issues amplifying once space travel is measured in terms of decades or centuries.

[Spaceception]

1 minute ago, LitaAlto said:

That's actually part of my concern. In stock career mode, if you get caught up in a long-term mission without paying attention to the rest of the game, you can have contracts expire, transfer windows missed, and in some cases, slingshotting if you're not careful to avoid your ship's orbit passing too close to a body with an SOI. I can only imagine these issues amplifying once space travel is measured in terms of decades or centuries.

In this case, something like Kerbal alarm clock, might be a much more important mod. Maybe important enough to become stock. Especially if you could make the game stop timewarp before any certain point so you don't miss things like that.

[magnemoe]

3 minutes ago, LitaAlto said:

Those are fair points. I just don't think we can presume that KSP2 developers will nerf Orion/Daedalus engines by default.

Also, I half-expect someone to create a Whackjob-inspired monstrosity. Although the overall mass if the ship, including fuel, may prevent sizable fractions of the speed of light, I can't rule out someone finding a way to kludge around that.

Maybe I *am* making much ado about nothing in the end, but all the same, this is the kind of thing I think about, maybe to the point of obsession, when playing games like this. Please don't get me started about Kerbin's density.  

You might be able to  do asparagus staging with orion pulse nuclear drives , that would be kind of Kerbal. 
Hold my beer. 
You could do it with the big fusion powered craft to  1+8 to top it. 
Jeb is running away visibly scared 

[kerbiloid]

In KSP orbits and planets are ~1/11 of their real size.
But their gravity acceleration is same.

Usually they say, the planets are dense.
I always insist that their density is normal, they are made of same rocks, but in the Kerbal Universe gravitational constant is increased.
(You don't know which part of GM is increased, so I believe that G).

Spoiler

M = (4/3)*pi*R³*density;
g = GM/R² = G * (4/3)*pi*R³*density / R² = G * (4/3)*pi*R*density = G*R*const;
So, as g is the same, R is 11 times less, then Kerbal gravitational constant G = G_IRL*11;

So, just decrease c, say, by an order of magnitude.
Then if Kerbal lightspeed = 30 000 km/s = 0.1 lightspeed IRL, then both Orion and Daedalus become relativistic starships.

[Spaceception]

3 minutes ago, kerbiloid said:

In KSP orbits and planets are ~1/11 of their real size.
But their gravity acceleration is same.

Usually they say, the planets are dense.
I always insist that their density is normal, they are made of same rocks, but in the Kerbal Universe gravitational constant is increased.
(You don't know which part of GM is increased, so I believe that G).

  Reveal hidden contents

M = (4/3)*pi*R³*density;
g = GM/R² = G * (4/3)*pi*R³*density / R² = G * (4/3)*pi*R*density = G*R*const;
So, as g is the same, R is 11 times less, then Kerbal gravitational constant G = G_IRL*11;

So, just decrease c, say, by an order of magnitude.
Then if Kerbal lightspeed = 30 000 km/s = 0.1 lightspeed IRL, then both Orion and Daedalus become relativistic starships.

3,000 is the new 30,000

Spoiler

I'd photoshop this if I could

 

 

[kerbiloid]

3 minutes ago, Spaceception said:

3,000 is the new 30,000

3000 can be an overkill for Orion and Daedaulus (they are up to 0.1 c irl = 30 000 km/s), as well for KSPI-E engines.

Edited August 22, 2019 by kerbiloid