Barycenters and non-spherical Volumes of Influence (an idea for binary planets/stars)

[Horn Brain]

Would this also affect crafts on rails?

No, it wouldn't. Everything would be fine. This would let you give the barycenter the proper mass and let the spacecraft treat it like a regular body (even though it's invisible).

This works, gentlemen.

[chaos_forge]

This works, gentlemen.

Seems so. Of course, without knowing the internal structure of the game it's hard to tell how hard this wuould be to implement. Whether it would be a one-update project or take longer . . . that'll most likely decide whether it's included or not.

[NeilC]

Seems so. Of course, without knowing the internal structure of the game it's hard to tell how hard this wuould be to implement. Whether it would be a one-update project or take longer . . . that'll most likely decide whether it's included or not.

Just the orbital period as a stored parameter? That's a pretty minor refactoring of code, impact would be pretty minimal and should be easy to test. There aren't too many ways to implement a keplerian orbit solver, pretty much "the right way" and "wait, what are you doing?" It's likely that the periods are calculated once and stored anyway, for performance reasons; we're just proposing storing them permanently.

All this VOI stuff, though, is non-trivial. It changes a fundamental assumption (all SOIs are spherical and parameterized by a single scalar), and it's possible that assumption is embedded in the SOI algorithm deeply enough that it would require major changes. Conceptually I know it seems simple, but it really depends on the data structures they're using and how they've optimized the sort that finds the closest body. That, we don't know enough about to comment.

[loknar]

No, it wouldn't. Everything would be fine. This would let you give the barycenter the proper mass and let the spacecraft treat it like a regular body (even though it's invisible).

This works, gentlemen.

Did I miss something or it does not work to replace 2 stars gravity effect with 1 barycenter for other objects in the system?

Off the top of my head, I would think if an object is in orbit that comes closer to the barycenter there are not in a stable orbit.

Edited August 8, 2013 by loknar

[NeilC]

Did I miss something or it does not work to replace 2 stars gravity effect with 1 barycenter for other objects in the system?

As long as the barycenter mass is the sum of the two star's masses, and the other objects are far enough away that they don't enter either star's SOI, then it will work fine.

It won't be 100% realistic because the pure-elliptical orbit won't be perturbed by any of the gravitational dynamics present in a real binary system. But it would work.

[loknar]

Oh the barycenter would be a 3rd SOI parent to each star's SOI?

I see. Yes it would "work"; I would be ok with it.

Sorry to ask - I should probably use the search function: Do we know if the game only considers the current SOI and none of the others?

[NeilC]

Sorry to ask - I should probably use the search function: Do we know if the game only considers the current SOI and none of the others?

What do you mean "considers"? You can only be in one SOI at a time. I wrote some discussion about how I think it works - traversing a tree - earlier in this thread. But that's just, like, my opinion, man.

[loknar]

What do you mean "considers"? You can only be in one SOI at a time. I wrote some discussion about how I think it works - traversing a tree - earlier in this thread. But that's just, like, my opinion, man.

I mean it seems trajectory calculations in game only use the current SOI. Am I correct?

That is slightly inaccurate even though the current SOI itself is moving within its parent(s). Example: having a stable orbit on the Mun is not realistic. (again, just my imagination of how physics work)

[NeilC]

Yeah, you should probably use "Search", or even "read this thread".

all on-rails trajectories are pure ellipses. No trajectory calculations, as such: just progress through a static Keplerian orbit.

all in-focus craft get physics solved, but G is just with respect to the current parent. there is no n-body.

[loknar]

oki thx

*sits down*

[Elros]

<...> We're not sure how deeply the constant-radius-sphere assumption is embedded in the orbital mechanics solver. Moving from spherical SOIs to rule-based VOIs is a significant change, and having to recalculate them every timestep is a bigger one. One measley calculation can affect performance if you have to do it often enough <...>

It only needs to be calculated when accelerating, excluding the effects of kepler's second law.

[chaos_forge]

It only needs to be calculated when accelerating, excluding the effects of kepler's second law.

"while accelerating" is a bit vague. Since the orbits are elliptical, the objects are always accelerating.

[Elros]

"while accelerating" is a bit vague. Since the orbits are elliptical, the objects are always accelerating.

*Using delta-v/Firing the engines. I wasn't sure how to say it in the post properly

EDIT: Kepler's second law is what says that it accelerates in an elliptical orbit.

Edited August 28, 2013 by Elros

[thereaverofdarkness]

Howabout arbitrary zones of influence engineered individually to best fit each particular multi-body system?

Here's an example for the Kerbin system:

In this example, the Kerbin influence region (ROI) is depressed from a sphere where the Mün is near it. The Mün would also be a little depressed in a similar fashion but I didn't depict it in this image. These ROIs would be static in shape but would rotate to face the other body at all times. This means it is possible to experience rapid ROI-swapping if you are on the border but moving with it, though unlikely as it won't be moving anywhere near orbital speeds. Minmus' ROI would just be a sphere, since it's far enough out there to be fine with such an arrangement.

The benefit of this structure is that it allows craft to swap into Kerbin-Mün barycenter influence before transiting into Mün influence.

[tntristan12]

Holy creeping necro of necros, batman! I looked into this mod a while ago and saw no readily obvious way to change the general shape of spheres of influence (this would by why they are spheres). I saw no way of overriding the body of influence either. I'm pretty sure patched conics are gonna be the best way to do this.

[Sigma88]

I've read part of the thread and it gives a lot of interesting ideas.

I you are still interested in make binary work in KSP right now I will be happy to discuss how I did it and why I took some decisions rather than others.

I will need some time to read the whole tread tho.

[chaos_forge]

I've read part of the thread and it gives a lot of interesting ideas.

I you are still interested in make binary work in KSP right now I will be happy to discuss how I did it and why I took some decisions rather than others.

I will need some time to read the whole tread tho.

You made a mod that makes binary work in KSP? I would certainly be interested in hearing about it . And while I certainly still think many of the ideas expressed in this thread are good ones, I haven't been as involved in the KSP community lately as I was a year or two ago . . .

I also just realized that I have no clue how the KSP forums handle necro-posting. Should I start a new thread, or is talking in this one okay?

[Sigma88]

You made a mod that makes binary work in KSP? I would certainly be interested in hearing about it . And while I certainly still think many of the ideas expressed in this thread are good ones, I haven't been as involved in the KSP community lately as I was a year or two ago . . .

I also just realized that I have no clue how the KSP forums handle necro-posting. Should I start a new thread, or is talking in this one okay?

I started with making a binary system myself

a Pluto-Charon analogue that started intended as an OPM expansion but now is compatible also with Stock

then, since I've received a lot of positive feedback about the system, I decided to make a mod that will let everyone turn a Planet-Moon system into a binary system. (both links are in my sig)

If you are interested in anything specific feel free to drop by and ask as many questions as you like

[CaptainKipard]

I also just realized that I have no clue how the KSP forums handle necro-posting. Should I start a new thread, or is talking in this one okay?

It's your thread. I'm sure no one will complain if you continue. I remember the discussion here was very interesting. I don't see the point in starting a new one.