Are KSP's physics, at this stage, a good approximation of this universe's physics?

[rodion_herrera]

Sorry, just saw it. But still gravity determines orbits.

LOL I honestly think Sarge is aware of that, but he meant "gravity" in a specific context, and that has to do with the "disappearing dimples"--he just was a loss for words on how to explain it But I'm a teacher so, I'm used to these things hehe

[asmi]

It's too complex, but really the real loss are Lagrange Points. Those things are useful...

It is not that complex. For example it is implemented in Orbiter. There is no universal analytical solution to the problem, but integration gives you VERY accurate results. Infact Orbiter's physics model is so accurate that I was able to get into Orbiter, set up ISS orbit and date to values from NASA site, and then fly Soyuz mission using MCC-M's cyclogram as if it would be a real-life mission.

[Rickenbacker]

Compared to 99% of games, it's very accurate. Not perfect, but close enough that real-life maneuvers and planning work in the game.

I really, really hate the "dimples in a rubber sheet" analogy - it's wrong on so many levels - but here it was actually used to explain the "only one gravity well at a time" limitation, so well done!

I believe the lack of Lagrange points is due to a Unity limitation?

[Specula]

Of course I know

When you are in orbit of Kerbin, you are attracted by Kerbol. It's just you don't feel it

Nope. You are in Kerbin's sphere of influence. Kerbin itself is in Kerbol's sphere of influence. This relationship does not form an enclosed triangle. It's a straight line. relationship A != relationship B here.

[rodion_herrera]

It is not that complex. For example it is implemented in Orbiter. There is no universal analytical solution to the problem, but integration gives you VERY accurate results. Infact Orbiter's physics model is so accurate that I was able to get into Orbiter, set up ISS orbit and date to values from NASA site, and then fly Soyuz mission using MCC-M's cyclogram as if it would be a real-life mission.

I don't want to thumbs-down your post, considering I'm an Orbiter user too, but however, I don't think the comparison is fair when it comes to talking about gravity computation, considering that, as mentioned in many posts above, there are indeed some "internal tweaks" to certain physical elements in the game, that perhaps if a realistic gravity model were to be considered, it would put too much a strain on the processor that it would render the game unstable/unplayable. So I understand that Squad perhaps are implementing workarounds to issues that would perhaps compound their development of the physics engine.

Now you might react with... "So why CAN Orbiter do it? Surely if M. Schweiger can, as one person, program it to work, Squad can implement it as well?" Well, Orbiter does have ONE feature of real life that is NOT implemented--OBJECT-TO-OBJECT or BODY-TO-BODY COLLISION. And KSC DOES HAVE COLLISION So you see? Perhaps those factors alone are already enough reason why the internal tweaks need to be done (i.e. no Lagrange points possible), if only to preserve how collision model works as it works now in the game. Additionally, Orbiter doesn't model "stack flexing" etc...

[Michael Kim]

Nope. You are in Kerbin's sphere of influence. Kerbin itself is in Kerbol's sphere of influence. This relationship does not form an enclosed triangle. It's a straight line. relationship A != relationship B here.

Ahhhh

I feel rude to say so but you should really pick up some high school physics before we continue the discussion

EDIT:

Well I still feel too rude

The thing is, Kerbol keeps you with Kerbin, and Kerbin makes you orbit around it

It's just like, when astronauts experience 0 g, they are still gravitational bounded to Earth right?

Edited May 30, 2013 by Michael Kim

[rodion_herrera]

While I can certainly see the eagerness of some of you to discuss intricacies of orbital mechanics, I hope you guys try not to focus on that or on who's "view" is correct--remember, we might be using different "jargon" to describe what we mean and that's bringing about the confusion. So please try to be cordial, or I will ask the mods again to close my thread. Thanks.

[asmi]

I don't want to thumbs-down your post, considering I'm an Orbiter user too, but however, I don't think the comparison is fair when it comes to talking about gravity computation, considering that, as mentioned in many posts above, there are indeed some "internal tweaks" to certain physical elements in the game, that perhaps if a realistic gravity model were to be considered, it would put too much a strain on the processor that it would render the game unstable/unplayable. So I understand that Squad perhaps are implementing workarounds to issues that would perhaps compound their development of the physics engine.

AFAIR Squad just doesn't WANT to implement that. To tweaks, just lack of proper implementation.

Now you might react with... "So why CAN Orbiter do it? Surely if M. Schweiger can, as one person, program it to work, Squad can implement it as well?" Well, Orbiter does have ONE feature of real life that is NOT implemented--OBJECT-TO-OBJECT or BODY-TO-BODY COLLISION. And KSC DOES HAVE COLLISION So you see? Perhaps those factors alone are already enough reason why the internal tweaks need to be done (i.e. no Lagrange points possible), if only to preserve how collision model works as it works now in the game. Additionally, Orbiter doesn't model "stack flexing" etc...

