raptop changed the topic of #principia to: READ THE FAQ: http://goo.gl/gMZF9H; The current version is Gateaux. We currently target 1.8.1, 1.9.1, and 1.10.1. <scott_manley> anyone that doubts the wisdom of retrograde bop needs to get the hell out | https://xkcd.com/323/ | <egg> calculating the influence of lamont on Pluto is a bit silly… | <egg> also 4e16 m * 2^-52 is uncomfortably large
<discord-> G​rass. — i want to test Principia, is good for noobs ( like me with almost 0 hours on RSS/RO/RP1 ) to start a career with it?
<discord-> b​adgermasher. — I think the answer is probably not immediately.
<discord-> e​xperimentalshells. — It's probably best to either try Principia after you get a few careers into RP1
<discord-> e​xperimentalshells. — Or, if you really want to try immediately, play with Principia on stock for a while
<discord-> l​pg. — counterpoint: with rp1, it'll be a while before principia makes any sort of difference
<discord-> G​rass. — ill try to do my first moon landing and then try principia
<discord-> G​rass. — ill try to do my first manned moon landing and then try principia (edited)
<discord-> S​umguy. — *2 real years later*
<discord-> S​umguy. — *2 real years and 13 restarts later* (edited)
<discord-> P​aculino. — Hey, Principia uses CPU right? Someone on the gregroxmods discord is saying it doesn't
<raptop> I mean, aren't GPU calcs usually pretty low precision?
<bees> they usually use 32-bit float
<bees> idk if Principia uses 64-bit
<raptop> Principia won't even compile on a 32bit machine
<raptop> And AFAIK, never did
<discord-> P​teropodidae. — > Hey, Principia uses CPU right? Someone on the gregroxmods discord is saying it doesn't
<discord-> P​teropodidae. — It most certainly does
<bees> that does not mean that 64-bit floats are used
<discord-> P​teropodidae. — > I mean, aren't GPU calcs usually pretty low precision?
<discord-> P​teropodidae. — Are GPUs even a good way to speed up the calculations Principia does?
<raptop> ah, I misread
<raptop> Yeah, there's also a limit on how much parallelization is practical with normal n-body
<raptop> I want to say that you need to do some weirdness with effective potentials or something for large-n?
<discord-> P​aculino. — @WarriorSabe#6571 had issues after 1400 years of time warping in a custom planet pack, and thinks it is because Principia uses GPU
<discord-> P​aculino. — @WarriorSabe#6571 had issues after 1400 years of time warping in a custom planet pack, and thinks it is because Principia uses GPU very heavily (edited)
<discord-> P​aculino. — > Then why would running it like that cause graphical errors and the screen to stop working, and ultimately crash the game with a "fatal error in gc" message wen the screen doesn't stop working?
<discord-> P​aculino. — > That sounds a lot like the GPU having problems
<discord-> P​aculino. — > It's not that [n-body instability], I know for sure
<discord-> P​aculino. — > Because everything was staying in its orbit perfectly fine
<discord-> P​aculino. — > It was 100% overtaxing some part of my computer
<discord-> P​aculino. — > The only thing that felt buggy was the endless apocalypse errors for a planet that wasn't moving from its orbit, but that was probably just my computer being unable to handle it properly
<discord-> P​aculino. — Can someone help Sabe?
<discord-> K​irk. — endless apocalypse errors doesn't sound "normal"
<discord-> P​aculino. — > I also don't play with full RO so I won't have any use for joining an RO server
<discord-> K​irk. — Also, GC stands for garbage collection
<discord-> P​aculino. — Yeah, but Sabe is certain it is GPU errors on a machine that runs high res planet packs
<discord-> K​irk. — crashing from "fatal error in GC" sounds like the log spam is filling up a bunch of ram and making the GC freak out
<discord-> K​irk. — That or just vessel/planet history
<discord-> K​irk. — That tends to enter the gigabyte range after a few decades, much less centuries
<discord-> P​aculino. — > Fatal error in GC is an error I crash with without Principia
<discord-> P​aculino. — 🤦
<discord-> K​irk. — In that case Unity is just dying from warping the game forwards 1400 years
<discord-> K​irk. — Principia is unrelated
<discord-> P​aculino. — Sabe always has that issue
<discord-> P​aculino. — Sabe always has that issue, and claims it is always due to GPU (edited)
<discord-> K​irk. — I can promise it's Unity badly handling RAM allocations until something kills GC
<discord-> K​irk. — Principia might speed it up
<discord-> K​irk. — But it is not GPU related at all
<discord-> K​irk. — At least, not directly
<discord-> P​aculino. — > i5-8250U for the processor, 8 gigs RAM, MX150 for the card
<discord-> P​aculino. — > How would you go about manually cleaning the history? I just set max history length to like 2 days
<discord-> P​aculino. — > And it started crashing external apps around year 1400
<discord-> P​aculino. — I think Sabe is just trying to use ksp for universe sandbox
<discord-> P​aculino. — 🤦
<discord-> P​aculino. — I'm surprised issues only started then
<discord-> K​irk. — KSP is fairly stable out into the centuries range
<discord-> P​aculino. — > Literally the only thing I was trying to do with it though was run as far into the future as I could to confirm that the stability it showed in US2 also existed with Principia
<discord-> K​irk. — But by the time you hit millenia, Unity starts giving out
<discord-> K​irk. — KSP will not be happy running 1400 years
<discord-> K​irk. — 32-bit KSP was hard limited to ~233 years
<discord-> K​irk. — 64-bit obviously raised that limit a lot
<discord-> P​aculino. — Sabe had no clue if their pc/game were 32 bit or 64
<discord-> K​irk. — But it doesn't stop Unity from giving up after a millenia or two
<discord-> K​irk. — But it doesn't stop Unity from giving up after a millennia or two (edited)
<discord-> K​irk. — It can't be 32-bit, it would've crashed at 233 years
<discord-> P​aculino. — They still think that it is GPU issues....
