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

[chaos_forge]

Like this?

Horn method:

HemiSOI method:

Original SOI method:

That actually does help a lot in seeing the error, good idea xD Now if I can get vector graphs to work . . .

EDIT: I don't like the way it handles the colors though, I'm gonna try and see if I can modify that.

EDIT2: corrected it, ignore the above comment

Edited July 17, 2013 by chaos_forge

[tntristan12]

Horn's method definitely looks the most viable. I wish somebody could mod this into the game to see what it's like.

[chaos_forge]

Or that we could get some feedback from the devs, developer-on-leave, or even a mod. I'd just like to know that they've seen the suggestion.

[tntristan12]

In the mean time, I'd really like to see how Horn's model handles existing situations like the Jool system or Kerbin-Mun-Minmus. Then, what if there are two bodies of comparable size, and a third small moon orbiting them? What about 3 similarly sized planets? For example: two in one part of the sphere and one in the other.

[leax256]

Or that we could get some feedback from the devs, developer-on-leave, or even a mod. I'd just like to know that they've seen the suggestion.

that's your cue nova.

[Horn Brain]

In the mean time, I'd really like to see how Horn's model handles existing situations like the Jool system or Kerbin-Mun-Minmus. Then, what if there are two bodies of comparable size, and a third small moon orbiting them? What about 3 similarly sized planets? For example: two in one part of the sphere and one in the other.

No system proposed can "handle" three similar-massed bodies in a realistic way. For Jool or Kerbin-Mun-Minmus, the primary (Jool or Kerbin) is large enough that even if you implemented my system it would look very much like the current system, except there would be VoI boundaries between the "inner" region and the "outer" region for each pair of bodies that essentially did nothing. That's why I think we should leave any planetary system with mass ratios above Duna/Ike levels alone and use the current system, since implementing my system there would only add extra VoI transitions which didn't act very differently at all from the current system. It would just be confusing for no reason.

If (for an example of how this would work, I'm not suggesting this)you made the Mun much more massive, you could implement my system for the Kerbin-Mun pair, since Minmus would be in the outer region and would just orbit the barycenter with no difficulties. You could also (although this is probably unrealistic from a stability perspective) have smaller moons orbiting either the beefier Mun or Kerbin such that they stayed within the VoI of one or the other. Think of Minmus in low Kerbin orbit.

Another hypothetical: What if we gave one of the forthcoming new gas giants a binary moon system? It could have two moons of equal mass orbiting one another very closely as their barycenter orbited the gas giant. As you approached this binary mun from orbit around the gas giant, you would first transition to the binary mun's barycenter's SoI, which would be determined exactly like the current system using the mass and semi-major axis of the binary pair's orbit about the gas giant to determine it's size. Then as you got closer, you would transition to the "inner" region of the binary and my system would be used to determine which VoI you were in.

Yet another, even more wacky proposal would be to have one large body and two smaller ones - say Duna and two Ikes. If you put both Ikes in orbit around Duna separately, you could use the current system and it would be ok. But imagine placing the Ikes in a binary! Their combined mass would make the mass-ratio about 8:1 (Duna to Ike+Ike), which would be perfect for using my system. So on an incoming trajectory from interplanetary space you would enter the Duna/Ikes barycenter VoI first, then (assuming you're going to aerobrake at Duna) you enter the inner region of the Duna/Ikes system on the Duna side and you are only affected by Duna's gravity. You aerobrake to be captured but your leave your apoapsis high enough that you enter the Ikes side of the Duna/Ikes inner region and enter the VoI of the Ike/Ike barycenter object. Then, as you approach one of the Ikes, you enter its inner region and solo VoI. That would certainly be confusing, and the conic patch draw limit had better be set pretty high or it will be hard to see what's going to happen next, but it would work perfectly well in my system.

So basically, you can do whatever you want with my system so long as you have at most two mutually-orbiting objects (barycenters included) in any given system which share a given order of magnitude of mass.

[chaos_forge]

Horn Brain is exactly right with, the small exception that it *would* be possible to have 3 objects of equal mass, if they follow an orbit like this: http://www.atlasoftheuniverse.com/orbit3.gif (essentially two stars orbit each other, and the third orbits the close pair from a significantly larger distance). That configuration would work exactly as you'd expect it to and as we've discussed before.

I proposed the system mainly to be able to implement multiple star systems in a way that made sense, but having a double planet in KSP would also be pretty cool. I think the double moon for the gas giant would have to be relatively far away from the gas giant (> pol's semimajor axis) for it's stability to makes sense from a real-life physics standpoint.

[Captain Sierra]

This would be very interesting to see as a physics expansion. It would definitely create interesting orbital mechanics. I wonder, with the barycenter basically being a Lagrange point (where two object's gravitys cancel to zero), could I thread the needle and use that as a gravitational slingshot? Double power?

[tntristan12]

This doesn't look like it would allow for legrange points as it isn't a solution to N-body physics. Actually it is more of an expansion on the existing patched comics system. It's probably not going to be implemented in the stock game, but might be made into a mod if the framework allows modders to go in that deeply.

