Jump to content

[WIP][1.8.1, 1.9.1, 1.10.1, 1.11.0] Principia—version Germain, released 2021-01-13—n-Body and Extended Body Gravitation


Recommended Posts

19 minutes ago, pleroy said:

but I am not sure what I am seeing on these pairs of screenshots: what is on the left and what is on the right

Thanks for asking me to clarify...indeed I left a lot out:  Simply stated: You & Egg are much more experienced in this domain than I, so I shared the observation in case seeing it sparked in your minds a 'go read x ,y, z papers" (or to catch if there was something obvious I was completely misunderstanding). 

Initially, I was skeptical of my understanding regarding the rather complex patterns of motions of Trappist1 in the skies of the various planets: for example, especially clear in the large drift in the 1d  & 1e plots.   I speculate these are the patterns of interaction between the planets on each other enhanced by their resonance frequencies, but perhaps you see  a more accurate interpretation.  For visual example, Trappist1 goes from a large relative motion in Trapiist1d's sky to almost stationary during the time period captured in the 1st video in the album. Also, adjusting the history plot in real time reveals some patterns in the less drifting more 'oval' wobbles of 1f & 1g.  So I am open to your insights if you would like to share.

Regarding the illumination difference: to reduce the possibility that I had messed up the Principia model I ported it to 1.8.1.  This forced me to check and to get a dramatically better understanding/experience with all the numbers & patching etc. (& some of what can go wrong when I mix things up).  I attempted first to port for myself Gregroxmun's nice SLIPPIST1 to 1.8.1 but kept failing to get the 1.8.1 Kopernicus to build all the SLIPPIST1 planets and I still can not figure out what is going wrong (I currently speculate a pqs or scaled space issue), so I switched to using the REX Trappist1 system CB models which were confirmed to work in 1.8.1 along with their scatterer & EVE choices...thus the difference in illumination between LEFT = 1.7.3 SLIPPIST1 models & RIGHT = 1.8.1 REX models for "Trappist1 for Principia"'s gravity model of that system in the album images.  I do hope that this personal use exploration port to 1.8.1 conveys my very sincere & great appreciation for the huge amount of beautiful work that each of you @GregroxMun  @eggrobin @pleroy @AndrewDraws have done.  With Principia's great new features of orbit analysis, CB plotting, plus the past big improvement of encoding saves in base64 (& hopefully soon the angular momentum) combined with the better garbage collection handling in KSP 1.8+ I am very much looking forward to finally building much more complex saves.  I expect to now be able to use (much more interesting) vessels to explore the gravitation dynamics & opportunities of this system on various long term vessel & station orbits as well as various types of transfers over 5 to 10 years of simulated time (on my decent but older computers).  How far out do you assess the Trappist1 gravity model provides a reasonable approximation of the system beyond the published data? (I have no way to check like I did with your implementation of up to J12 in RSS which makes it a very good model even after 80 years).  Thanks again!

Link to post
Share on other sites

As Frege didn't have a 1.9 build (released before the new ksp version), I waited for the new moon. But Frenet (released after the new ksp version) never seemed to manage to build for 1.9 either? Or am I somehow not seeing a link in the github README that links to the 1.9 build? Nothing is mentioned anywhere (github readme / forum OP) about skipping 1.9, so I am guessing this is an oversight?

ps: this threads title probably need an update. Still only lists up to 1.8.

Link to post
Share on other sites
10 hours ago, unregistered_user said:

As Frege didn't have a 1.9 build (released before the new ksp version), I waited for the new moon. But Frenet (released after the new ksp version) never seemed to manage to build for 1.9 either?

My observation is that the mod makers have been focused on working on some very remarkable additional features for the Principia mod...that takes time & focus.   A change in KSP version requires checking for changes in the API etc.  that can affect a mod...that also takes time.  Additionally, Frenet may introduce by itself new bugs due to the new features in the works not just due to a change in KSP version thus using it in familiar KSP versions is prudent.

If you are familiar with Principia already, one way you may help other Principia users is by verifying Frenet performs as you expect in 1.8.1: see:  "This release has quite a few (invisible) changes to the management of parts and vessels, so we invite users to upgrade to help us find any problems/bugs that we could have introduced.""  The mod authors have a very good history of supporting new KSP versions when the work is ready to do so.

Link to post
Share on other sites
  • 2 weeks later...

I'm using Principia with RO on version 1.7.3. and I noticed that my persistent file has ballooned in size after launching 4-5 satellites to low Earth orbit. File has gone from around 400-700 kB to 200+ MB. Is this normal behavior? Scene changes now take several minutes, if the game doesn't crash outright or becomes unresponsive.

Link to post
Share on other sites
8 minutes ago, viktor19 said:

I'm using Principia with RO on version 1.7.3. and I noticed that my persistent file has ballooned in size after launching 4-5 satellites to low Earth orbit. File has gone from around 400-700 kB to 200+ MB. Is this normal behavior? Scene changes now take several minutes, if the game doesn't crash outright or becomes unresponsive.