<discord-> P​aculino. — 🤦 This is bothering me more than it should
<discord-> K​irk. — Why do they think its GPU issues?
<discord-> P​aculino. — they say that GC errors always accompany heavy GPU use
<discord-> P​aculino. — I'm off to bed now
<raptop> the apocalypse thing sounds like questionable planet configurations
<discord-> K​irk. — > they say that GC errors always accompany heavy GPU use
<discord-> K​irk. — That's an interesting belief
<discord-> P​aculino. — > If you run on such low specs too, and seem to know a fair bit about Principia, do you know why it would give an apocalypse warning every few months for a planet which doesn't deviate from its orbit in two millenia?
<discord-> P​aculino. — Apparently they kept going until 1700 years
<discord-> K​irk. — Because its configured badly and is making Principia unhappy. I have no idea why
<discord-> P​aculino. — Apparently they kept going until 1700 years, at which point they had to restart the game (edited)
<discord-> K​irk. — As for GPU and GC, there is almost no correlation between GPU usage and GC errors
<discord-> K​irk. — With how low powered his PC is it's probable that the GPU is running hard all the time
<discord-> P​aculino. — The graphics card is better than mine
<discord-> W​arriorSabe. — ok, I joined for now at least
<discord-> P​aculino. — Goodnight
<discord-> K​irk. — Oh huh, MX150 actually isn't terrible
<discord-> K​irk. — Anyway, the crashes are probably Unity GC being terrible as usualy
<discord-> W​arriorSabe. — If the GC errors aren't actually GPU related, it must have just been a coincidence that it was coming with the GPU stuff then
<discord-> P​aculino. — Good luck
<discord-> K​irk. — Anyway, the crashes are probably Unity GC being terrible as usual (edited)
<discord-> W​arriorSabe. — Anyways, any clue why I'd get apocalypse errors if my planet isn't crashing into anything or otherwise falling out of orbit?
<discord-> K​irk. — The apocalypse errors are the planet pack not being properly configured for Principia. You'd be better off asking the devs about the planet pack you're using
<discord-> W​arriorSabe. — That'd be me; I'm testing the stability of my planet pack
<discord-> W​arriorSabe. — But the system is gravitationally stable
<raptop> Yeah, you'd probably want to simulate the planet in MERCURY6 or something and see if there's a close encounter or something
<discord-> W​arriorSabe. — Like, I went to 1700 years without any of my bodies leaving their orbits
<discord-> W​arriorSabe. — The apocalypse warning was called on the same planet every time
<discord-> W​arriorSabe. — About 2000 times over that 1700 year run
<discord-> W​arriorSabe. — Would it be related to the really high timestep I was using to get to 1700 years?
<discord-> K​irk. — If the errors coincide with your timestep then sure\
<discord-> K​irk. — If the errors coincide with your timestep then sure (edited)
<discord-> K​irk. — I don't think principia was ever tested with month long timesteps
<discord-> W​arriorSabe. — The first one happened on the exact same date for both runs when I was at a different timewarp for each
<discord-> W​arriorSabe. — The second time was 30 million timewarp, but the first time I was only at the default max timewarp by the time of the first error
<discord-> K​irk. — Well I have no idea then, that's a better question for the principia devs, probably tomorrow
<discord-> K​irk. — Since I don't think egg is awake at the moment
<discord-> W​arriorSabe. — 3.3936 million seconds was the time reported
<discord-> W​arriorSabe. — I'll try going that far without using a BTW timestep again real quick
egg|cell|egg has quit [Ping timeout: 204 seconds]
<discord-> W​arriorSabe. — Alright, I caught it right as it happened and it appears that it did a weird skip, like a missed frame or something, and then returned to its normal orbit
<discord-> W​arriorSabe. — At that exact same time again too
<discord-> W​arriorSabe. — Weird thing is, that skip doesn't seem to have been caused by any interactions, and has no permanent effect on its orbit
<discord-> W​arriorSabe. — Definitely timestep
<discord-> W​arriorSabe. — I went back, and warped up to right before the exact time, and it didn't happen at 1x; only when I hit 10000x timewarp did it start the weird 4 broken orbits thing'
<discord-> W​arriorSabe. — Unless the time reported was wrong
<discord-> W​arriorSabe. — I'll go back and warp a few days past but at a lower step
<discord-> W​arriorSabe. — Yeah, no it does actually happen at 1x
Mike` has quit [Ping timeout: 194 seconds]
<discord-> W​arriorSabe. — Exactly at the turn of day 135
<discord-> W​arriorSabe. — It popped up the exact second the clock ticked over, which combined with the complete lack of any apparent triggering event, seems quite suspicious
Mike` has joined #principia
<discord-> K​irk. — That looks very close
<discord-> K​irk. — I can see why Principia thinks it's unstable
<discord-> W​arriorSabe. — So it just doesn't like the proximity?
