Adjust request priority for close globe tiles

[lilleyse]

For #6243
Fixes #5954

If the camera is inside a globe tile's bounding rectangle, that tile will be requested with highest priority. This prevents 3D Tiles requests from possibly starving imagery/terrain requests. This is especially important for the root globe tile and other large tiles where distance is a bad metric and they should really be loaded right away.

I'm not sure if I've truly seen 3D Tiles starve imagery/terrain (#5954 is more of an edge case), but this would prevent that from happening.

[cesium-concierge]

@lilleyse, thanks for the pull request! Maintainers, we have a signed CLA from @lilleyse, so you can review this at any time.

⚠️ I noticed that CHANGES.md has not been updated. If this change updates the public API in any way, fixes a bug, or makes any non-trivial update, please add a bullet point to CHANGES.md and comment on this pull request so we know it was updated. For more info, see the Pull Request Guidelines.

---

I am a bot who helps you make Cesium awesome! Contributions to my configuration are welcome.

🌍 🌎 🌏

[pjcozzi]

Update CHANGES.md

Do you think the heurtisc should also take into account dot(view-direction, tile surface normal) so that the more the viewer is looking at the tile, the more important it is?

[lilleyse]

Do you think the heurtisc should also take into account dot(view-direction, tile surface normal) so that the more the viewer is looking at the tile, the more important it is?

This could maybe be good in some situations. It would have to be tuned pretty carefully since it might de-prioritize tiles that are not straight on but still take up a lot of pixel space.

I'll add this idea to #6243.

[lilleyse]

Update CHANGES.md

I think I'm going to have all the 3D Tiles PR's go into request-performance and write a more general entry in CHANGES.md.

[puckey]

This seems to only prioritise the tile in the very center of the screen – perhaps it would be an idea to prioritise all tiles visible to the camera? Or even discard those in the queue that aren't visible?

[pjcozzi]

@kring will this be replaced by your latest work?

[kring]

@kring will this be replaced by your latest work?

Yes I'd say so. The modified function has been removed entirely in my branch.

[pjcozzi]

Gotcha.

@lilleyse please coordinate if there is anything worth saving/discussing with @kring and then delete this branch when ready. Thanks.

[lilleyse]

Fine to delete this branch.