I have noticed the same, and thought it was strange as my PC handled Principia pretty well on older KSP versions. Overall feels sluggish and scene changes take 2-3 mins.

Link to post
Share on other sites
18 minutes ago, viktor19 said:

I'm using Principia with RO on version 1.7.3. and I noticed that my persistent file has ballooned in size after launching 4-5 satellites to low Earth orbit. File has gone from around 400-700 kB to 200+ MB. Is this normal behavior? Scene changes now take several minutes, if the game doesn't crash outright or becomes unresponsive.

6 minutes ago, hypervelocity said:

I have noticed the same, and thought it was strange as my PC handled Principia pretty well on older KSP versions. Overall feels sluggish and scene changes take 2-3 mins.

This is known and expected, and tracked in #2400: the underlying issue is that the past trajectories of vessels accumulate over the years, and end up being very voluminous.

We are working on a solution, but this will take a while, and will likely happen in multiple steps*.

In the meantime, the suggested workaround is to delete old vessels (including debris) if you don't need them.

* We have experimented with a quick fix (using a lossy floating-point compression algorithm) that reduces savefile size by a factor of 4. I am currently looking at papers on compact representation of ephemerides by Jean Chapront, Vũ Dương Tuyền, and Сергей Михайлович Кудрявцев, with whose methods I hope to gain several orders of magnitudes by taking advantage of the highly periodic behaviour of orbits.

Link to post
Share on other sites
12 hours ago, eggrobin said:

This is known and expected, and tracked in #2400: the underlying issue is that the past trajectories of vessels accumulate over the years, and end up being very voluminous.

We are working on a solution, but this will take a while, and will likely happen in multiple steps*.

In the meantime, the suggested workaround is to delete old vessels (including debris) if you don't need them.

* We have experimented with a quick fix (using a lossy floating-point compression algorithm) that reduces savefile size by a factor of 4. I am currently looking at papers on compact representation of ephemerides by Jean Chapront, Vũ Dương Tuyền, and Сергей Михайлович Кудрявцев, with whose methods I hope to gain several orders of magnitudes by taking advantage of the highly periodic behaviour of orbits.

 

12 hours ago, eggrobin said:

@viktor19, @hypervelocity: by the way, we are interested in your large saves: looking at them will allow us to test the effectiveness of our solutions on real orbits. Please post them on #2400.

I would like to contribute with my save file.

Is posting to github the only way? I wouldn't open an account just for just this one thing, if it isn't necessary.

Link to post
Share on other sites
On 3/9/2020 at 1:40 PM, viktor19 said:

I would like to contribute with my save file.

Is posting to github the only way? I wouldn't open an account just for just this one thing, if it isn't necessary.

@viktor19: Google Drive works too: just upload a save and post the link here.

Link to post
Share on other sites
  • 2 weeks later...

If anyone is curious release Frenet runs fine as best I can tell when built from source for 1.9.  Passes all tests in compile and works flawlessly.  I've been using it (along with my personal unifiedcamera build of scatterer) for some time now.

Edited by R-T-B
Link to post
Share on other sites

For the new moon (lunation number 250), the new release (Frobenius) is out.

After nearly a year of work and 96 pull requests, Principia now enforces the conservation of angular momentum, as had been alluded to earlier.

This means, among other things, that:

  • the long-standing bug #1639, reported by @Damien212 in 2017, whereby the orientation of vessels changed when they underwent an SoI transition, is resolved;
  • vessels will rotate as rigid bodies, following Euler’s equations, in time warp; in particular they may exhibit the Джанибеков effect in time warp, as illustrated in the short video below;
  • changes in mass distribution within the vessel, e.g., due to fuel transfer, will affect the angular velocity, as illustrated in the short video below.

     

See the change log for more details.

We support two sets of versions of KSP: downloads are available for 1.8.1, and for 1.5.1, 1.6.1, 1.7.x. Make sure you download the right one (if you don't, the game will crash on load).

For the convenience of our Chinese users, the binaries can be downloaded either from Google Drive or from 腾讯微云.

Link to post
Share on other sites

Absolutely phenomenal work, once again! Can't wait to use this release, since this means that vessels in near circular orbits and told to hold prograde will only drift slightly with timewarp because of their extant rotation!
This will make space stations look even better!

Link to post
Share on other sites

Important notice: we have discovered an issue in Frobenius that renders the game unplayable (rockets occasionally get flipped back-to-front while being launched).

As a result, we have retracted the Frobenius binaries while we investigate the issue; Frobenius will be re-released once we have resolved it.

Link to post
Share on other sites

Can't wait to check it out once it is re-released! I forgot to ask this last time, is the angular momentum conserved across scene changes? It seems like a safe assumption given that Principia will completely manage any not-landed vehicle (and of course I can just wait and check it out myself later, but who has the patience, right? :sticktongue:)

Link to post
Share on other sites

Frobenius is available again.