<discord-> W​arriorSabe. — Orbital period is a few hours
<discord-> S​ir Mortimer. — "fatal error in gc" doesn't sound "a lot like the GPU having problems" at all
<discord-> S​ir Mortimer. — > The only thing that felt buggy was the endless apocalypse errors for a planet that wasn't moving
<discord-> S​ir Mortimer. — your solar system isn't stable. that planet collided with something that was not a vessel. fix that first.
<discord-> S​ir Mortimer. — that orbit there might be way too close. the satellite would probably break up
<discord-> S​ir Mortimer. — and since they are equal in size, so would the main body
<discord-> W​arriorSabe. — No, they are outside their roche limits, I did the math
<discord-> W​arriorSabe. — And nothing is collidong - all errors are for the thing onscreen that is not colliding
<discord-> W​arriorSabe. — Like I said
<discord-> W​arriorSabe. — > for a planet that isn't moving from an orbit
<discord-> W​arriorSabe. — You can't tell me the solar system isn't stable when I surveyed the entire system after the errors and found nothing out of place
<discord-> W​arriorSabe. — What happens is, the second it turns to day 135, I get an apocalypse error from Shard while it is still in its orbit, perfectly fine. Then, it suddenly snaps away from its orbit (despite nothing having come close to it and with no gradual disturbance leading up to that) and follows a non-Keplerian path for 4 orbits, before snapping back into its old orbit as if nothing has happened
<discord-> W​arriorSabe. — Also, we already established that the fatal error in GC errors were just a coincidence
<discord-> W​arriorSabe. — And nothing is colliding - all errors are for the thing onscreen that is not colliding, and stability has been tested to 1700 years while that happened at just 135 days (edited)
<discord-> W​arriorSabe. — Sorry if I sound a little snappy, you're just not the first person to tell me its planets colliding after I already established that nothing has left their orbit or collided with anything
<discord-> W​arriorSabe. — The error itself says that, so if it was true, I wouldn't have brought it up as seeming buggy
<discord-> S​ir Mortimer. — no worries, i just read the backlog and felt the need to mention that GPU does not actually stand for garbage proessing unit 😛
<discord-> S​ir Mortimer. — I guess egg is going to ask for log files first.
<discord-> W​arriorSabe. — It's graphics processing unit, yeah, I just didn't know what gc stood for and so assumed graphics card since it always happened at the same time my gpu was being really stressed and was associated with the resolution going all weird
<discord-> e​gg. — Hm?
<discord-> S​ir Mortimer. — uh oh
<discord-> S​ir Mortimer. — *hides*
<discord-> e​gg. — You got an apocalypse ?
<discord-> W​arriorSabe. — Basically, getting apocalypse errors for a world in a stable orbit
<discord-> e​gg. — The orbit is too fast for the default numerical settings then
<discord-> W​arriorSabe. — Here's a screenshot from almost the second it happened (Shard is the one to the left)
<discord-> W​arriorSabe. — Happens the second the day 135 starts, without fail
<discord-> W​arriorSabe. — The period is just a few hours
egg|cell|egg has joined #principia
<discord-> e​gg. — You'll want to stop the lengthy confused speculation and to adjust the time step of the ephemeris in the numerics blueprint
<discord-> W​arriorSabe. — How do I do that?
<discord-> e​gg. — See the wiki page on the configuration files
<discord-> W​arriorSabe. — I don't really know anything about Principia from the config side
<discord-> W​arriorSabe. — Oh, there's one? Alright
<discord-> e​gg. — And ask gregrox who has some experience with that
<discord-> W​arriorSabe. — Well... tbh I don't really know what I'm looking at in the wiki
<discord-> W​arriorSabe. — It's a bit technical in a way I'm not familiar with
<discord-> W​arriorSabe. — Appears to be referencing certain code libraries or something?
<discord-> W​arriorSabe. — Yeah, I don't understand anything that's saying
<discord-> W​arriorSabe. — That's what I'm reading
<discord-> e​gg. — Look for the description of the blueprint
<discord-> W​arriorSabe. — It uses a bunch of words relating to something I'm not familiar with, presumably internal to Principia's interator
<discord-> W​arriorSabe. — It uses a bunch of words relating to something I'm not familiar with, presumably internal to Principia's integrator (edited)
egg|laptop|egg has joined #principia
<discord-> W​arriorSabe. — THe thing this example is for?
<discord-> e​gg. — yes, though it should be noted that this is not actually the default, it is the one used for RSS
<discord-> W​arriorSabe. — What should I change it to?
<discord-> e​gg. — the default has a different integrator and a time step of 35 minutes
<discord-> W​arriorSabe. — And how would you recommend I do that via MM?
<discord-> e​gg. — @WarriorSabe we are not in the job of debugging solar systems, whether they go too fast or are unstable; figure out what works for yours.
