45

Kerbal Space Program is an independent spaceflight simulation game, which has become quickly popular due to being (kind of) precise at simulating actual spaceflights.

But how precisely? How close is Kerbal Space Program to reality in the aspect of difficulty of flights?

NattyBumppo
  • 164
  • 4
Zoltán Schmidt
  • 5,223
  • 7
  • 34
  • 69

3 Answers3

48

Kerbal Space Program is somewhat of a medium fidelity simulation. It manages a few things quite well, and a few things not as well. Let me try and give a list (which might be a bit out of date):

The Good:

  • The orbit simulation is quite accurate, including how to change inclination, raise/lower orbits, leave a planet, and approach a new planet.
  • The staging is somewhat accurate.
  • The use of fuel, acceleration, mass, etc. are pretty accurate.
  • Structural stability is pretty accurate, at least modeling the key components.
  • The new Aerodynamic's model is pretty good. It doesn't model everything perfectly, but is pretty good overall.

The Bad:

  • Only one body affects an orbiting object, the object of most influence. It might even do something like a Hill sphere. However, there are plenty of multi-object systems which aren't managed at all correctly; Lagrange points aren't taken into account, for instance.
  • The aerodynamics model doesn't model well things like breaking the sound barrier.
  • Reaction wheels aren't modeled properly (Can do with a mod, but...)
  • A number of effects aren't really modeled, but can be included with mods (Communication delay, rocket fuel settling, etc)
  • Re-entry heating is now modeled, although it isn't great.

The Ugly:

  • No life support requirements.

For more, see Wikipedia.

PearsonArtPhoto
  • 121,132
  • 22
  • 347
  • 614
  • 3
    Just wanted to add the aerodynamics, life support and jet engine issues can actually be improved with some mods. Nothing for multi-body systems yet (as far as I know). –  May 10 '14 at 16:18
  • People are working on that. – Zoltán Schmidt May 10 '14 at 16:20
  • 9
    Also, there are no random part failures (so redundancy is less important and you don't have to worry about complexity decreasing reliability), and rockets have full throttling capability and can be restarted infinitely many times. – cpast May 10 '14 at 19:30
  • Besides engines, there are no other external forces, which can exert a force on a ship, other than one gravity body and (smiple) atmospheric drag/lift. So no tidal forces, solar pressure, ect. – fibonatic May 10 '14 at 22:46
  • 1
    and do remember that the game is still in beta, aerodynamics are on the list of things to improve. Multi body physics might be, but may well have been left out to improve performance. List support is I think deliberately left out to reduce complexity and increase playability. – jwenting May 13 '14 at 07:17
  • The current release 1.0 did a complete overhaul of the aerodynamics system. It is now a lot more realistic. – Philipp Apr 29 '15 at 14:58
  • Updated to support the 1.0 release. It's about time the aerodynamics system is updated;-) – PearsonArtPhoto Apr 29 '15 at 15:36
  • 1
    I must object on the re-entry heating being modeled well. I'm playing 1.2.2 as I write this, I've got an early Mun rocket coming home. It's now made two passes through the fire tail first, burning off half the heat shield. I decided to try going nose first and while things are getting hot nothing is in danger. Obviously on the pass where I actually come out of orbit I will have to be tail first but I can aerobrake this way without burning my heat shield. – Loren Pechtel Dec 20 '16 at 05:13
  • Followup to yesterday's comment: I lost count of the number of passes it took to bleed off the speed but it was something like a dozen. The game had no problem with it although in the real world it would have been catastrophic--a battery pack up front was routinely running 700K and the cockpit core temperature exceeded 600K on the last pass before I came down. – Loren Pechtel Dec 20 '16 at 22:54
  • Sub vs supersonic modelling is improved in 1.2.The supersonic drag modelling is a bit different to subsonic, for example at (very) subsonic speeds drag caused by a leading and trailing blunt surface is identical, but at transonic speeds the leading surface becomes a lot more important with a blunt leading surface being much more draggy than a blunt trailing surface. I think this is meant to represent eddy formation at subsonic speeds and flow separation at supersonic speeds. So it's no longer true to say it doesn't model breaking the sound barrier, though it might do so a little poorly. – Blake Walsh Dec 21 '16 at 10:41
  • It's basically the "fun and exciting" part of space exploration with some bits kept authentic for challenge, and other parts simplified to gloss over the "make a single mistake and you either die or are never seen again" part of space exploration. – Omaha Dec 21 '16 at 15:50
  • There are plenty of you die and are never seen again moments in the game, but that's part of what makes it interesting. – PearsonArtPhoto Dec 21 '16 at 16:27
  • @cpast I disagree, yes there are no *random* failures, but you won't catch me putting one antenna on a lander. I'll always put 2 #NoRevertNoSaveLife – Magic Octopus Urn Mar 30 '18 at 21:34
14

One thing that is different is the lack of orbital decay. In kerbal, if you get above 68km, you will stay there forever. In real life, even extremely high orbits are subject to decay, since our atmosphere is much more complicated than KP can model.

Things like the ISS have to constantly make adjustments to stay in orbit. Things like solar flares have a tendency to spew atmospheric particles up into higher orbits and make orbiting objects slow down.

I think the game is definitely much easier with its simplified atmospheric model, and I for one am glad to not have to worry about my orbit changing once I achieve one.

Community
  • 1
captncraig
  • 241
  • 2
  • 5
  • 2
    When not time-warping orbits will decay due to the numerical integration which is used. However the opposite side of the orbit will actually increase while the near side decreases. So by switching time-warp on and off on the right time allows you to make and orbit more or less eccentric. – fibonatic May 10 '14 at 22:54
2

A couple of other things:

Kerbal is very nice about parts working. In real life, parts fail, fuel boils off after being stored for so long, electronics short out, rockets can only throttle so low, and hardware slowly decays from continuous exposure to solar UV and cosmic rays. It can be a very slow process, but it's a very there process.

Second, someone on reddit pointed out that in real life, Asparagus staging isn't really practical because the thrust to weight ratios of rockets and the wet-dry mass ratios of fuel tanks in real life are much higher, so there's more penalty for having dead weight in KSP, but more margin with part performance. In real life--parts can easily be destroyed by off-nominal launches, and don't have much margin for failure. Titan rockets couldn't even stand up under their own weight without being fueled up.

Michael Cox
  • 21
  • 1