[WIP][1.8.1, 1.9.1, 1.10.1, 1.11.0–2, 1.12.2–5] Principia—version ‎‎Kronecker, released 2024-11-01—n-Body and Extended Body Gravitation

[olekopyto]

How To Install?

[Dolin]

Is there a version working in windows x64?

I've tried both the win32 and the linux versions from Espernet IRC  and, unsurprisingly, principia doesn't work. No crashes, it just works as normal KSP.
 

Is there a way or should I try running a ubuntu distro? (i really don't want to roll back to win32)

Thank you for your time.

Dolin

[eggrobin]

@Dolin, the next version, Буняковский, which will target 1.1, will work with win64; as you may be aware, win64 is unsupported for now.

Edited March 27, 2016 by eggrobin

[Nydoc]

It has been reported that the Chinese science administration plan to put a communication relay satellite in a halo orbit around Earth-moon L2 in 2018. A satellite in this orbit would have constant line of site to Earth and to the far side of the moon. This seems like a very useful orbit. I found a paper online describing the characteristics of a similar orbit. Has such an orbit been demonstrated using Principia?

[En_Dotter]

For quite a few days i tried to think of a "worthy" set of words to post here and i cannot find any... This is... well... how to simply put it... maybe... divine? When i heard of KSP for the 1st time i was hoping to see N-body simulation, but deep in my mind i was sure Squad wouldnt "bother" with it cus its too complex, both conceptually and possibly computationally. Then i found out about this mod (tnx to Scott Manley). Usually ppl say why a certain mod isnt stock, but i would like to ask when is this mod gonna be stock?

Keep up the magnificent work and i really hope transition to Unity 5 wont be a big of an issue for you.

I didnt read the entire thread (maybe 10 pages or less), and im quite interested in your numerical math series. I regret "losing" a professor that would teach me some numerical math in uni due to some arbitrary internal politics and having the course replaced with something else. But i guess molecular biology students dont suffer much by having that course replaced (except i do...).

[Jim Meillente]

I'd like to just say "Thank You".

[Guest]

Hey eggrobin, just made an account so I could comment. Do you think you can write a document outlining the various symplectic integrators and implementations? I'm currently taking a numerical methods course and thought this would be really relevant.

Additionally, do you think you'll implement any additional terms beyond J2 in the potential model of the planets? Thank you for your hard work!

[NathanKell]

@Neo63 there used to be a pdf in the OP discussing it, IIRC. It may still be on the repo, but @eggrobin would know best.

[Pyromartian]

Hello, I have been trying out this mod and it has been going great! I recently put a ship in Kerbin-Mun L1 for science purposes. Thank you for your hard work!

 

Unfortunately I also had a crash, but I don't know if it is worth reporting so close to 1.1 

 

I also wanted to ask, how do you pronunce  Буняковски?

[Autochton]

10 hours ago, Pyromartian said:

I also wanted to ask, how do you pronunce  Буняковски?

It transliterates to 'Bunyakovskiy'.

[eggrobin]

23 hours ago, Neo63 said:

Hey eggrobin, just made an account so I could comment. Do you think you can write a document outlining the various symplectic integrators and implementations? I'm currently taking a numerical methods course and thought this would be really relevant.

Additionally, do you think you'll implement any additional terms beyond J2 in the potential model of the planets? Thank you for your hard work!

It's SRKNs and explicit embedded RKNs (really only one of the latter). The implementations (including obsolete ones, the ones with abbreviations in the name) can be found in https://github.com/mockingbirdnest/Principia/tree/master/integrators (https://github.com/mockingbirdnest/Principia/blob/master/integrators/symplectic_runge_kutta_nystr%C3%B6m_integrator.hpp are the available SRKNs, https://github.com/mockingbirdnest/Principia/blob/master/integrators/embedded_explicit_runge_kutta_nystr%C3%B6m_integrator.hpp is the available variable-step embedded explicit RKN).

Additional documentation on composition methods can be found in https://github.com/mockingbirdnest/Principia/blob/master/documentation/Composition%20methods.pdf.

Interesting overviews of these subjects can be found in some of the references in the bibliography (along with completely unrelated aerospace and astronomy technical reports and normative documents), see for instance Mclachlan, Reinout, and Quispel, Geometric Integrators for ODEs.

As for terms beyond J₂, maybe (for instance, the Moon has a fairly significant C²₂).

[ss8913]

will this mod have a binary distribution, post 1.1?

[NathanKell]

It does now.

[Dolin]

Hello, I've had a quick go at a TLI in RSS and I have to say this is a true game changer, from someone who is considering a career in astrophysics after graduation, I bow to you.