<discord-> e​gg. — and again, you can ask gregrox, who has experience with these settings.
<discord-> W​arriorSabe. — Well, I know its not unstable since I went to 1700 years without instability, I just wasn't sure what a good step size is
<discord-> e​gg. — you don’t know anything
<discord-> W​arriorSabe. — If 35 is default and it uses 10 for RSS, I'll try 10
<discord-> W​arriorSabe. — wdym?
<discord-> e​gg. — you know that if you ignore the warning that tells you « we cannot trust what we show you », nothing obviously wrong happens
<discord-> W​arriorSabe. — Well, fair, though I know that the thing its for is stable for at least 500 years in US2
<discord-> W​arriorSabe. — Though I also get that there are differences; that was the whole reason I went to far with my testing in ksp
Mike` has quit [Ping timeout: 198 seconds]
<discord-> W​arriorSabe. — Either way, I know nothing was out of place at the moment the error occurerd
<discord-> W​arriorSabe. — Either way, I know nothing was out of place at the moment the error occurred (edited)
<discord-> e​gg. — Gregrox has regularly had issues with replicating things in US2; I think the issue mostly lies in transferring the initial states between one and the other, but it may also use questionable methods
<discord-> W​arriorSabe. — Like I said, I get that there are differences and if I didn't I probably wouldn't have gone 1700 years in ksp to check
<discord-> W​arriorSabe. — But greg's also not online rn
<discord-> W​arriorSabe. — For now, how do I edit that blueprint via a MM patch? Do I @ it or not, and should I use a specific pass specifier?
<discord-> e​gg. — I don’t think there is anything for you to @, but also this isn’t an MM support line, you probably know how to use MM better than I do
<discord-> S​tandecco. — you need to declare it first, if there wasn't one defined already, and I don't see why there would be one
<discord-> e​gg. — By the way, the defaults are actually equivalent to
<discord-> e​gg. — ```
<discord-> e​gg. — principia_numerics_blueprint {
<discord-> e​gg. — ephemeris {
<discord-> e​gg. — fixed_step_size_integrator = BLANES_MOAN_2002_SRKN_14A
<discord-> e​gg. — integration_step_size = 35 min
<discord-> e​gg. — fitting_tolerance = 1 mm
<discord-> e​gg. — geopotential_tolerance = 0x1.0p-24
<discord-> e​gg. — }
<discord-> e​gg. — }
<discord-> e​gg. — ```
<discord-> W​arriorSabe. — Not defined by default? I'll go no @ then
<discord-> e​gg. — and the documentation literally has an annex explaining what you should be doing to try to make things work https://github.com/mockingbirdnest/Principia/wiki/Principia-configuration-files#annex-choosing-a-fixed-step-size-integrator
<discord-> e​gg. — > Solar system designers should test the actual stability of their system with a symplectic Runge-Kutta-Nyström method and a small time step, and then try increasing the step size and switching to a linear multistep integrator to reduce computational cost for their users.
<discord-> S​ir Mortimer. — the default is to calculate orbits to 1mm?
<discord-> W​arriorSabe. — idk what that means but I can see if any of the ones in that list look like that name
<discord-> S​ir Mortimer. — as in milli meters?
<discord-> W​arriorSabe. — which they don't
<discord-> W​arriorSabe. — I know nothing about integrators
<discord-> W​arriorSabe. — So idk which of those are the right kind
<discord-> W​arriorSabe. — Unless there's another page listing that?
<discord-> e​gg. — this will tell you what you want
<discord-> e​gg. — e.g., that BlanesMoan2002S6 is a SymplecticPartitionedRungeKutta method, from a paper by Blanes and Moan (2002),
<discord-> e​gg. — called *Practical symplectic partitioned Runge–Kutta and Runge–Kutta–Nyström methods*
<discord-> e​gg. — @WarriorSabe but just stick with BLANES_MOAN_2002_SRKN_14A for the SRKN and QUINLAN_TREMAINE_1990_ORDER_12 for the linear multistep method
<discord-> W​arriorSabe. — So do the blanes one and like a 1min timestep for checking stability, then the quinlan and whatever the highest timestep that works is after I've confirmed it is?
<discord-> e​gg. — it gets a bit more fun than that
<discord-> e​gg. — because the multistep method will need a shorter timestep than the single-step one
<discord-> W​arriorSabe. — Still, highest that works is what I'm looking for, right?
Mike` has joined #principia
<discord-> e​gg. — vague heuristic: try blanes 14A with 1 min, try increasing that timestep, then divide it by 14 and explore quinlan around there
<discord-> e​gg. — but be very careful with the quinlan: it can easily be broken in ways that don’t warn about apocalypses
<discord-> e​gg. — you can have weirdly shaky planets that still fit
<discord-> W​arriorSabe. — Like, seconds for the quinlan? Wasn't the rss one 10 min?
