00:36 <awang> nornagon: Yep, running on 1.3.0 right now

00:37 <awang> Mostly curious since I've been intending on looking into some bugs and was hoping that 1.3.1 would provide a good excuse to stop my rp-0 career

01:08 <awang> egg|work|egg: Would a 10Mm apogee be far enough to have significant gravitational effects from the moon?

15:54 <awang> egg|work|egg: Is something like this normal for an orbit with an apogee of roughly 10Mm and roughly in the same plane as the moon's orbit?

15:54 <awang> https://imgur.com/a/0h0XL

15:54 <kmath> https://i.imgur.com/Uanelz6.jpg

17:37 <egg> awang: uh this looks confusing

17:37 <egg> awang: there seems to be a perfectly normal elliptic orbit, and also something weird?

17:38 <egg> awang: what does the prediction look like

17:38 <egg> UmbralRaptor: any idea what we're seeing?

17:40 <awang> egg: The weird thing is what I was wondering about

17:40 <UmbralRaptor> 10 Mm isn't *that* much in the real world. Uh, which reference frame is this? o_O

17:40 <awang> I'll have to get back to you with the prediction. In the middle of a long flight

17:41 <awang> This is ECI, IIRC

17:41 <awang> ECEF is even weirder

17:42 <awang> Unless there's a way of seeing the prediction without switching craft?

17:42 <egg> well ECEF will be weird because the earth rotates of course

17:42 <egg> awang: what did the craft do in the past? because we're seeing the history here

17:43 <awang> egg: Hard to say for sure. That was my first artifical satellite, so I stayed with it long enough to get it into orbit and transmit some science, then left to start building the next craft

17:44 <awang> So stayed with the craft for an orbit at most

17:44 <egg> did you at some point timewarp a lot?

17:44 <awang> Yes, when waiting for another rocket to build

17:44 <egg> (e.g. from the space centre view at max warp or something)

17:44 <awang> Yep

17:44 <egg> awang: hmm, then that might be extreme downsampling

17:44 <egg> and then cubic interpolation thereof

17:45 <egg> #228

17:45 <Qboid> [#228] title: Keep the whole history | For the moment we compute the histories using `sampling_period == 0`, which results in large gaps in the trajectory at high timewarp (rendered as long straight line segments, see below). We should use `sampling_period == 1` instead, and downsample ourselves as needed when rendering.... | https://github.com/mockingbirdnest/Principia/issues/228

17:47 <awang> Any way for me to get you more info to see if that is actually what is going on?

17:47 <egg> we need to kill that bug at some point

17:47 <egg> awang: well you should see that by launching a sat in LEO, maxwarping in space centre view, and looking at the silly trajectory

17:48 <egg> awang: we really need to stab that bug, it's silly

17:48 <egg> stabbity stab

17:48 <egg> certainly won't make it in Chasles

17:48 <egg> maybe something for 陈景润

17:49 <awang> Is it just a rendering thing, or could it actually result in something like a satellite near the 140km atmospheric cutoff falling into the atmosphere?

17:49 <egg> it's strictly a rendering thing

17:49 <egg> the actual computation is no less accurate

17:49 <egg> we just drop intermediate values and thus we render shit

17:49 <awang> Hmmm

17:49 <awang> Guess that's also why I don't see actual straight lines in my screenshot?

17:53 <egg> awang: what do you mean?

17:54 <egg> the reason why there are no straight lines is that we do cubic interpolation, so you get cubics

17:54 <awang> So why the straight lines in the screenshot in the issue?

17:54 <egg> older plotting implementation, which did linear interpolatino

17:55 <egg> issue's old

17:55 <awang> Ah

17:55 <awang> Makes sense

17:55 <egg> cubics are kinda worse when the data's too sparse tbh

17:55 <egg> (they yield those silly loops)

18:06 <GH> [Principia] pleroy pushed 2 new commits to rotating-world: https://git.io/vd2Fu

18:06 <GH> Principia/rotating-world 4e8c3ee Pascal Leroy: Merge pull request #1589 from eggrobin/reunification-point-8...

18:06 <GH> Principia/rotating-world 84b278d Robin Leroy: compatibility with existing FAR, ignore the silly flags

18:43 <egg> awang: re. 1.3.1, try recompiling and changing the version check from 1.3.0 to 1.3.1, if that works I'll make an additional build config and make a triple release I guess (blarg)

18:48 <awang> egg: Is the version check in ksp_plugin_adapter.cs?

19:18 <egg> awang: yeah

19:18 <egg> #if KSP_VERSION_1_2_2 etc

19:51 <GH> [Principia] eggrobin opened pull request #1590: A crash when crashing due to binding before freeing (rotating-world...reunification-point-8) https://git.io/vdakv

20:18 <awang> egg: Here's another satellite after some timewarping

20:18 <awang> https://i.imgur.com/bqkqtNz.jpg

20:18 <awang> This one is in a polar orbit

20:27 <egg> awang: yeah see how it does a weird loop from a place that's on its orbit to another place that's three quarters of an orbit later? that's an interpolating cubic interpolating too long a segment

20:28 <egg> awang: we'll try looking at that for 陈景润

20:28 <egg> at least with Chasles moon landings will work nicely

20:31 <awang> egg: I see. Makes sense

20:31 <awang> I think moon landings are slightly more important than weird rendering :P

20:32 <awang> Also, remember that SIGILL issue I've been having?

20:32 <awang> Seems that it occurs when compiled at -O3, but not at -O0

20:32 <awang> So either got some undefined behavior or miscompilation

20:32 <awang> I think

20:42 <egg> Ꙩ_ꙩ

21:57 <GH> [Principia] eggrobin commented on issue #228: @aw1621107 reports impact; the visual appearance has changed with the Чебышёв interpolated plotting, we now get weird loops instead of straight segments:... https://git.io/vdaCN

22:29 <awang> egg: Get SIGILL with -O1 at least

22:29 <awang> Trying -Og