[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

[123nick]

3 minutes ago, eggrobin said:

It is also the only mod that will cause actual crashes (as opposed to just breaking and throwing exceptions), that has its own logs, that requires the user to download different binaries for different platforms (even within Linux, things vary, the Ubuntu build seems to have trouble on arch), that requires the user to install the appropriate C++ runtime (this is probably what @FreeThinker is running into by the way, missing or misinstalled C++ runtime), etc.

As a result, a (small) potential barrier, and one that hopefully heightens the chances of having the user read the FAQ, is welcome while the mod is still very unstable:

In addition, the IRC channel allow users to get support fairly quickly with the usual stuff (e.g. misinstalled C++ libs), and facilitates bugreporting (since I can quickly specify which logs are needed, how to start a journalling instance to replicate the bug, etc.).

When we consider that things are stabler, we will switch to a more conventional release mechanism. In the meantime this minor inconvenience helps us fix issues more quickly by interacting with users, which helps us find and diagnose the bugs.

oh ok, i guess that makes sense. also, a mod that requires such? like, a mod with its own log , different binaries for different platforms,  appropriate C++ runtimes? wow, thats alot ! this mod must be extremely technically complicated and extremely ambitious too require such. i also would assume it doesent play well with some other mods that add in .dlls? simply because how big this mod is.

1 minute ago, Svm420 said:

You must not frequent the popular mod threads. Maybe only few now, but tolerating the attitude only perpetuate the false assumption that that it is ok. That is whwy I said I would love for more of the most popular mods to adopt this idea. If only  

 

Egg Robin beautifully sums it up. Straight from the horses mouth.  

well, i guess that now he replied, it does sorta makes sense. i mean, the only mod that requires different binaries for different platforms,  has its own logs, and requires the appropriate C++ runtime, it would be appropriate for it too also have a different download method, for such a expansive and big and technologically complex mod.

[ss8913]

I'm assuming that this is one of those things where waiting for 1.1 will probably greatly improve the experience, since my hardware won't be doing dumb things in the CPU that are the GPU's job, and will have the ability to math a lot harder for the Hamiltonian physics required?   Also, will installing this mod break existing saves, and if so, in what ways/are there workarounds?

[FreeThinker]

1 hour ago, eggrobin said:

It is also the only mod that will cause actual crashes (as opposed to just breaking and throwing exceptions), that has its own logs, that requires the user to download different binaries for different platforms (even within Linux, things vary, the Ubuntu build seems to have trouble on arch), that requires the user to install the appropriate C++ runtime (this is probably what @FreeThinker is running into by the way, missing or misinstalled C++ runtime), etc.

As a result, a (small) potential barrier, and one that hopefully heightens the chances of having the user read the FAQ, is welcome while the mod is still very unstable:

In addition, the IRC channel allow users to get support fairly quickly with the usual stuff (e.g. misinstalled C++ libs), and facilitates bugreporting (since I can quickly specify which logs are needed, how to start a journalling instance to replicate the bug, etc.).

When we consider that things are stabler, we will switch to a more conventional release mechanism. In the meantime this minor inconvenience helps us fix issues more quickly by interacting with users, which helps us find and diagnose the bugs.

I have 19 C++ Redistributes installed (2005 - 2013). Exactly which on am I supposed to have installed?

[eggrobin]

1 minute ago, FreeThinker said:

I have 19 C++ Redistributes installed (2005 - 2013). Exactly which on am I supposed to have installed?

The 2015 ones for x86, linked in the FAQ: https://github.com/mockingbirdnest/Principia/wiki/Installing,-reporting-bugs,-and-frequently-asked-questions#its-not-starting-dllnotfoundexception.

[goncaloeaguiar]

Great mod this is. 

I've been experimenting with it and I found it amazing how near the mun the stock approximation is so invalid. Even a simple transfer to the mun, you have to be carefull and correct your trajectory, or else you end up crashing into it! 

Now as for the Lagrange point exploitation I've been experimenting along with the Kerbin - mun system, and to my surprise L4 and L5 are easy to get, just Homan transfer to the approximate position and turn the 2nd frame of reference option on, and you see your craft orbiting about that sweet spot. I tried to get to the Kerbin - Mun L1 and L2 and it was pretty difficult to even get the right burn perdition to achieve it, let alone actually burn into it. Basically what I did is to aim more or less to the boundary of the mun hill sphere, or 14,194,000 km height from Kerbin. The best attempt I had was a direct transfer to two halo revolutions around the L1. 

