Disappointed in the physics.

[Setu]

I'm new to Kerbal, just got first flight to the Mun. On the way back I got a nice trajectory towards the mother planet, but when I got in the SOI of the planet, my trajectory turned 90 degrees! I've tried to find out what happened, and I'm told this is because the three body physics is too hard to solve - but surely none is involved? Three body physics is where the path of body A is affected by B & C, but the path of B is affected by A &C, and the path of C is affected by A & B, so any tiny change in starting conditions rapidly changes the outcome,  ( and even then that doesn't mean its incalculable, only that we cannot know the starting conditions accurately enough to calculate anything useful - which of course  is not true in a simulation), anyway, the changes in the paths of the moons and planets is surely ignored in Kerbal, so this not a three body problem. Maybe the physics is harder than it looks, but reducing it to a series of one-body problems with a step change between zones, seems to me to be an over simplification?

What other things are over simplified?

[steuben]

fuel flow
ISP
general structural design, both static and dynamic
science
manufacturing
research and development
commuincations
hiring

Now, to the question at hand.

Define ninety degree turn.

But my guess is that what you saw was an apparent change of trajectory based upon the change of frame of reference caused by the change of the SOI. 

 

[Sitting Duck]

6 hours ago, Setu said:

when I got in the SOI of the planet, my trajectory turned 90 degrees!

Are you sure you aren't confusing the NavBall readout vs actually making an (impossible) 90 degree turn in space at speed, relative to the previous body's SOI and your current trajectory? The NavBall display flips when you cross SOI boundaries, which is perfectly normal.

 

[Ultimate Steve]

1 hour ago, Sitting Duck said:

Are you sure you aren't confusing the NavBall readout vs actually making an (impossible) 90 degree turn in space at speed, relative to the previous body's SOI and your current trajectory? The NavBall display flips when you cross SOI boundaries, which is perfectly normal.

 

Yes. The navball orients to up and down based on what body's sphere of influence the craft is in. When you cross from one sphere of influence to another, up and down usually change.

Your trajectory on the map might also make it seem like you are making a turn. However, depending on what mode you have your patched conics predictions set to and which body you are focusing on, your trajectory will be shown relative to either the body you are around or the body you are on your way to. It can also change from the former to the latter when you cross the boundary.

Your total speed and direction remain the same, but it is just being displayed relative to a different body.

Unless you are experiencing a bug, but this is unlikely.

There was an older bug where if you crossed a sphere of influence at high time warp you could get flung off in a random direction because the game couldn't simulate fast enough, but this has been mostly fixed for a while.

[winged]

8 hours ago, Setu said:

I'm new to Kerbal, just got first flight to the Mun. On the way back I got a nice trajectory towards the mother planet, but when I got in the SOI of the planet, my trajectory turned 90 degrees! I've tried to find out what happened, and I'm told this is because the three body physics is too hard to solve

No, it didn't and it has nothing to do with n-body physics. You have just switched to a different frame of reference. Your velocity is now being displayed relative to Kerbin (previously it was being displayed relatively to the Mun).

[KerikBalm]

pictures would help, but I suspect this is simply a frame of reference change, and there's actually no problem

[Setu]

Why do I think I turned ninety degrees? Is it due to frame of reference? Possibly. I'm judging from the map. By eyeball I had a trajectory directly at Kerbin, and then suddenly I was in an orbit around Kerbin, 2/3rds of the way to Mun.

Ok I can see that it is frame of reference phenomenon, thanks. Like trying to throw a ball into a bin from a moving car, you have throw it backwards more than sideways, so a trajectory from Mun towards Kerbin will miss it and go into orbit.

[kerbiloid]

And btw that's the only purpose of the SOI at all, to switch the orbit shape on the reference body switch.

Edited November 10, 2019 by kerbiloid

[ezaroo]

15 minutes ago, Setu said:

Why do I think I turned ninety degrees? Is it due to frame of reference? Possibly. I'm judging from the map. By eyeball I had a trajectory directly at Kerbin, and then suddenly I was in an orbit around Kerbin, 2/3rds of the way to Mun.