These two things are completely unrelated. Again, there are no "tweaks" that only permits latter, but not former, or vise versa. It simply is ABSENCE of a feature. Collisions in Orbiter are trivial and relatively easy to implement, but author has different priorities right now and frankly I argee with him that collisions are not THAT important since the space is HUGE and EMPTY and due to lack of asteroids there is almost nothing to collide with.

[ComradeGoat]

Ahhhh

I feel rude to say so but you should really pick up some high school physics before we continue the discussion

I don't think any of this is quite right.

The way the game is implemented, gravity is only calculated from the body you are directly orbiting. The sun is indeed ignored when you are in Kerbin orbit.

HOWEVER, Kerbin itself is set to follow a path ("on rails", as it were), around Kerbol which, although not done using the game's gravity model AIUI, is the same path that would be followed by a spacecraft placed in that position, with that velocity, when subject to the gravity model.

So you are not "feeling" the pull of the sun while orbiting Kerbin, but you're still beng dragged around that solar orbit by Kerbin itself. If you were able to feel both, you'd be able to extend Kerbin's SOI sufficiently to have Lagrange points, but that simply won't work, no matter how big you make the SOI, because the game won't simulate gravity from two objects.

[Michael Kim]

HOWEVER, Kerbin itself is set to follow a path ("on rails", as it were), around Kerbol which, although not done using the game's gravity model AIUI, is the same path that would be followed by a spacecraft placed in that position, with that velocity, when subject to the gravity model.

So you are not "feeling" the pull of the sun while orbiting Kerbin, but you're still beng dragged around that solar orbit by Kerbin itself. If you were able to feel both, you'd be able to extend Kerbin's SOI sufficiently to have Lagrange points, but that simply won't work, no matter how big you make the SOI, because the game won't simulate gravity from two objects.

So I am right.

[rodion_herrera]

AFAIR Squad just doesn't WANT to implement that. To tweaks, just lack of proper implementation.

These two things are completely unrelated. Again, there are no "tweaks" that only permits latter, but not former, or vise versa. It simply is ABSENCE of a feature. Collisions in Orbiter are trivial and relatively easy to implement, but author has different priorities right now and frankly I argee with him that collisions are not THAT important since the space is HUGE and EMPTY and due to lack of asteroids there is almost nothing to collide with.

I disagree re collision--docking in Orbiter is hugely unrealistic for now, due to lack of collision. While perhaps, the effects of a ship-to-ship collision is neglible when viewed from the perspective of gravity in general, I still think it matters. Same thing goes for effects like "stack stretching or warping" when the vehicle is on ascent--that happens in reality, and has to be taken into account--right now, Orbiter doesn't model that, and a Saturn V behaves like a dense diamond hurtling up to orbit

[asmi]

I disagree re collision--docking in Orbiter is hugely unrealistic for now, due to lack of collision. While perhaps, the effects of a ship-to-ship collision is neglible when viewed from the perspective of gravity in general, I still think it matters. Same thing goes for effects like "stack stretching or warping" when the vehicle is on ascent--that happens in reality, and has to be taken into account--right now, Orbiter doesn't model that, and a Saturn V behaves like a dense diamond hurtling up to orbit

No you again mixes things up, and I KNOW these things are unrelated. Orbiter's model is different in a sence that it models only vessel's motion, and it's up to vessel itself to implement flexing. All APIs and other things for that are in place. I've actually had collisions implemented at some experimental build of my GC (I'm one of D3D11Client's devs if you haven't figured it out by now ) and it was working so I know it's possible to do.

[haltux]

The FAR mod provides an excellent drag model, however, the current part catalogue is not adequate for building aerodynamic rockets. until fairings, larger engines, wider parts etc are introduced, realistic drag just makes everything a nightmare.

I would hope that squad doesn't gimp the final game just for the sake of some people who have been playing the alpha to death and have got used to the weird idiosyncrasies of an unfinished game.

You are right, Splode, and I agree with comham. But I think than when you release an update with flags and chairs (no offense), you don't plan to completely rewrite half of you game (the rocket design), which I think would be required for a "realistic rocket" design.

[rodion_herrera]

No you again mixes things up, and I KNOW these things are unrelated. Orbiter's model is different in a sence that it models only vessel's motion, and it's up to vessel itself to implement flexing. All APIs and other things for that are in place. I've actually had collisions implemented at some experimental build of my GC (I'm one of D3D11Client's devs if you haven't figured it out by now ) and it was working so I know it's possible to do.

I have used collision *.dlls in Orbiter, and frankly, they don't model collision very well. In many ways, KSP's collision model is better.

[Good_Apollo]

I don't really get that statement anyway, just because another 'game' does it does not mean it isn't complex and potentially too complex for Unity and/or Squad.

