UmbralRaptor changed the topic of #principia to: READ THE FAQ: http://goo.gl/gMZF9H; The current version is del Ferro. We currently target 1.5.1, 1.6.1, and 1.7.x. <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|zzz|egg has quit [Remote host closed the connection]
<discord->
DRVeyl. — @egg I'm trying to work on a bug that @lpg and I think @neph noticed with RealAntennas and Principia. The behavior is not seen outside of Principia. RA has a Unity gameobject placed on the surface of the homeworld, it's transform is parented by the CelestialBody's transform. In a stock universe, it rotates around with the planet and properly moves through space. With Principia, everything is cor
<discord->
DRVeyl. —
<discord->
DRVeyl. — Stock CommNet has an occlusion test, which is "just" a raycast between two Vectors hitting or missing an occluder (object defined by position & radius). I get an unexpected result from the occlusion test between the vessel and the surface gameobject, and if I draw a line between them, the answer is clearly wrong.
<discord->
lpg. — neph was seeing something else I misinterpreted, I think
<discord->
DRVeyl. — Does Principia try to manipulate (some?) gameobject transforms?
<discord->
neph. — I don't believe I've seen that yet, but it's possible I have encountered it and just not noticed. I don't keep good track of where ground stations are located.
<discord->
DRVeyl. — @egg I'm trying to work on a bug that @lpg ~~and I think @neph ~~ noticed with RealAntennas and Principia. The behavior is not seen outside of Principia. RA has a Unity gameobject placed on the surface of the homeworld, it's transform is parented by the CelestialBody's transform. In a stock universe, it rotates around with the planet and properly moves through space. With Principia, everything i
<discord->
DRVeyl. —
<discord->
DRVeyl. — Stock CommNet has an occlusion test, which is "just" a raycast between two Vectors hitting or missing an occluder (object defined by position & radius). I get an unexpected result from the occlusion test between the vessel and the surface gameobject, and if I draw a line between them, the answer is clearly wrong. (edited)
<discord->
DRVeyl. — @egg I'm trying to work on a bug that @lpg ~~and I think @neph ~~ noticed with RealAntennas and Principia. The behavior is not seen outside of Principia. RA has a Unity gameobject placed on the surface of the homeworld, it's transform is parented by the CelestialBody's transform. In a stock universe, it rotates around with the planet and properly moves through space. With Principia, everything i
<discord->
DRVeyl. —
<discord->
DRVeyl. — Stock CommNet has an occlusion test, which is "just" a raycast between two Vectors hitting or missing an occluder (object defined by position & radius) centered on the planet. I get an unexpected result from the occlusion test between the vessel and the surface gameobject, and if I draw a line between them, the answer is clearly wrong. (edited)
Jesin has quit [Ping timeout: 190 seconds]
<discord->
DRVeyl. — Some test notes from lpg with 2 vessels, and wherein "broken" means the vessel is connecting to sites that do not seem to occlude correctly:
<discord->
DRVeyl. — - before launch: vessel 1 in orbit is broken
<discord->
DRVeyl. — - vessel 2 on the pad: vessel 1 in orbit is fixed
<discord->
DRVeyl. — - switch to vessel 1: vessel 1 is broken
<discord->
DRVeyl. — - switch to vessel 2 in the pad: vessel 1 is re-fixed
<discord->
DRVeyl. — [revert to pre-launch and re-launch, to remove any x factors]
<discord->
DRVeyl. — - vessel 2 in orbit: both vessels are fine
<discord->
DRVeyl. — - go to space center, then tracking station: both vessels are broken
<discord->
DRVeyl. —
<discord->
DRVeyl. — It seems like there are some rotations that are happening at some times but not others.
<discord->
DRVeyl. — Some test notes from lpg with 2 vessels, and wherein "broken" means the vessel is connecting to sites that should be occluded, and not to ones that should not be occluded:
<discord->
DRVeyl. — - before launch: vessel 1 in orbit is broken
<discord->
DRVeyl. — - vessel 2 on the pad: vessel 1 in orbit is fixed
<discord->
DRVeyl. — - switch to vessel 1: vessel 1 is broken
<discord->
DRVeyl. — - switch to vessel 2 in the pad: vessel 1 is re-fixed
<discord->
DRVeyl. — [revert to pre-launch and re-launch, to remove any x factors]
<discord->
DRVeyl. — - vessel 2 in orbit: both vessels are fine
<discord->
DRVeyl. — - go to space center, then tracking station: both vessels are broken
<discord->
DRVeyl. —
<discord->
DRVeyl. — It seems like there are some rotations that are happening at some times but not others. (edited)
Mike` has quit [Ping timeout: 202 seconds]
Mike` has joined #principia
egg|zzz|egg has joined #principia
egg|zzz|egg has quit [Ping timeout: 378 seconds]
Jesin has joined #principia
<discord->
Sir Mortimer. — What do you need that object for?