Ok I can see that it is frame of reference phenomenon, thanks. Like trying to throw a ball into a bin from a moving car, you have throw it backwards more than sideways, so a trajectory from Mun towards Kerbin will miss it and go into orbit.

Yep, you just have a lot to learn on orbitals mechanics  don’t worry everyone starts the same! It’s not entirely intuitive, go watch some videos by KSP youtubers, Scott Manley has good videos explaining things, they are a bit old now but still totally valid the UI will just look a bjt different.

But the short reason is outside of some very specific situations accelerating directly at something is the least efficient way to get to it in space (because you’re travelling in a circular path).
 

To get back from to Kerbin from the Mun you do not want to point at Kerbin from the mun - that will as you noticed leave you lower than the Mun on one side and higher on the other - you want to point the opposite direction to your motion around Kerbin awhile accelerating - this means if you’re orbiting with the muns rotation you want to accelerate prograde on the Kerbin facing side of the mun or if you’re orbiting against the muns rotation you want to accelerate prograde on the side facing away from Kerbin. 
 

In both cases you are accelerating 90 degrees away from where Kerbin is. 

 You’ll get the hang of it! And yes changing reference frames can make things like weird but you have continued on the exact same path.

[terryna]

19 hours ago, steuben said:

fuel flow
ISP
general structural design, both static and dynamic
science
manufacturing
research and development
commuincations
hiring

Now, to the question at hand.

Define ninety degree turn.

But my guess is that what you saw was an apparent change of trajectory based upon the change of frame of reference caused by the change of the SOI. 

 

Nice

[pandaman]

Yes, as others have mentioned.  The change of direction is an illusion caused by changing the body you are in the SOI of.

If you look at your initial orbit path (blue trail) it will show your orbit as it is now and you may see another coloured path that shows what it will be, relative to the new body, when you leave your current SOI and enter the new one.  Once you cross the SOI boundary the 'new' path turns blue as it is now the current one.  Yout direction hasn't changed, just your point of reference.

A bit like multiple cameras showing the same car from opposite sides of the road.  One will appear to move from left to right the other right to left, so if you switch cameras the car will seem to swap dirrctions.

Edited November 10, 2019 by pandaman

[Guest]

If it makes you feel any better @Setu, I was incredibly confused about this exact thing when I first started playing KSP. It is not noob-friendly when it comes to explaining orbital mechanics (or indeed much at all).

Just wait 'til you get to the point where you need to learn to RV and dock. Took me days and I was screaming in frustration. Hadn't found these forums yet, either...

[Guest]

deleted

Edited March 29, 2020 by Guest

[Snark]

Moving to Gameplay Questions, since it looks as though this is an issue about "how to play the game". 

[Kerbart]

On 11/9/2019 at 5:28 PM, Setu said:

Maybe the physics is harder than it looks, but reducing it to a series of one-body problems with a step change between zones, seems to me to be an over simplification

Not for NASA. You’re aware that this how they plan their interplanetary missions?

[Jestersage]

On 11/10/2019 at 1:47 PM, <noname> said:

• Conic Patch Draw Mode
  • Available Settings: Local to Body, Local at SOI Entry, Local at SOI Exit, Relative, Dynamic
  • Default: Relative

So what are the difference, and which one is "the best"?

[HansAcker]

10 hours ago, Jestersage said:

So what are the difference, and which one is "the best"?

As <noname> said, "Best" depends on what you prefer.

 

[Jestersage]

1 hour ago, HansAcker said:

As <noname> said, "Best" depends on what you prefer.

 

Thanks. However, that one, being made in 2012, doesn't explain "dynamic", unless it corresponde to mode 4 "Lerped mode"?

And default=relative is mode 3, right?

Edited November 12, 2019 by Jestersage

[Guest]

deleted

Edited March 29, 2020 by Guest

[Guest]

deleted

Edited March 29, 2020 by Guest