KSP2 Hype Train Thread

[darthgently]

40 minutes ago, tstein said:

The hard thing is PREDICT the orbit  in future without incurring a large computational cost

That is what I meant by performance hit.  Without the ability to predict orbits KSP would not have any customers

[tstein]

6 minutes ago, darthgently said:

That is what I meant by performance hit.  Without the ability to predict orbits KSP would not have any customers

I am sure it would have a few... We would all see competitions to   get to certain bodies with minimum Dv at an extreme level.

[darthgently]

1 hour ago, tstein said:

I am sure it would have a few... We would all see competitions to   get to certain bodies with minimum Dv at an extreme level.

With Kepler orbits and no orbit prediction, I'll give you a few.  N-body with no orbit prediction?  Who? Lol

[magnemoe]

3 hours ago, GoldForest said:

I'm really interested in seeing how they handle the twins spinning around each other, while being tidally locked, while also being N-bodied, while also orbiting around Deb Deb. Or was it To Be Announced?

The two bodies would be on rails even if they used n-body inside the SOI, The idea of using n-body on planets and moons is nuts in a game like KSP. 
Apollo was planned with patched conics.  Artemis 1 used n-body to get into orbit around the moon cheaper.  And I'm pretty sure you need n-body to send an probe to other planets and you obviously need it for more advanced orbital stuff. 
But I doubt they use n-body to calculate Mars position as we know its orbit. Jupiter's moons was used to synchronize clocks back in the 19th century. 

But one pretty weird for people who are not astronomers or play lots of KSP  is that we can not predict an comet like halley's comet more than an couple orbits. 
Reason is simple Pe is very high compared to Ap. Tiny changes to velocity because of n-body interaction will have much larger effect on Pe and therefor orbital period who changes how it interact with other bodies on the next pass so the uncertainty increase.

[Vl3d]

48 minutes ago, magnemoe said:

The two bodies would be on rails even if they used n-body inside the SOI, The idea of using n-body on planets and moons is nuts in a game like KSP. 
Apollo was planned with patched conics.  Artemis 1 used n-body to get into orbit around the moon cheaper.  And I'm pretty sure you need n-body to send an probe to other planets and you obviously need it for more advanced orbital stuff. 
But I doubt they use n-body to calculate Mars position as we know its orbit. Jupiter's moons was used to synchronize clocks back in the 19th century. 

But one pretty weird for people who are not astronomers or play lots of KSP  is that we can not predict an comet like halley's comet more than an couple orbits. 
Reason is simple Pe is very high compared to Ap. Tiny changes to velocity because of n-body interaction will have much larger effect on Pe and therefor orbital period who changes how it interact with other bodies on the next pass so the uncertainty increase.

I think people just want some Lagrange points.

[whatsEJstandfor]

6 hours ago, GoldForest said:

Patched conics, except for Rask and Rusk. The twins get N-body, iirc. 

I don't think it's true n-body, but more of a restricted 3-body solution, since that could be solved for directly, so it wouldn't have the chaos of a general n-body solver. I also extremely think that Rask and Rusk are going to be just one (two?) of several bodies with custom solutions like this. I would bet all my remaining organs that, if Rask and Rusk were the only bodies with unique physics, they wouldn't have been in the trailer. IG would have saved that as a surprise for the full release.

[LHACK4142]

4 hours ago, Vl3d said:

I think people just want some Lagrange points.

I wonder if you could hack in some decently realistic Lagrange points even with patched conics.

[Pthigrivi]

49 minutes ago, LHACK4142 said:

I wonder if you could hack in some decently realistic Lagrange points even with patched conics.

I ran mini-soi’s at lagrange points by our resident programmers and they said its probably okay but they did have some issues. Maybe smarter people are more informed? 

[Superfluous J]

I still don't see why you can't set up a small sphere at L4 and L5 as if there was a tiny planet there, with enough "gravity" to keep something there in a little orbit.

To fix the problem where you're essentially creating a black hole, put a 2nd patched conic in there with 0 or negative gravity, half the size of the Lagrange patched conic.

Other than confusion when you pass through one unintentionally, I see no down sides to this idea.

[Bej Kerman]

1 hour ago, Superfluous J said:

I still don't see why you can't set up a small sphere at L4 and L5 as if there was a tiny planet there, with enough "gravity" to keep something there in a little orbit.