Now I will try the distant retrograde orbit (DRO)! Anyone as any clue of how to achieve them? 

[FreeThinker]

I love it so far, but I have te relearn some ksp space faring skills, especialy docking is a lot harder now. I used to be able to tweak the orbits in such a way I would intercepting as vessel a few hundred meter, now I have to strugle to get in near a kilometre

Edited March 1, 2016 by FreeThinker

[FreeThinker]

21 minutes ago, goncaloeaguiar said:

Great mod this is. 

I've been experimenting with it and I found it amazing how near the mun the stock approximation is so invalid. Even a simple transfer to the mun, you have to be carefull and correct your trajectory, or else you end up crashing into it! 

Now as for the Lagrange point exploitation I've been experimenting along with the Kerbin - mun system, and to my surprise L4 and L5 are easy to get, just Homan transfer to the approximate position and turn the 2nd frame of reference option on, and you see your craft orbiting about that sweet spot. I tried to get to the Kerbin - Mun L1 and L2 and it was pretty difficult to even get the right burn perdition to achieve it, let alone actually burn into it. Basically what I did is to aim more or less to the boundary of the mun hill sphere, or 14,194,000 km height from Kerbin. The best attempt I had was a direct transfer to two halo revolutions around the L1. 

Now I will try the distant retrograde orbit (DRO)! Anyone as any clue of how to achieve them? 

It would be nice if the 5 larange point could be visualised, so that you can aim for them and get into that perfect orbit.

Edited March 1, 2016 by FreeThinker

[goncaloeaguiar]

Guys I managed to get a mun distant retrograde orbit after a few attempts.

 

Here's what I have achieved:

I will try now exploring minmus space

 

Thanks again for the great mod!

Edited March 2, 2016 by goncaloeaguiar

[goncaloeaguiar]

I can't get more excited!

Check out Earth-Minmus L4:

Look how perturbed it is by the mun!

[ss8913]

I've been perturbed by the mun for years even without this mod...

[nilof]

A few neato Kerbin-Mun orbits orbits in stock KSP + principa:

http://imgur.com/a/u41SA

First one is a Mun almost-capture, where the craft goes far enough out that Kerbin perturbation turns a retrograde munar orbit into a prograde munar orbit. The second one is a Kerbin-Mun Halo orbit around their L2 point. Stationkeeping delta-v needed should be less than 2 m/s per month, but it needs constant baby sitting as in 4-5 stationkeeping burns per orbit. If you don't babysit it for half a halo orbit it will either fall into an unstable prograde mun orbit or get flung out into a solar orbit.

 

First one is plotted in an inertial Mun-centered frame, second one is obviously Kerbin-Mun corotating.

 

Edited March 6, 2016 by nilof

[goncaloeaguiar]

1 hour ago, nilof said:

First one is a Mun almost-capture, where the craft goes far enough out that Kerbin perturbation turns a retrograde munar orbit into a prograde munar orbit. The second one is a Kerbin-Mun Halo orbit around their L2 point. Stationkeeping delta-v needed should be less than 2 m/s per month, but it needs constant baby sitting as in 4-5 stationkeeping burns per orbit. If you don't babysit it for half a halo orbit it will either fall into an unstable prograde mun orbit or get flung out into a solar orbit.

 

First one is plotted in an inertial Mun-centered frame, second one is obviously Kerbin-Mun corotating.

 

Very neat! How do you manage the L2 Halo orbit? You need some inclination right?

[nilof]

21 hours ago, goncaloeaguiar said:

Very neat! How do you manage the L2 Halo orbit? You need some inclination right?

First, get close enough to where you need to be. There aren't many visual references, I personally did it by picking an orbit that just grazed the Mun SOI and canceling my velocity relative to Mun surface when I saw Kerbin reappearing from behind the mun. After that, I performed a 20 m/s burn that gives you a bit of velocity towards the kerbin-mun axis and a bit of inclination, until I saw a corkscrew-ish pattern towards the mun or outwards which is round rather than elliptical. Incidentally, I had an easier time seeing it when the trajectory was falling outwards.