<discord-> e​gg. — yes, but this was tuned specifically for RSS
<discord-> W​arriorSabe. — It's a pretty big difference, but I can just binary search and so it probably won't affect how many checks it'll take
<discord-> W​arriorSabe. — It's a pretty big difference, but I can just binary search and so it probably won't significantly affect how many checks it'll take (edited)
<discord-> e​gg. — I want to insist that you want to check not only that you don’t get an error message, but also that you can’t tell which configuration you are using by looking at where the planets are along their orbits after a while
<discord-> W​arriorSabe. — Yeah, I'd check that they're in the right place either way
<discord-> e​gg. — it’s easy to have a poorly-converged method that still doesn’t trip the apocalypse check
<discord-> W​arriorSabe. — Something getting thrown into the abyss is much more common than things smashing into each other
<discord-> e​gg. — well, they will be in their right place initially, and they will likely stay on their orbits, that is not what I mean
<discord-> e​gg. — what I mean is where they are along their orbits after, say, 10 years
<discord-> W​arriorSabe. — Oh, like mean anomalies?
<discord-> e​gg. — yeah
<discord-> S​tandecco. — is numerical things you can get results that look correct but aren't actually correct, and depend on the method that you're using; you need to check that both methods have the same _exact_ results
<discord-> W​arriorSabe. — Should be simple enough to check
<discord-> W​arriorSabe. — I can just write down where everything is at some chosen time, and then check against that
<discord-> e​gg. — I see @Standecco has played with numerical methods
<discord-> e​gg. — how’s uni
Mike` has quit [Ping timeout: 204 seconds]
<discord-> S​tandecco. — bit tough, but interesting
<discord-> S​tandecco. — mostly getting into electromagnetic fields and electronics atm, I switched from aerospace to physics engineering
<discord-> S​ir Mortimer. — > > 3.14
<discord-> S​ir Mortimer. — > < 3.14
<discord-> S​ir Mortimer. — > > 3.14 * 2
<discord-> S​ir Mortimer. — > < 6.28
<discord-> S​ir Mortimer. — > > 3.14 * 2.2
<discord-> S​ir Mortimer. — > < 6.908000000000001
<discord-> S​ir Mortimer. — ^ numerical things.
<discord-> W​arriorSabe. — Anyways, it's too late tonight to check that stuff except maybe start a game running overnight with a low timestep. I'd also like to define gravitational potential models for some of the more aspherical bodies, but I don't really know anything about the spherical harmonics I'm told it uses. Just how complicated is it?
<raptop> Having all of our back end calculations in a different base from what we want to represent them in is a good idea, and cannot cause any problems at all
<discord-> e​gg. — @WarriorSabe stop rambling about low timestep
<discord-> W​arriorSabe. — What?
<discord-> e​gg. — timewarp does not control the time step
<discord-> W​arriorSabe. — You just said to use a low timestep for initial stability testing
<discord-> e​gg. — ah yes, in that sense
<discord-> W​arriorSabe. — I didn't say anything about timewarp
<discord-> e​gg. — not awake and replying to your forum post
<discord-> W​arriorSabe. — Oh, that was a long time ago
<discord-> e​gg. — which was confused about timestep vs. timewarp
<discord-> e​gg. — 6 hours
<discord-> W​arriorSabe. — Before you explained how the integrators work
<discord-> S​tandecco. — where are you now, egg? Tomorrow we go into lockdown for the second time
<discord-> e​gg. — I am in CH
<discord-> e​gg. — where they have no idea what they are doing, but federally
<discord-> W​arriorSabe. — Long time in the context of this conversation
<discord-> e​gg. — I should have gone back to pet my cat in FR
<discord-> S​tandecco. — oh, poor cat
<discord-> S​tandecco. — > where they have no idea what they are doing, but federally
<discord-> S​tandecco. — that would also describe the US really well
<discord-> W​arriorSabe. — Anyways, about my other question - how complicated is setting up a gravity model, given that I don't have any prior experience regarding spherical harmonics (which I'm told is used in there)
<discord-> W​arriorSabe. — Anyways, about my other question - how complicated is setting up a gravity model, given that I don't have any prior experience regarding spherical harmonics (which I'm told is used in there)? (edited)
<discord-> e​gg. — well, how else would you define it
<discord-> W​arriorSabe. — What? I was just asking how complicated it was
<discord-> e​gg. — what is « it » here :-p
<discord-> e​gg. — i.e. if you are not starting from an expansion in spherical harmonics, how do you even define your gravity model to start with (regardless of the difficulty of translating that to spherical harmonics)
<discord-> W​arriorSabe. — I don't understand what you're asking, but you seem to be trying to rebut an argument that it wouldn't need spherical harmonics, which was never an argument being made
<discord-> W​arriorSabe. — I was just asking how complicated the spherical harmonics used to create the gravity model was to learn and use
<discord-> e​gg. — no
<discord-> e​gg. — I am asking you what you are trying to do
<discord-> W​arriorSabe. — Oh, make the gravity correct for aspherical bodies
<discord-> e​gg. — define correct
<discord-> e​gg. — have you defined the mass distribution inside those bodies
<discord-> W​arriorSabe. — Like, more closely approximating than assuming it is a sphere
<discord-> S​tandecco. — he's saying that spherical harmonics are a tool to model a gravitational model that isn't perfectly spherical, but you first need a model
<discord-> W​arriorSabe. — I mean, I know what the mass distribution should look like
<discord-> W​arriorSabe. — What kind of model specifically?