To fix the problem where you're essentially creating a black hole, put a 2nd patched conic in there with 0 or negative gravity, half the size of the Lagrange patched conic.

Other than confusion when you pass through one unintentionally, I see no down sides to this idea.

I don't see how this even approaches realism.

[Kalessin1]

BTW anybody has idea why they choose 24 February? Is this important date in history of space conquer?

[obney kerman]

2 hours ago, Superfluous J said:

I still don't see why you can't set up a small sphere at L4 and L5 as if there was a tiny planet there, with enough "gravity" to keep something there in a little orbit.

To fix the problem where you're essentially creating a black hole, put a 2nd patched conic in there with 0 or negative gravity, half the size of the Lagrange patched conic.

Other than confusion when you pass through one unintentionally, I see no down sides to this idea.

The issue is... with patched conics, lagrange points aren't really necessary? The main appeal of them is gravitational stability, which we have by default in all orbits, so... why would we need a complicated and difficult system like this?

[GoldForest]

6 hours ago, whatsEJstandfor said:

I don't think it's true n-body, but more of a restricted 3-body solution, since that could be solved for directly, so it wouldn't have the chaos of a general n-body solver. I also extremely think that Rask and Rusk are going to be just one (two?) of several bodies with custom solutions like this. I would bet all my remaining organs that, if Rask and Rusk were the only bodies with unique physics, they wouldn't have been in the trailer. IG would have saved that as a surprise for the full release.

Well, we know we're getting more systems after To Be Announced. I'm sure they have a Binary star system they want to add. Maybe even a trinary star system. I know those are extremely rare, but would be nice to have. So the N-body physics won't just be for Rask and Rusk if they do add either of those. 

It also depends on how well the hack or fakery is. If it stands up to scrutiny, then they might use it again. If it is more pain than it's worth to expand it on to other celestial bodies, they may abandon it and leave Rask and Rusk as the only planets with semi N-body physics. 

And yes, I know it won't be true N-body, I just don't put "fake N-body" or "N-body hack."

Hmmm. I wonder how hard it would be to get into a figure 8 orbit around the twins. 

[Bej Kerman]

8 minutes ago, GoldForest said:

Hmmm. I wonder how hard it would be to get into a figure 8 orbit around the twins. 

It'd be impossible. Even if you started in a perfect figure-8 orbit, the limited accuracy that the game can describe the position of your ship with will cause the orbit to fall apart soon after establishing it.

[GoldForest]

30 minutes ago, Bej Kerman said:

It'd be impossible. Even if you started in a perfect figure-8 orbit, the limited accuracy that the game can describe the position of your ship with will cause the orbit to fall apart soon after establishing it.

So, a natural figure 8 would break down. So, if you want to stay in the figure out, you'd basically have to use RCS or the engine to make the craft switch from one planet to another.

[Superfluous J]

3 hours ago, Bej Kerman said:

I don't see how this even approaches realism.

It helps simulate an observable phenomenon.

2 hours ago, obney kerman said:

The issue is... with patched conics, lagrange points aren't really necessary? The main appeal of them is gravitational stability, which we have by default in all orbits, so... why would we need a complicated and difficult system like this?

They're not necessary, of course. The game is perfectly playable without them. However, you cannot put a ship in one of those spots in the base game and have it experience the stability that exists in real life. If you edit your save file or use the set orbit tool maybe.

I'm not saying it should be in the base game or anything. I'm just wondering with all the people who want it, why no modder (who also wants it) has tried.

(for what it's worth I tried to do it once and failed miserably)

[Bej Kerman]

25 minutes ago, Superfluous J said:

4 hours ago, Bej Kerman said:

I don't see how this even approaches realism.

It helps simulate an observable phenomenon.

Of course it does, but I'm still not seeing how it does it in a way that's not wildly inaccurate.

[Vl3d]

Hey guys, I think you should check out my thread in the KSP2 suggestions about parallel-sequential missions. I may have stumbled upon the mother of all KSP2 innovations. I think we're in for a very big surprise.

Edited February 2, 2023 by Vl3d

[tstein]

12 hours ago, Vl3d said:

I think people just want some Lagrange points.

naaa the fun part is the  composite orbits.  You can  make some nefty tricks  of changing your orbit for much much reduced Dv.  Lagrange points can be implemented without n-body You just need artificial  SOI on vacuum aroudn those points. The shape of the "points"orbit will be wrong although.