<discord->
Damien. — @DRVeyl 90° turn sounds like the issue I had with spin stabilised vessels not having their orientation changed to match the universe rotation principia does to do axial tilt
<discord->
egg. — @DRVeyl the problem is that we don't rotate some things while rotating the universe around then, rather than the reverse
<discord->
egg. — I don't think you thing can work nicely with our axial tilt; you would have to use the scaled space transform
<discord->
egg. — Also as @Sir Mortimer says, why do you need the object instead of just the body
<discord->
DRVeyl. — @egg Need the object because it is a ground station for communication, and being obstructed/occluded by the body surface is important. Ie it's location on the surface is relevant.
<discord->
egg. — I mean, you could just do the maths yourself instead of delegating it to the Unity object logic
<discord->
egg. — but anyway, you want to use the scaled body, that one will have the correct orientation
<discord->
egg. — (obviously it lives in scaled space, so you can't just have an object there and work with non-scaled things elsewhere, but this should give you the proper orientation and the rest is a simple matter of geometry)
<discord->
Sir Mortimer. — Kerbalism does the same to locate the geomagnetic pole
<discord->
Sir Mortimer. — all I need to know is lat/lon, the rest is calculations.
<discord->
egg. — simple matter of geometry
<discord->
Sir Mortimer. — "simple" is relative 🙂
<discord->
neph. — watch it be a stock mechanic in ksp 2
<UmbralRaptor>
The KSP2 devs seem to have said that it won't have stock n-body (Rask & Rusk use some sort of clever hack), so…
egg|zzz|egg has joined #principia
egg|zzz|egg has quit [Remote host closed the connection]
egg|zzz|egg has joined #principia
<discord->
lamont. — my guess would be that rask + rusk are 2-body physics within their SOI, planets themselves run on "rails" where they orbit each other. not sure what they do for ships-on-rails in that SOI at high warp, but they could use a higher order integrator rather than solving keplers problem on rails.
<discord->
Sir Mortimer. — Nobody is way too hard for a game.
<discord->
Sir Mortimer. — And for the devs I might add
<discord->
Sir Mortimer. — Nbody is way too hard for a game.
<discord->
Sir Mortimer. — And for the devs I might add (edited)
egg|zzz|egg has quit [Remote host closed the connection]
<UmbralRaptor>
I mean, granted, Principia won't compile without a 64bit processor, and long term ephemeris calcs will use 4+ GiB of RAM…
egg|zzz|egg has joined #principia
<discord->
egg. — @lamont what are r[au]sk?
<discord->
egg. — or umbralraptor
<discord->
lamont. — KSP2
<discord->
egg. — meow
<UmbralRaptor>
KSP2 planets in a very close binary
<discord->
egg. — ah
<discord->
egg. — yeah I suppose they went the Sigma route
<discord->
egg. — huh, sigma isn't on this server
<discord->
DRVeyl. — I can recalculate, but I also need to maintain the basic functionality of CommNet because I don't completely replace it, I rewrite several specific functions. Ideally, I am doing things the way they do.
<discord->
DRVeyl. — On the plus side, it means if Principia doesn't cause this same problem with stock CommNet stations, then I need to hunt more closely for how/why my stations are different.
<egg>
stock as in not RSS?
<egg>
RSS + Principia has tilt, stock + Principia doesn't
<egg>
and, well, Principia having tilt where stock doesn't, by definition, violates some assumptions that are true in stock; if you build upon something that relies on those assumptions, you can't be compatible with Principia
<egg>
(I have no idea whether stock commnet works correctly with principia anyway)
<egg>
or if commnet does rely on things that are not true, at the very least you (or we) would need to fool it into believing something more appropriate
egg|zzz|egg has quit [Remote host closed the connection]
egg|zzz|egg has joined #principia
<discord->
DRVeyl. — Stock as in not RSS, but my test can be converted to RSS by removing all but one of the defined ground stations that Kopernicus will instantiate.
<discord->
DRVeyl. — Well, sort of. I'd trust the true stock CommNet objects more than Kopernicus' creation of it.
<discord->
DRVeyl. — I guess the short form is if you rotate the CelestialBody's transform directly, or if you just trick that.