On Rails Physics.

[JackleOfSparta]

I was just wondering, what would happen if the planets and moons weren't on rails and could have legitimate physics?

[Tex_NL]

The game would probably grind to a halt due to the excessive CPU power needed to calculate it all.

[Boffinbox]

your computer would basically tell you NOPE and stop working. every time you apply any thrust to your ship, the whole solar system would have to be recalculated (because your spacecraft would also have to have "legitimate" physics)

[Brotoro]

I suspect the layout of planets/moons would not remain stable over time.

[Strategia]

The first thing that would happen is a challenge thread being posted for the shortest time to crash the Mün into Kerbin.

[Stealth2668]

More like a challenge thread to see how long your computer survives.

[MR4Y]

You would essentially need a cluster to play the game. Or one machine just to run the enviroment, other for the ships and a third one for the physics calculations.

[Strategia]

Well, I guess it depends on what you define as "legitimate" physics. Having full, 100% realistic physics, yeah, that'd screw over your computer something fierce. But the thread mentions the planets being on rails specifically. That should, I think, be easier to change. If the planets are treated as one single, spherical (or almost spherical) object, at orbital velocity, that'd require much less calculation than simulating them in the tiniest detail. And it'd allow players to do things like play Kerbol Pinball or otherwise rearranging the solar system.

[jfx]

Kerbol system is inherently unstable, planets would collide or be ejected from the system after some time. Someone did this in universe sandbox

[RawChicken]

Kerbin will end up with 15 moons...

[NovaSilisko]

The solar system would destabilize itself over time with no other gameplay implications. Kerbin weighs 52.91 quintillion tons. Even gilly weighs 124.2 trillion tons. You ain't movin' that.

[rkman]

It would probably work in real time (aside from problems with stability of the system), but it would be uselessly inaccurate at high timewarp.

[MR4Y]

Well, I guess it depends on what you define as "legitimate" physics. Having full, 100% realistic physics, yeah, that'd screw over your computer something fierce. But the thread mentions the planets being on rails specifically. That should, I think, be easier to change. If the planets are treated as one single, spherical (or almost spherical) object, at orbital velocity, that'd require much less calculation than simulating them in the tiniest detail. And it'd allow players to do things like play Kerbol Pinball or otherwise rearranging the solar system.

Not even the world's fastest super computers can simulate real life physics with 20% accuracy.

[jheriko]

Why does everyone believe that its impossible to calculate n-body gravity quickly? There are very few bodies and the calculations would be trivial. It would break all the gameplay but it certainly wouldn't strain your computer... it wouldn't even be a strain for a 20 year old computer for that matter.

If we are talking about modelling a pointlessly accurate gravity with GR and taking into account the shapes of planets etc. then yes it would be difficult, but using newtonian gravity and spheres - which is a fantastic approximation - is very practical.

The alternative, one used by astronomers and space programs is to actually calculate everything in advance and create rails that include all the small perturbations, like VSOP87 or DE206 etc. I know from experience (i.e. I have done it and can point at other software that does, e.g. celestia, stellarium... ) that these things can be very happily run at interactive framerates with small data files if you are willing to sacrifice accuracy a little...

Edited June 10, 2013 by jheriko
grammar

[protoz]

First thing i would do is attach 1 ion engine to each planet forcing them to thrust at each other, then i wait.

[Stealth2668]

First thing i would do is attach 1 ion engine to each planet forcing them to thrust at each other, then i wait.

You'd die of old age before you notice any change

[Sapphire]

The solar system would destabilize itself over time with no other gameplay implications. Kerbin weighs 52.91 quintillion tons. Even gilly weighs 124.2 trillion tons. You ain't movin' that.

That sounds like a challenge!

*turns on unlimited fuel*

*edits Mainsail parameters to 9999 thrust*

[um3k]

Yeah, the actual computation required for simulating the orbits of the planets and moons themselves is trivial for modern computers, even with a spacecraft being simulated at the same time. The real issue would be, as mentioned, the inherent instability of the Kerbal solar system.

[Fel]

That sounds like a challenge!

*turns on unlimited fuel*

*edits Mainsail parameters to 9999 thrust*

10000kN of thrust against a 50000000000000000Mg planet... comes out to be.... 200 atto (10 ^ -18) m/s^2

In comparison, you can think of an atom as being within the pico (10^-12) and nano(10^-9) range.

And, to change that orbit by 1 cm/s, you'll need to run that engine for 115.74 days

Edited June 10, 2013 by Fel

[Astronomer]

The planets would collide and assemble into a Kraken-like figure.

[Sapphire]

10000kN of thrust against a 50000000000000000Mg planet... comes out to be.... 200 atto (10 ^ -18) m/s^2

In comparison, you can think of an atom as being within the pico (10^-12) and nano(10^-9) range.

And, to change that orbit by 1 cm/s, you'll need to run that engine for 115.74 days

Ah, but who says I'd only be using one engine?

I could land like 10 of those and dock them with KAS... now I'd just need some giant RCS thrusters, then I could fly Gilly around!

Edited June 10, 2013 by Sapphire

[Strategia]

Actually, the inherent instability might be interesting if other solar systems were implemented/modded in. There could be a challenge to get X amount of Kerbals to a colony in another system before the Kerbol system eats itself.

now I'd just need some giant RCS thrusters, then I could fly Gilly around!

On the scale of Gilly, those would be your RCS thrusters

[Fel]

Ah, but who says I'd only be using one engine?

I could land like 10 of those and dock them with KAS... now I'd just need some giant RCS thrusters, then I could fly Gilly around!

Fine, 480 engines with 9999kN of thrust against the 124 Eg Gilly... you'll need 12 min of constant thrust to change the orbit by 1 cm.

But, I guess if you were REALLY determined you could have 1,000,000 engines producing a total of 10 PN of thrust and be able to move Gilly 1 cm in 11 seconds. (Or w/e... I'm tired and confused by my own math)

Edited June 10, 2013 by Fel

[DisarmingBaton5]

Melted computer.

[WhiteWeasel]

Ah, but who says I'd only be using one engine?

I could land like 10 of those and dock them with KAS... now I'd just need some giant RCS thrusters, then I could fly Gilly around!

Kinda off topic but this makes me think of that one episode of invader zim? Know what I'm talking about?