When you have that, keep adjusting your trajectory with burns directly along the kerbin-mun line (use the plotting frame navball) every so often to make you fall towards the mun if you were escaping, or towards escape if you were falling towards the mun. To some extent you are spending fuel on hovering, but if you are close to the Lagrange point it will be almost nothing and you should be able to bring it down to single digit or less than single digit m/s per orbit.

I had a much harder time doing it in RSS, since KSP needs all planet rotation axes to be aligned with each other, so the Moon's rotation axis relative to its orbit is wrong and Moon surface velocity is no longer helpful. = (

TLDR: Halo orbits are self correcting in the directions perpendicular to the Kerbin-Mun axis, but you need to stationkeep to not fall outwards or towards the Mun. Away from the lagrange points this is equivalent to hovering. Close to the lagrange point the delta-v cost of "hovering" is tiny. Also, staying in a halo orbit is quite a lot easier than trying to balance at the actual lagrange point.

Edited March 7, 2016 by nilof

[nilof]

Eyeballed Farquhar transfer!

I'm working on getting better at quickly putting stuff into good halo orbits so I can incorporate L2 comsats into my next career mode save with remote tech. With ion engine fine tuning I was able to get down to ~1 stationkeeping burn per munar orbit @ ~0.2 m/s.

Edited March 9, 2016 by nilof

[eggrobin]

@nilof: very nice orbits!

Unrelated: a solar eclipse (RSS by @NathanKell et al. for the bodies, Scatterer by @blackrack for their shadows, Principia for their motions). It goes without saying that without Principia, real-life eclipses cannot be observed in RSS, since in stock RSS the nodes of the Moon's orbit do not precess, whereas they do so in 18.60 years in real life. The shadow looks very nice, for comparison see the recent Himawari-8 images (video by @illectro).

Note that this required a bug fix to Earth's rotation (which is off by a day per year in RSS since 1.0.5).

[nilof]

That is incredibly cool! I'm guessing you can also observe Jupiter/Saturn multi-eclipses and get fairly accurate timings for those?

Edited March 10, 2016 by nilof

[FreeThinker]

I wonder has anyone had the problem not beeing able to continue with a compaign yet while using Principa?

[blackrack]

13 hours ago, eggrobin said:

@nilof: very nice orbits!

Unrelated: a solar eclipse (RSS by @NathanKell et al. for the bodies, Scatterer by @blackrack for their shadows, Principia for their motions). It goes without saying that without Principia, real-life eclipses cannot be observed in RSS, since in stock RSS the nodes of the Moon's orbit do not precess, whereas they do so in 18.60 years in real life. The shadow looks very nice, for comparison see the recent Himawari-8 images (video by @illectro).

Note that this required a bug fix to Earth's rotation (which is off by a day per year in RSS since 1.0.5).

Wow, that is quite accurate actually.

[TotallyNotSatan]

Are Brouwer or Buffon out for the Mac?

[RaginCaucasian]

Wow. This is amazing. I can't wait to try it!

[goncaloeaguiar]

I guess now with Scott Manley's advertisement of this mod running, it won't be long before the IRC gets flooded.

I have been waiting for a long time for this video, but still... No lagrange point demonstration, I was quite disappointed. But I guess he will try that some other time! 

[Sophistry]

Hmm. Is there an idiot's guide to playing with Principia somewhere (Or an accessible real life text that would help)? And does anyone have any suggestions for mods that complement it?

[PugzInSpace]

This is the best mod anybody has created so far. Thank you.

[WillThe84th]

This sounds like an amazing mod, but I have a few questions.

How would this affect planet packs such as OPM, would the systems of Sarnus, Urlum and Neidon become unstable and the moons would fly off into space?

How much does this mod affect the performance of the game? Does it dramatically slow down the game, or just a little?

[goncaloeaguiar]

58 minutes ago, WillThe84th said:

How would this affect planet packs such as OPM, would the systems of Sarnus, Urlum and Neidon become unstable and the moons would fly off into space?

Just try it.

 

How much does this mod affect the performance of the game? Does it dramatically slow down the game, or just a little?

 Like this

It is not noticeable if you have at least a 4-5 year old PC (mine is in that range). If you have a lot of crafts the game crashes before it stats to lag to the point of becoming unplayable.