Fix z fighting of HiPS levels

[10110111]

Description

When a tile is drawn in a planetary survey, and then its child of a higher level is drawn over, we can often observe the lower-level tile shining through the higher-level one, and the shape that shines through depends on the position of the planet, which resembles z fighting.

To combat this, we now draw only the deepest HiPS level requested for the current zoom level, substituting lower-resolution textures for the tiles whose native-resolution textures haven't been loaded yet.

This makes it much harder to support fading between levels, so fading support is removed.

Type of change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• This change requires a documentation update
• Housekeeping

How Has This Been Tested?

Test Configuration:

• Operating system: Ubuntu 20.04
• Graphics Card: Intel UHD Graphics 620

Checklist:

• My code follows the code style of this project.
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation (header file)
• I have updated the respective chapter in the Stellarium User Guide
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• Any dependent changes have been merged and published in downstream modules

[github-actions]

Great PR! Please pay attention to the following items before merging:

Files matching src/**/*.cpp:

• Are possibly unused includes removed?

This is an automatically generated QA checklist based on modified files.

[10110111]

Here's a screencast of how the issue looks (a bit overcompressed, but enough to get an idea): https://youtu.be/gKLiFmzM9k8.

This might be hard to reproduce with low-order HiPS, I reproduce it with an order-6 survey of the Moon that I'm preparing.

[alex-w]

Hmm... I see an empty place when HiPS for Moon is enabled.

P.S. Probably this is internet connection issue

[gzotti]

I also see no image, but the logfile screams:

[    40.645][DBG ] Load allsky QUrl("https://data.stellarium.org/surveys/moon-normal/Norder0/Allsky.webp?v=2458533")
[    40.666][DBG ] Load allsky QUrl("https://data.stellarium.org/surveys/moon/Norder0/Allsky.webp?v=2458519")
[    40.666][DBG ] got allsky
[    40.686][DBG ] got allsky
[    40.686][DBG ] Load allsky QUrl("https://data.stellarium.org/surveys/moon-normal/Norder0/Allsky.webp?v=2458533")
[    40.707][DBG ] Load allsky QUrl("https://data.stellarium.org/surveys/moon/Norder0/Allsky.webp?v=2458519")
[    40.707][DBG ] got allsky
[    40.727][DBG ] got allsky
[    40.727][DBG ] Load allsky QUrl("https://data.stellarium.org/surveys/moon-normal/Norder0/Allsky.webp?v=2458533")
[    40.747][DBG ] Load allsky QUrl("https://data.stellarium.org/surveys/moon/Norder0/Allsky.webp?v=2458519")
[    40.748][DBG ] got allsky
[    40.768][DBG ] got allsky

[10110111]

Hmm, I didn't test on a survey with Allsky used.

[10110111]

Hmm... I see an empty place when HiPS for Moon is enabled.

Well, I get the normal Moon with the default Stellarium's survey.

I also see no image, but the logfile screams

Hmm, I wonder why... What I get is this and nothing more:

[    34.293][DBG ] Load allsky QUrl("https://data.stellarium.org/surveys/moon/Norder0/Allsky.webp?v=2458519")
[    34.323][DBG ] Load allsky QUrl("https://data.stellarium.org/surveys/moon-normal/Norder0/Allsky.webp?v=2458533")
[    34.404][DBG ] got allsky
[    34.419][DBG ] got allsky
[    34.422][WARN] Normal map survey's hips_order=2 is less than that of the color survey: 3. Will reduce the total order accordingly.

Do you maybe have some other surveys enabled?

[10110111]

I also see no image, but the logfile screams

Do you also get this in master?

[gzotti]

I also see no image, but the logfile screams

Do you also get this in master?

Uh, indeed! As soon as Survey is selected, HiPS drawing activated and Moon is in view. There are no other (deepsky) HiPS active.

[10110111]

Uh, indeed! As soon as Survey is selected, HiPS drawing activated and Moon is in view. There are no other (deepsky) HiPS active.

So HiPS is broken for you in master? The symptoms you show may indicate that Allsky is downloaded, but can't be read as an image, i.e. QImage::fromData(data) returns a null image. What if you try downloading the URL logged via a browser — does this give a viewable image?

[10110111]

Please try again, the log should calm down and report an error once (per file) in case of a failure to decode the image.

[gzotti]

OK, that seems to work. But the next strange obervation: select a crater near the rim, then zoom in. (FoV in range of 1-2 arcmins - just where the HiPS looks still inviting...). Suddenly the Moon vanishes. This does not happen in 25.1, and not with the default Moon.

[10110111]

I reproduce vanishing sectors that have no corner inside the screen in both this branch and the older versions. This is a years old problem with Stellarium's HiPS. I wonder why you see a different behavior between versions. Just to make sure, what crater did you select? And did you set the time to the same values when testing each version?

[gzotti]

Riccioli, 2 25-7-12 12:56 CEDT
This PR: Vanishes when FoV<43".
25.1: Ah -- Poof! Same issue, indeed. I just scrolled around between east and west rims before, so apparently it works sometimes.

[10110111]

So, is this PR OK to go in then?

[alex-w]

Thanks!

[gzotti]

Yes, should be OK from my side.