If I may suggest a possible feature in the next version, Bunyakovsky, it would be a true help if we could edit any manoeuvre node at one time; for people who, like myself, are accustomed to eyeballing their sequences of burns until they look right, it would enable us to add or adjust a previous manoeuvre point on the go for best results.

Of less importance, but still useful, would be a periapse-apoapse indicator displaying altitude, measured to the surface of the parent body selected in the frame of reference window

I apologise if these are already in the works or if they've already been suggested, and I overlooked.

Thank you very much!

Edited April 1, 2016 by Dolin

[SergeantSnickers]

Hi,

for a space enthusiast like me this mod is just awesome. It's so much fun to just play around with all the different kinds of orbits and maneuvers that are possible.

I also managed to place my probes in different Lagrange points and was so fascinated that I already created a small video about flying to the Kerbin-Mun-L4 Point: Find the Video here if you want to see how to go there :)

 

Only thing I would really like to have is, like Dolin said, a periapsis, apopasis and relative inclination indicator. This would help me alot as a guy who is eyeballing nearly every maneuver.

 

Thanks alot for this awesome mod!

[FreeThinker]

Hey, I would like to make KSPI-E compatible with Principia. The ptoblem is Warp trave and persistant trust effects with low trust engines. While warping itself works well, once you return back to normal space, you recieve large amount of latent speed. The persitant trhust effect which supposed to be aplied durring time work don't work at all. I wonder if I can somehow use Principia library to make them function again as intended.

[eggrobin]

On 01/04/2016 at 4:03 PM, Dolin said:

edit any manoeuvre node at one time; for people who, like myself, are accustomed to eyeballing their sequences of burns until they look right, it would enable us to add or adjust a previous manoeuvre point on the go for best results.

This will not be in Буняковский, because it would require some significant modifications to the logic of the flight plan and its UI.

On 01/04/2016 at 4:03 PM, Dolin said:

Of less importance, but still useful, would be a periapse-apoapse indicator displaying altitude, measured to the surface of the parent body selected in the frame of reference window

1 hour ago, SergeantSnickers said:

Only thing I would really like to have is, like Dolin said, a periapsis, apopasis and relative inclination indicator. This would help me alot as a guy who is eyeballing nearly every maneuver.

Apsides, on the other hand, will be in Буняковский (they have already been implemented). They are shown with respect to the fixed body in the reference frame, if any: in the barycentric rotating frame neither body is fixed, since the barycentre is fixed instead, so no apsides are shown; in the body-centred non-rotating frame the apsides are shown with respect to that body, just like in stock. In future (post-Буняковский) versions we plan to add the surface frame of a body (there the apsides will be shown with respect to that body, obviously), and the frame centred on a (secondary) body and fixing the direction of its primary (e.g. the frame fixing the centre of the Moon and the Earth-Moon line), where apsides will be shown with respect to the secondary (in that example, with respect to the Moon).

1 hour ago, SergeantSnickers said:

I also managed to place my probes in different Lagrange points and was so fascinated that I already created a small video about flying to the Kerbin-Mun-L4 Point: Find the Video here if you want to see how to go there :)

Nice video! It showcases the various reference frames and the use of flight planning very well.

One remark concerning the Frenet trihedron and navball, once you're in space you probably want to check "Fix navball in plotting frame"; that aligns the navball with the reference frame (so that you can see the Kerbin-Mun orbital plane on the navball), and more importantly, it makes the tangent/normal/binormal (or prograde/radial/normal in stock KSP terminology) correspond to the plotted trajectory.

If you check that option, you'll see that at your L4 orbit insertion manœuvre you will have the navball markers aligned with the tangent, normal and binormal markers shown for your manœuvre, rather than aligned with the Kerbin-centred trajectory. Since at that point on your trajectory the tangent direction in the rotating frame is nearly opposite stock KSP's prograde marker, confusion can easily result from keeping the stock markers.

We should really document that option and make it more intuitive to use. 

[SergeantSnickers]

20 hours ago, eggrobin said:

Nice video! It showcases the various reference frames and the use of flight planning very well.

One remark concerning the Frenet trihedron and navball, once you're in space you probably want to check "Fix navball in plotting frame"; that aligns the navball with the reference frame (so that you can see the Kerbin-Mun orbital plane on the navball), and more importantly, it makes the tangent/normal/binormal (or prograde/radial/normal in stock KSP terminology) correspond to the plotted trajectory.