[CaptainKipard]

This doesn't look like it would allow for legrange points as it isn't a solution to N-body physics. Actually it is more of an expansion on the existing patched comics system. It's probably not going to be implemented in the stock game, but might be made into a mod if the framework allows modders to go in that deeply.

I hope you're not speaking for the devs because that wouldn't be very smart. I don't know how things will develop but right now there's nothing that would give players an intermediate challenge. We have the Kerbol system, and then suddenly we have talk of FTL drives and flying to other systems in a matter of days or something. Where's the middle ground? Well it's right here in the form of an expanded star system. (Actually it's here)

[chaos_forge]

As far as FTL is concerned, I'd rather not have it in KSP (a DLC would be fine though). I like my sci-fi hard. Plus Orion Drives are so much more Kerbal than FTL. With an Orion Drive, traveling around a multi-star system starts looking much more feasible.

This would be very interesting to see as a physics expansion. It would definitely create interesting orbital mechanics. I wonder, with the barycenter basically being a Lagrange point (where two object's gravitys cancel to zero), could I thread the needle and use that as a gravitational slingshot? Double power?

With the original system you could get ridiculous gravitational slightshots, but that's physically unrealistic, so that's why we came up with the fixed system. And this system does not model any of the behavior associated with Lagrange points, so don't get your hopes up.

[tntristan12]

I hope you're not speaking for the devs because that wouldn't be very smart.

I never presume to speak for the devs, but I've been around long enough that I can be reasonably confident in predicting what they're likely to say.

[Horn Brain]

I never presume to speak for the devs, but I've been around long enough that I can be reasonably confident in predicting what they're likely to say.

He's right. The devs probably saw "N-Body Physics" in the title of this thread (if they saw it at all) and ignored it. They've heard so much nonsense about N-body stuff I don't blame them for assuming that this thread is useless. I really want to see a double moon, though. Or at least Pluto/Charon with a lower mass ratio.

[CaptainKipard]

He's right. The devs probably saw "N-Body Physics" in the title of this thread (if they saw it at all) and ignored it. They've heard so much nonsense about N-body stuff I don't blame them for assuming that this thread is useless. I really want to see a double moon, though. Or at least Pluto/Charon with a lower mass ratio.

O.O So that's like judging a book by half of a cover. Surely they're not that stupid. If they have any sense at all they'll look at the 5-star rating and number of pages at least. And if Skunky is doing his job he'll notice the unusually high quality of this thread and pass this suggestion on.

[tntristan12]

I think the best we can hope for is that the mods are reading this thread. Probably right now. Especially since N-Body Physics is on the title, and the do-not-suggest list. They're just waiting for any excuse to shut this thread, so let's keep it on topic, shall we?

I would also recommend changing the title since we are no longer talking about the titular "barycenter object," and have instead opted to focus on non-spherical volumes of influence.

Anyway, can somebody who has more modding experience than I say whether or not this is within the scope of the SDK, and can in fact be modded into the game? Or are we wholly reliant on the devs implementing it into the stock game?

[chaos_forge]

I would also recommend changing the title since we are no longer talking about the titular "barycenter object," and have instead opted to focus on non-spherical volumes of influence.

That's probably a good idea. How do I do that? Also I should probably summarize the last 12 pages of discussion in the first post.

[CaptainKipard]

I think you just edit the original post and go to advanced. It should allow you to change the topic.

Also the barycenter VOI is still there isn't it?

[Horn Brain]

I think you just edit the original post and go to advanced. It should allow you to change the topic.

Also the barycenter VOI is still there isn't it?

Yeah, that's the outer region of my model.

[chaos_forge]

Yeah, that's the outer region of my model.

The model. The idea doesn't belong to anyone, we're all helping refine it.

In other news, I finally got the vector plots to work, so just let me know if you want to see one.

EDIT: and yeah, the outer barycenter region is still there

Edited July 22, 2013 by chaos_forge

[Kerbal01]

So the barycenter is kinda like a Lagrange Point?

[tntristan12]

Not at all. The barycenter and L1 point are only the same if both planets are of equal mass. The barycenter is, by definition, the center of mass of the two planets. What the initial suggestion (which we have since abandoned) was, was to have an invisible SOI that would have just the planets orbiting it. The idea has since become more about non-spherical SOIs, to try to implement double body systems. This would still not allow L points as it isn't a real way of implementing N-body physics.

[chaos_forge]

changed the title and added a summary of the current progress to the first post. Also, Horn Brain, I used the chart that you posted to show the shape of the non-spherical VOI's because it was easier than making my own, but if it's not okay with you I can remove it.

[tntristan12]

First post looks good. I hope someone who can contribute more usable information on the concept contributes!

[CaptainKipard]

Good summary. You should explain what VOI means under "terminology"

Also (in the title) Infulence is spelled Influence.

Edited July 23, 2013 by Cpt. Kipard

[chaos_forge]

Good summary. You should explain what VOI means under "terminology"

Also (in the title) Infulence is spelled Influence.

Whoops, thanks for pointing out that typo xD it should be fixed now.