[rodion_herrera]

We're all only speculating here. Unless HarvesteR himself or one of the devs share their ideas on how the physics was developed for KSP, we can only keep speculating.

[Specula]

So I am right.

... No, he's saying you're completely wrong. just like I am. He's going so far as to say that Kerbol HAS no gravitational influence at ALL over ANYTHING. Not even the planets.

I'm also a programmer, I understand some basic physics kiddo. This game is not an accurate representation of RL physics concerning multiple gravitational influences. You're acted upon by the gravitational pull of one thing in game at a time and nothing else. The point's been reiterated enough. If it hasn't sunk in yet for you, it never will.

/argument.

Edited May 30, 2013 by Specula

[rodion_herrera]

So you're saying it's useless to teach kids certain basic principles of spaceflight, using KSP (which is the point and aim of this thread, actually, and not trying to see who has a better grasp of Newtonian physics), yes? If so, question answered, and I will ask a mod to close the thread now. Thanks for participating.

[Specula]

So you're saying it's useless to teach kids certain basic principles of spaceflight, using KSP (which is the point and aim of this thread, actually, and not trying to see who has a better grasp of Newtonian physics), yes? If so, question answered, and I will ask a mod to close the thread now. Thanks for participating.

Sorry, my last message was for Michael Kim not you.

[khyron42]

To steer back to the original question - other minor differences from real-world physics that I haven't seen mentioned...

- Collisions between parts in the same ship are dampened. As long as you don't apply enough force to break the inter-part connector, your ship's parts don't crush each other.

- gravitational pull is applied as a force to the ship as a whole, at the center of gravity. Since each part doesn't feel a separate pull, there's no ability to demonstrate tidal stabilization of a space station. (or tidal stresses pulling apart a ship.)

- parts and inter-part connections all have a simple "impact tolerance." exceeding that tolerance destroys the part, or breaks the inter-part connection. Beyond some flexibility in the inter-part connections, there's no modeling of things getting bent, broken but still attached, or suffering non-destructive failures.

The first and third ones would only matter if you're trying to show engineering aspects of force distribution. Kerbal is still the only modular ship-design game I know of, so it's the only one that can show force distribution, but it was more accurate in the very earliest versions - collisions of parts in the same ship were removed to gain performance and improve rocket stability, I think around 0.15?

Gravity per-part was also there in the very earliest versions, but calculating pull and applying forces to each part separately was a huge performance roadblock, and it was switched over to calculate & apply forces to the craft as a whole (I think in the same update that added fuel lines.)

Edited May 30, 2013 by khyron42

[Splode]

You are right, Splode, and I agree with comham. But I think than when you release an update with flags and chairs (no offense), you don't plan to completely rewrite half of you game (the rocket design), which I think would be required for a "realistic rocket" design.

Actually, fairings are planned. It isn't all that difficult to add the parts we need for 'realistic rocket design'.

We already have nosecones, which proves they are planning to fix it (not to mention they have repeatedly said they will re-do the drag model). Currently, nose cones add weight and drag!

You can already experience this sort of aerodynamic planning by downloading KSPX part pack, using Fairing Factory, and installing FAR. It's just an effort.

[rodion_herrera]

Sorry, my last message was for Michael Kim not you.

Thanks for clearing that up. Next time, use the QUOTES button--it was created for that purpose.

[rodion_herrera]

Ok a legitimate question for the hardcore physics and astrophysics guys--do you think that KSP cannot demonstrate rigid body dynamics and/or n-body problems at its current state right now, even at the most rudimentary level?

[haltux]

There is a recent thread on n-body physics:

http://forum.kerbalspaceprogram.com/showthread.php/29938-A-question-on-n-body-physics?highlight=lagrange

Developers said (in a lost post I guess) that n-body physics would not be implemented because it raises problems with time warp.

This can be easily understood. Patched Conics Approximation (current KSP model) allows exact predictions. It means that you can easily know where your ship will be any time in the future. Real n-body modelling does not allow exact prediction, you can only make approximations.

Consequently, the behaviour of your ship will depend on the time warp mode. To know where you ship will be at the next frame, in normal speed, you compute the position of your ship a fraction of a second later. In maximum time warp, you compute the position of your ship (let's say) one minute later. The cumulative error will be much higher in the second case after the same amount of (game) time, whatever prediction algorithme you use.

In terms of gameplay, knowing that the time acceleration changes the trajectory is difficult to accept. I don't know how Orbiter is handling that problem.

[Holo]

So you're saying it's useless to teach kids certain basic principles of spaceflight, using KSP (which is the point and aim of this thread, actually, and not trying to see who has a better grasp of Newtonian physics), yes? If so, question answered, and I will ask a mod to close the thread now. Thanks for participating.

Nope, it is completely useful. Professionals also sometimes use patched conics, and everything works pretty much how it does in real life. Other people in this thread are essentially saying "KSP is not 110% accurate to real life, it's entirely useless".