Misleading patched conics

[tuguley]

Don't get me wrong, I love the patched conics system as much as everyone else does. But this would even make Jeb scream.

I think my conics mode was 3

with a conic limit of 5

http://imgur.com/a/YEgl3

But quick thinking by Shepger Kerman saved the ship, which is now sitting in polar orbit of Ike.

[Kiershar]

When you're past 3 orbits calculations you can only rely on its apoapsis reading. I'm guessing that path you see in purple is what your orbit looks like to an observer watching you from the gravity source of your current orbit.

[Pyre]

It's not misleading, it's just confused. Seems to happen whenever you are 2 or more SoI changes from a solar orbit.

[Pingonaut]

Does it happen with every conics mode or just 3? I might reconsider using that mode.

[0jam3290]

Uuhhhhhhhh........ I believe that has happened to me in .16 with unaltered patched conics, but thats just crazy.

[Yeomans]

I have never personally had this problem, but I do understand (for the most part) why it occurs.

KSP does not currently support n-body math. This means it has issues understanding how 2 or more bodies with significant influence affect another body, or each other. Because of this, when you are using conics in situations such as above, you get funky patterns like this. That same reason explains why there are currently no Lagrangian points in KSP.

[vexx32]

No, that's not the case. It's not that KSP doesn't understand how more than one body affects a craft, it actually doesn't care. In other words, at the basic game level, n-body physics is ignored entirely. Gravity is calculated only with reference to the body you're currently orbiting. Thus, the patched conics will have nothing to get confused over to start with.

As such, I have no idea why this happens, since it's predicting a physically impossible trajectory, n-body physics or not.

[SteevyT]

It is your orbit drawn with whatever body you are orbiting as the reference point. Nothing wrong with it, just looks really bizarre.

[tuguley]

The purple orbit conic was supposed to be predicting a solar orbit. (I moused over the AP, PE icons and they said Sun apo / peri).

I am using conics mode 3.

I may be wrong, but I do remember harv commenting that conics 3 was still buggy and under development and that's why it wasn't the default. (current default being mode 1)

I have previously noticed other garbled patched conics with mode 3. But those were on very complex paths (very high speeds, very low altitude, with 4 SOI changes).

But this was just plain surprising and very comical. I guess there is a reason mode 1 is the default.

Here is another weird mode 3 conic: http://i.imgur.com/nP0SR.png

I was attempting an aerocapture on Jool, after the Mun interfered with my orbit and changed my velocity.

[SteevyT]

I know in the original post the purple chunk is in the planet's reference frame even for the solar portion of the orbit.