<discord-> e​gg. — are they homogeneous cows of uniform density? Is there a metallic core?
<discord-> W​arriorSabe. — For a place to start, how would I go about it for an oblate spheroid?
<discord-> e​gg. — no again you are missing the point
<discord-> e​gg. — you are talking about the figure
<discord-> e​gg. — but you need the mass distribution
<discord-> W​arriorSabe. — Metallic core, rocky mantle, icy crust with a subsurface ocean
<discord-> W​arriorSabe. — Like a spun-up Enceladus
<discord-> W​arriorSabe. — Like a spun-up Enceladus with a metallic core (edited)
<discord-> W​arriorSabe. — I know I need the mass distribution
<discord-> e​gg. — OK, so assuming you quantify that properly you would need to somehow plug it into something that knows how to compute the gravitational field of your mass distribution, and then expand that into spherical harmonics
<discord-> W​arriorSabe. — All I was originally asking was how hard that was to do
<discord-> e​gg. — I have no idea ?
<discord-> e​gg. — I just fetch the spherical harmonics that various astronomical agencies give for the solar system
<discord-> W​arriorSabe. — Which I would follow up with, how exactly would that be done and what tools would do that? But, if you don't know, do you know anyone who does?
<discord-> W​arriorSabe. — At a bare minimum I would at least like to approximate bodies with homogenous ellipsoids. Preferably more, but that may be too complex to do
<discord-> e​gg. — I know that there is a group at the astronomical institute of the university of Bern who works on a question that involves this for Callisto
<discord-> W​arriorSabe. — Well, that doesn't really do much for me but imply that it's not a very easy thing to do
<discord-> e​gg. — but I should note that for practical purposes (low degree and order), the geopotential will not really be related to the topography
<discord-> W​arriorSabe. — Yeah, ik
<discord-> e​gg. — so basically you can completely ignore what your model looks like, which should make your life much easier
<discord-> W​arriorSabe. — For most worlds, yeah
<discord-> W​arriorSabe. — There's only a handful I'm concerned with
<discord-> W​arriorSabe. — Such as this one
<discord-> e​gg. — (and you can invent gravity anomalies as appropriate to make your bodies more interesting)
<discord-> e​gg. — ooookay yeah that is not round.
<discord-> e​gg. — what is the internal composition of that thing?
<discord-> W​arriorSabe. — That's the one I'm most concerned with getting a model for, at least a barebones one that ignores mascons
<discord-> W​arriorSabe. — Gimme a sec, I made a composition diagram of all my worlds including that one
<discord-> S​tandecco. — osmium crust with hydrogen core
<discord-> W​arriorSabe. — Metallic core, rocky mantle, ice crust with subsurface ocean
<discord-> W​arriorSabe. — Inner layers are less oblate due to their higher densities
<discord-> e​gg. — ookay we are not out of the woods.
<discord-> e​gg. — maybe we can assume that the non-oceanic part behaves a bit like earth, and stretch something like GRS80 on it, but then we have to deal with that ocean
<discord-> W​arriorSabe. — I can probably approximate the oblateness of an equivalent constant-density spheroid
<discord-> W​arriorSabe. — But once I have that, I'm not sure how to turn it into a gravity model. Presumably an oblate spheroid with no mascons would be simple to do, at least relatively speaking
<discord-> W​arriorSabe. — With it being so oblate, I'd expect that the oblateness would be a much larger driver than any localized mascons
<discord-> W​arriorSabe. — And just approximating it with a single spheroid would be akin to approximating the earth as a sphere
<discord-> e​gg. — OK so that would basically be going for the GRS80 approach.
<discord-> e​gg. — Define a reference ellipsoid, compute the normal potential for it, use that.
<discord-> W​arriorSabe. — How much work would that be? It's too late atm to start work, so mostly I just want to know how much it'll take so I know where I can put it in my priority list
<discord-> e​gg. — (let’s use standard terminology and say « ellipsoid », we don’t need more confusion)
<discord-> e​gg. — if you have the axes of your reference ellipsoid, you will get the coefficients from its normal potential by standard formulae, somewhere around here https://geodesy.geology.ohio-state.edu/course/refpapers/00740128.pdf
<discord-> e​gg. — p. 130 of that PDF has the zonal harmonics
<discord-> W​arriorSabe. — So it's just a couple formulae? Shouldn't be too much then, unless they're particularly atrocious. I'll take a look
<discord-> e​gg. — beware of the normalization though
<discord-> W​arriorSabe. — How come?
<discord-> e​gg. — it is a complete mess, and I should note that Wikipedia is a particularly bad offender there
<discord-> W​arriorSabe. — Oh, so it's spherical rather than elliptical? Should be easy enough to convert; I can probably just adapt the math I used to write the generator that made the moon's ellipsoid map in the first place to convert backwards
<discord-> e​gg. — what are you talking about
<discord-> W​arriorSabe. — Unless it's to be done somewhere internally
<discord-> e​gg. — what
<discord-> W​arriorSabe. — Or if I'm misunderstanding it
<discord-> e​gg. — I am not sure what you are purporting to understand even
<discord-> W​arriorSabe. — The thing you linked me
<discord-> W​arriorSabe. — The last one
<discord-> e​gg. — it is telling you about the normalization
<discord-> W​arriorSabe. — You said to beware; were you referring to what this part is saying?