4 hours ago, Bej Kerman said:

It'd be impossible. Even if you started in a perfect figure-8 orbit, the limited accuracy that the game can describe the position of your ship with will cause the orbit to fall apart soon after establishing it.

O woudl nto think so. I  implemented an N-body  simulation   just for  fun after the last discussion in the topic and   could   easily put things in lagrange points that  lasted   quite a lot of orbits (in real life they are NOT eternal  either. And I used no  advanced tricks at all just  Float 64 and  position in cetimeters, mass in grams of  real  universe sun, earth and moon and data from a site I found with specific  and clear data of the position and speed needed. It is NOT hard, once again  N-Body is not a hard problem it doe snto demand more precision than patched conics.  All its problem is related to  PLANING AHEAD and the  huge gameplay impact that the lack of long term predictions create.

[The Aziz]

There were dev talks about semi stable lagrange points between the twins.

[GoldForest]

5 minutes ago, The Aziz said:

There were dev talks about semi stable lagrange points between the twins.

I wonder if one of those points is right where the twin's SOI's meet. 

[Bej Kerman]

29 minutes ago, GoldForest said:

I wonder if one of those points is right where the twin's SOI's meet. 

A craft situated between them would quickly fall to either side of the system as the point where their gravity is balanced is infinitesimal.

[The Aziz]

Fair enough but lagrange points aren't exactly points. If there's a possibility that a craft can orbit around a small sphere simulating somewhat even pulls from R&R, I'll take it.

Quote from a long time ago (nevermind the typo, it was like that)

 "in the case of Rask and Rusk, we’ll be calculating the gravitational pull of multiple bodies on our Kerbal vessels, so that developing a stable orbit in complex conditions like a binary planet system becomes a new and exciting challenge! In addition, attempting a landing on Rask or Rusk will be a different experience depending on the location of the sister planet in relation to your target for touchdown, and yes, there will be an astable Lagrage point between the two planets (if we pull this off correctly)"

[tstein]

35 minutes ago, Bej Kerman said:

A craft situated between them would quickly fall to either side of the system as the point where their gravity is balanced is infinitesimal.

Yes and no... the lagrange points are very missnamed making people think it is a place where you balance in top of a needle. They are local minima or local maxima (in fact cell points in the function) potential and kinetic energy wise  (that is why you have L4 , L3 and L5 even  when these are not aligned in direct counter forces) , so   you tend to get dumped into  energy function minimum around them and get a lots of comes and goes until you  can be finnaly pulled out of them and while on that process minimal dV adjustments can  return you  to the deepest point in the local energy  function, the result is that  they are SEMI stable.

Edited February 2, 2023 by tstein

[Bej Kerman]

3 hours ago, tstein said:

4 hours ago, Bej Kerman said:

A craft situated between them would quickly fall to either side of the system as the point where their gravity is balanced is infinitesimal.

Yes and no... the lagrange points are very missnamed making people think it is a place where you balance in top of a needle. They are local minima or local maxima (in fact cell points in the function) potential and kinetic energy wise  (that is why you have L4 , L3 and L5 even  when these are not aligned in direct counter forces) , so   you tend to get dumped into  energy function minimum around them and get a lots of comes and goes until you  can be finnaly pulled out of them and while on that process minimal dV adjustments can  return you  to the deepest point in the local energy  function, the result is that  they are SEMI stable.

GoldForest was referring to the point directly between Rask and Rusk. So yeah.

Now to discuss simulating lagrange zones as small SOIs as a placeholder, well it's stupid. But I think there's a clear middleground that'd make everyone happy, continue using rails for most orbits but enable multiple-body gravitation for vessels within zones surrounding the lagrange points. AFAIK, the game should easily be able to figure out where the lagrange zones are for each body and moon. You still get orbital stability but vessels also act in a realistic manner when in proximity to a lagrange point and can hold realistic orbits. Maybe if the player lets a vessels' RCS drain, the game will lock its orbit and keep it in a fixed halo so a player can leave a vessel in a halo orbit without needing to manually keep it there. Sounds similar to SOIs, yeah, but it isn't; the kind of fixed orbit I propose would use an already-established halo orbit as a basis rather than a weird Keplerian orbit around a black hole that doesn't behave as an orbit should when it's being constantly perturbed to form a halo around an imaginary point.