Inaccurate atmospheric temperature calculations

[prestja]

https://forum.kerbalspaceprogram.com/index.php?/topic/194936-191-kopernicus-continued/&do=findComment&comment=3809504

[Keith-OHara]

The panelsFix2 branch at src/Kopernicus/Components/KopernicusStar.cs recomputes GetAtmoThermalStats for some reason, in the new code from commit 07230f3 "CalculateFlux => CalculateFluxAt"

[R-T-B]

Good catch. I'm sure that was done as part of the multistar support for solar panels, but atmospheres really don't need their thermal stats recomputed in this situation, do they?

EDIT: That code would be broken in multistar situations anyways, as it only uses one "Sun" vector. I recomend removing lines 283-291 from the present master of KopernicusStar, and testing. They are commented under the heading "// Get Thermal Stats." This should allow the stock game to calculate flux properly.

We will still only get atmospheric flux calculations from one star, but honestly, that's not a huge deal right now as long as it's the closest one, and it should be in most instances (isn't that what stock does?). I hope I'm right anyways, Kopernicus is pretty foreign to me.

I'll look into it in detail tomorrow. I don't recommend acting on this until I have had time to test.

EDIT EDIT: At the end of a long night, I see no reason for that entire 07230f3 commit. I recommend removing it's changes, pending testing of course.

[R-T-B]

Don't revert that afterall, it's needed for multistar support.

I think I fixed it, but I need a reliable way to test this issue. Is there a test scenario I can setup easily for it?

[R-T-B]

Tested and fixed in PR #7

[R-T-B]

Or so I thought, it's still not right. It appears to be applying temps to the vessel component and not the general atmosphere, as confirmed by AeroGUI. Still working on this.

[Keith-OHara]

Shucks. Must be something else in PanelsFix2.
The timeline could be confusing: this issue was actually opened a day before the merge (PanelsFix2 6022e3b) that brought the into this branch.
Release 1.9.1-1 has the air temperature cold at the poles, warm at the equator, and 1.9.1-2 has uniform air temperature,

[R-T-B]

Unfortunately my proposed PR is not good enough. EC is not functioning correctly at all.

I'm working on fixing that. For now, do not merge.

[R-T-B]

This was fixed with last merge and will be included in next release, so closing.