<discord-> e​gg. — No, that is not the section to which I linked
<discord-> W​arriorSabe. — The part before didn't have anything that seemed like I'd need to "beware"; it was just showing the formatting
<discord-> e​gg. — I liked to the section that starts with the words
<discord-> e​gg. — > The coefficients 𝐶𝑛𝑚 and 𝑆𝑛𝑚 may be given [...]
<discord-> W​arriorSabe. — Oh wait it went to a different spot the second time
<discord-> W​arriorSabe. — The link just messed up for some reason then
<discord-> W​arriorSabe. — Sent me to the wrong part of the page
<discord-> e​gg. — you clicked on the big thingy expanded by discord, presumably
<discord-> e​gg. — that loses the anchor of the link
<discord-> W​arriorSabe. — discord didn't expand anything; it just seemed to scroll down a section for some reason
<discord-> W​arriorSabe. — wikipedia and atomic rockets does that to me sometimes too
<discord-> W​arriorSabe. — Still, just looks like a bunch of rather straightforward formulae, so it shouldn't be too bad, at least for a simple ellipsoid
egg|laptop|egg has quit [Remote host closed the connection]
<discord-> W​arriorSabe. — Though, one issue - the part you linked shows it in the code, but the MM syntax I'm looking at for the included RSS configs doesn't seem to match that well at all, so when I get to that I'll need to know how to convert. That's for when I get time to do that though
egg|laptop|egg has joined #principia
<discord-> W​arriorSabe. — So, last thing - are these referring to the rotation axis?
<discord-> e​gg. — yes, again see the documentation of the config, it refers to a document that defines what those angles are
<discord-> W​arriorSabe. — Oh, right, the config page; forgot about that. It is getting pretty late, I should probably just wait until tomorrow to ask the rest of my questions
<discord-> e​gg. — it is 10:31
<discord-> W​arriorSabe. — Not here
<discord-> e​gg. — it is 10:31 eggtime
<discord-> e​gg. — that is early
<discord-> W​arriorSabe. — It's late relative to when I woke up
<raptop> Given the lockdown comment, I'm assuming that it's 09:33 where sabe is
<discord-> e​gg. — I thought @Standecco was the one who commented on lockdown?
<discord-> e​gg. — but standecco is in eggtime
<discord-> W​arriorSabe. — Yeah, it was standecco
<discord-> S​tandecco. — I am in eggtime, yeah
<discord-> S​tandecco. — I'm also almost in egglocation
<discord-> S​tandecco. — but not quite
<discord-> e​xperimentalshells. — you mean in ECI reference frame?
<raptop> ah
<discord-> W​arriorSabe. — I'm probably about a third of the way around the world from there
* raptop blames it on being 04:36 my time
<discord-> e​xperimentalshells. — interestingly, my latitude is almost exactly the same as egg
<discord-> e​gg. — @experimentalshells Seattle? Budapest? Innsbruck? Ростов-на-Дону? Nantes?
<discord-> W​arriorSabe. — 40-50N seems to be where a majority of users live
<discord-> W​arriorSabe. — I'm at 46
<discord-> e​xperimentalshells. — somewhat close to seattle, not very
<discord-> e​xperimentalshells. — *in the general area*
<discord-> W​arriorSabe. — southeastern washington
<discord-> e​gg. — 齐齐哈尔 was also an option
<discord-> S​ir Mortimer. — Zürich?
<discord-> e​gg. — yes
<discord-> S​ir Mortimer. — 92 minutes from here.
<discord-> S​ir Mortimer. — as in drive time, not arc minutes.
<raptop> also not minutes of RA
<discord-> W​arriorSabe. — or minutes of latitude, presumably
<discord-> W​arriorSabe. — or lightminutes
UmbralRaptop has quit [Quit: Bye]
UmbralRaptop has joined #principia
<discord-> S​ir Mortimer. — 124km on asphalt
<discord-> W​arriorSabe. — I should probably get to bed
<discord-> S​ir Mortimer. — But it warns me that the route crosses the EU border.
<discord-> S​ir Mortimer. — (which, from experience, noone cares about)
<discord-> e​gg. — technically correct
<discord-> e​gg. — (but it’s been an intraschengen border for a while so indeed no-one cares)
<discord-> e​gg. — (unless you are a truck because it is a customs border)
raptop has quit [Ping timeout: 189 seconds]
<discord-> S​ir Mortimer. — I crossed that border up to 4 times a day for a while, because it was the shortest route to work. Nobody seemed to be interested.
<discord-> S​ir Mortimer. — Only once was I stopped, and he didn't ask for my ID, he asked for my driving license.
<discord-> e​gg. — that does tell them you are not swiss though, because CH is weird with that
<discord-> S​ir Mortimer. — One time I came back from St. Gallen with a car filled to the brim with IKEA furniture, definitely more than the allowance. On that day, they checked everyone - except me.