If you check that option, you'll see that at your L4 orbit insertion manœuvre you will have the navball markers aligned with the tangent, normal and binormal markers shown for your manœuvre, rather than aligned with the Kerbin-centred trajectory. Since at that point on your trajectory the tangent direction in the rotating frame is nearly opposite stock KSP's prograde marker, confusion can easily result from keeping the stock markers.

We should really document that option and make it more intuitive to use. 

Thank you! I will try the "Fix navball in plotting frame" the next time. It seems to be much more useful than i thought, so thanks for the explanation!

I will probably upload a new video tomorrow where i show how to reach the L1 point. Hope you guys check it out once it's uploaded!

[TheKSPDesigner]

Quick question!  I understand that this mod is built to calculate multiple physics bodies but how far in the future does that go?  There must be a limit to how far in the future the program can accurately predict N-Body physics, right?  Because I was thinking of a grand tour type mission where I waited until the planets got into a good alignment (I study astrodynamics ) and try to get gravity assists through them all to visit every planet and though, will this go in the future enough to get an accurate maneuver?  Anyway, thanks a bunch for helping, keep up the good work!  

[Pyromartian]

7 hours ago, TheKSPDesigner said:

Quick question!  I understand that this mod is built to calculate multiple physics bodies but how far in the future does that go?  There must be a limit to how far in the future the program can accurately predict N-Body physics, right?  Because I was thinking of a grand tour type mission where I waited until the planets got into a good alignment (I study astrodynamics ) and try to get gravity assists through them all to visit every planet and though, will this go in the future enough to get an accurate maneuver?  Anyway, thanks a bunch for helping, keep up the good work!  

It does depend on your computer on how far into the future and how accurate the prediction path is.

However, I have noticed that just like in the stock game floating point errors will screw up your path anyways. It is also super sensitive to conditions, so chances are you wouldn't be able to burn as accurately.

Download it and give it a shot. You should be able to have a feel of its limitations as soon as you set up a flight plan and make a burn.

[SergeantSnickers]

Hi guys,

I just uploaded a new video of KSP with Principia where I explore the Mun L1 and L2 point. So if you are interested jut check it out!

Mun L1 and L2 Point

Can't say it often enough, I just love this mod!

Edited April 6, 2016 by SergeantSnickers

[TheKSPDesigner]

11 hours ago, Pyromartian said:

It does depend on your computer on how far into the future and how accurate the prediction path is.

However, I have noticed that just like in the stock game floating point errors will screw up your path anyways. It is also super sensitive to conditions, so chances are you wouldn't be able to burn as accurately.

Download it and give it a shot. You should be able to have a feel of its limitations as soon as you set up a flight plan and make a burn.

Alright! Thanks dude

[Guest]

On 3/31/2016 at 1:28 PM, eggrobin said:

It's SRKNs and explicit embedded RKNs (really only one of the latter). The implementations (including obsolete ones, the ones with abbreviations in the name) can be found in https://github.com/mockingbirdnest/Principia/tree/master/integrators (https://github.com/mockingbirdnest/Principia/blob/master/integrators/symplectic_runge_kutta_nystr%C3%B6m_integrator.hpp are the available SRKNs, https://github.com/mockingbirdnest/Principia/blob/master/integrators/embedded_explicit_runge_kutta_nystr%C3%B6m_integrator.hpp is the available variable-step embedded explicit RKN).

Additional documentation on composition methods can be found in https://github.com/mockingbirdnest/Principia/blob/master/documentation/Composition%20methods.pdf.

Interesting overviews of these subjects can be found in some of the references in the bibliography (along with completely unrelated aerospace and astronomy technical reports and normative documents), see for instance Mclachlan, Reinout, and Quispel, Geometric Integrators for ODEs.

As for terms beyond J₂, maybe (for instance, the Moon has a fairly significant C²₂).

Sweet ligatures in your documentation  I'll look into your references, thanks. Do you know of any implementations that have different adaptive time steps for each object? 

[SergeantSnickers]

Everytime I play this mod it makes me love it even more! 

I was thinking about doing a career paly with Principia but am a bit concerned about the performance. Does anybody of you has experience playing Principia with multiple vessels in orbit?

And for the people still interested in some Video footage, I just put a probe in the crazy looking Minimus L5 point: Check it out!

[_Augustus_]

3 hours ago, SergeantSnickers said:

Everytime I play this mod it makes me love it even more! 

I was thinking about doing a career paly with Principia but am a bit concerned about the performance. Does anybody of you has experience playing Principia with multiple vessels in orbit?

And for the people still interested in some Video footage, I just put a probe in the crazy looking Minimus L5 point: Check it out!

It'll cause lag and potentially crashes on modded 32bit, with x64 it should be fine.