Note that if you downloaded it before March 24, 21:00 UTC, you probably have a buggy version (see #2503). Please download it again. If you have the correct version, the version string in the Principia UI says 2020032409-Frobenius-0-g21e623384639abd4bdd277a9fb722e94ee9763e4. We apologize for the inconvenience.

Also note that, as mentioned in the change log for Frobenius, Principia is now incompatible with PersistentRotation, as Principia now handles rotation. Please uninstall PersistentRotation if you have been using it, otherwise the two mods will fight each other.

Link to post
Share on other sites

Thank you for the quick fix!
I have already played around with the rotations a bit (I might have gotten my hands on a bad copy before it was removed) and it is quite a nice addition indeed!

Link to post
Share on other sites
On 3/23/2020 at 7:45 PM, R-T-B said:

If anyone is curious release Frenet runs fine as best I can tell when built from source for 1.9.  Passes all tests in compile and works flawlessly.  I've been using it (along with my personal unifiedcamera build of scatterer) for some time now.

@R-T-BNote that the Kerbol system has changed in 1.9, so your Principia won't recognize it and won't stabilize the Jool system.  Expect to see planetary collisions after a few days.

Link to post
Share on other sites
4 minutes ago, pleroy said:

 Note that the Kerbol system has changed in 1.9,

It has? The only things I am aware of were the texture/shader updates. That counts as a changed system?

Edited by Delay
Link to post
Share on other sites
3 minutes ago, Delay said:

It has? The only things I am aware of were the texture/shader updates. That counts as a changed system?

It seems that the max_radius has changed for a number of celestials.  Everything counts as a change, we need a way to unambiguously detect the stock system.

Link to post
Share on other sites
3 hours ago, pleroy said:

@R-T-BNote that the Kerbol system has changed in 1.9, so your Principia won't recognize it and won't stabilize the Jool system.  Expect to see planetary collisions after a few days.

Would that happen in a save game?

If not, I should probably revert and update the fingerprints to force the systemfixup.  I don't think it's an issue though:  I still have retrobop and the planets/moons have hung around for nearly a game year in Frenet.

Edited by R-T-B
Link to post
Share on other sites

我发现了一个bug。我用的是KSP 1.8.1.2694,添加了RP-1和许多Mod,同时安装的是最新版的Principia。打开游戏后显示“The Principia DLL failed to load. Dependencies, namely the Microsoft Visual C++ 2015-2019 Redistributable (x64) -14.22.27821, were not found.  Warning: don’t load a Principia save before you have fixed this error, it might get damaged.”我创建了一个新的存档,在下边没有看到Principia的图标。之后我尝试发射了一枚火箭,在导航球上并没有看到Principia的参考系。(我在没有用任何Mod的时候也是这样)。我下载的是Google里的“Principia Frenet for 1.8.1”

Edited by Peter JY
Link to post
Share on other sites
1 hour ago, Peter JY said:

打开游戏后显示“The Principia DLL failed to load. Dependencies, namely the Microsoft Visual C++ 2015-2019 Redistributable (x64) -14.22.27821, were not found.  Warning: don’t load a Principia save before you have fixed this error, it might get damaged.”

  

On 2/5/2014 at 1:19 AM, eggrobin said:

请阅读常见问题解答

Quote

If you get the message "Dependencies, namely the Microsoft Visual C++ 2015-2019 Redistributable, were not found", well, you need to install the Microsoft Visual C++ 2015-2019 Redistributable.

请安装Microsoft Visual C++ 2015-2019 Redistributable。

Edited by eggrobin
Link to post
Share on other sites
9 hours ago, pleroy said:

It seems that the max_radius has changed for a number of celestials.  Everything counts as a change, we need a way to unambiguously detect the stock system.

Would a fingerprint before and after sucesfully running the patcher on 1.9 help?

I found out why my build is working, it's been so long since i compiled it I forgot I updated (and noted!) the fingerprints lol:

Straight out of Principia\Principia\astronomy\solar_system_fingerprints.hpp in my build tree:

constexpr std::uint64_t KSPStockSystemFingerprint      = 0x63A5CF9AAB1676AE;
constexpr std::uint64_t KSPStabilizedSystemFingerprint = 0x9DCFF6CCCD13551F;

That's for 1.9.

Result on fresh game:

This is the Jool system retrobop'd after the patcher runs on 1.9, after ~2.5 years max time acceleration, max history shown.  My game has been running a similar time with no issues, btw.  Seems similar to your analysis:  Chaotic but stable.

screenshot506.png

screenshot507.png

screenshot508.png

 

Hope these details are helpful!

 

You can view the complete log here as well:

http://glacialsoftware.net/PrincipiaLog.txt

Edited by R-T-B
Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...