<discord-> e​gg. — you need to turn your driving license into a CH one within a couple of years of immigrating
<discord-> e​gg. — if you fail to do so, you have a license with which you can drive everywhere except in iCH
<discord-> e​gg. — if you fail to do so, you have a license with which you can drive everywhere except in CH (edited)
<discord-> W​arriorSabe. — I'm going to leave you guys to your EU talk then
<discord-> e​gg. — CH is not in the EU :-p
<discord-> W​arriorSabe. — idek what CH is referring to
<discord-> e​gg. — Switzerland
<discord-> e​gg. — *Confoederatio Helvetica*
<discord-> e​xperimentalshells. — well, CH hasn't had a long history of being involved in anything really
<discord-> W​arriorSabe. — Well, I live an ocean and a continent away, so none of this means anything to me or is relevant to me. I'll come back when I need more help with Principia configuration
<discord-> e​gg. — goodnight
<discord-> e​gg. — @Sir Mortimer 124 km is 67 nautical miles, so that would be about 67 minutes of arc :-p
<discord-> e​gg. — @Sir Mortimer 124 km is 67 nautical miles, so that would be about 67 arcminutes :-p (edited)
egg|laptop|egg has quit [Remote host closed the connection]
Blu3wolf has joined #principia
<discord-> S​ir Mortimer. — homeworking this afternoon. i nicked a fountain pen from the kids, and now that I've used it I'm thinking about getting one for myself 😛
<discord-> S​ir Mortimer. — always had a weakness for stationery
<discord-> P​aculino. — You can make a nib rather easily with a piece of metal
egg|laptop|egg has joined #principia
<discord-> e​gg. — fountain pens are nice
<discord-> P​aculino. — Yeah
egg|laptop|egg has quit [Remote host closed the connection]
<discord-> T​heSquirrelPatrol. — Is anyone having issues with principia where having it installed makes a new rp-1 career start in 2014, not 1951?
<discord-> P​aculino. — Someone on reddit had that problem and asked about it today
egg|cell|egg has quit [Ping timeout: 189 seconds]
egg|cell|egg has joined #principia
<discord-> T​heSquirrelPatrol. — Any clue on whether this is more likely to be an actual principia issue, or something in the new rp-1 release?
<discord-> T​heSquirrelPatrol. — but its definitely the combo of the two, with principia, my career starts at 2014, without, 1951
raptop has joined #principia
egg|laptop|egg has joined #principia
<discord-> e​gg. — @Specimen.Spiff what exactly did you install
<discord-> e​gg. — @Specimen.Spiff more specifically, what did you do to install Principia
<discord-> T​heSquirrelPatrol. — I grabbed the Principia-2020101620-Gauss.zip from github, unzipped it, renamed folder to principia and put it in gamedata
<discord-> e​gg. — I regret to inform you that you failed the reading comprehension test
<discord-> e​gg. — please read the readme https://github.com/mockingbirdnest/Principia/tree/master/README.md
<discord-> e​gg. — and the FAQ
<discord-> e​gg. — (you ended up:
<discord-> e​gg. — 1. not getting Principia (the binaries are not in the repository that you downloaded),
<discord-> e​gg. — 2. getting a configuration file for the TRAPPIST-1 system which changes the date to December of 2014)
<discord-> T​heSquirrelPatrol. — I went to the relase page, which had a binaries link, and I dont think the readme was obvious from there
<discord-> e​gg. — the releases page did not have a binaries link.
<discord-> e​gg. — It had a source code link.
<discord-> e​gg. — That is only useful if you want to build Principia yourself, which is an interesting can of worms, but probably not what you are looking for.
<discord-> T​heSquirrelPatrol. — nope, I get enough of that at my paying job 🙂
<discord-> T​heSquirrelPatrol. — next lunation is in what, a week or so?
<discord-> e​gg. — ten days apparently
egg|laptop|egg has quit [Remote host closed the connection]
<discord-> T​heSquirrelPatrol. — yep, just googled.
<discord-> T​heSquirrelPatrol. — I know you release each lunar month, is on the day exactly, or just as close as works out for you?
<discord-> T​heSquirrelPatrol. — actuyally rephrase that to "I believe"
<discord-> S​tonesmile. — The mod is released *at the correct minute*, according to the observatory in Paris IIRC
egg|laptop|egg has joined #principia
Mike` has joined #principia
Mike` has quit [Ping timeout: 194 seconds]
Mike` has joined #principia
Mike` has quit [Ping timeout: 198 seconds]
Mike` has joined #principia
<discord-> e​gg. — at least we aim for that
<discord-> e​gg. — we have been a few hours late on a couple of occasions
<discord-> e​gg. — the upgrade reminder goes off at the correct minute so that is an incentive for having it in time :-p
Mike` has quit [Ping timeout: 189 seconds]
Mike` has joined #principia
<_whitenotifier-9244> [Principia] zroug starred Principia - https://git.io/JTpWd
Mike` has quit [Ping timeout: 194 seconds]
Mike` has joined #principia
Blu3wolf has quit [Quit: Konversation terminated!]
Blu3wolf has joined #principia
Blu3wolf has quit [Quit: Konversation terminated!]
egg|laptop|egg has quit [Remote host closed the connection]
egg|laptop|egg has joined #principia
egg|laptop|egg has quit [Remote host closed the connection]
egg|laptop|egg has joined #principia
raptop has quit [Ping timeout: 189 seconds]