Why? This makes it unnecessarily difficult to set up blending effects.
Importance is exposed in both the built-in and universal RP and serves a valuable purpose there. It’s not like the HDRP has somehow revolutionized the way reflection probes work that would render importance useless, it is clearly used in the exact same way as the other RPs. The only difference is that the value is inaccessible, which forces you to create reflection probes in the exact order of the importance hierarchy you’re trying to create and name the probes in a way that reflects their inaccessible importance value so you don’t lose track, and also pray that those values are actually stable and not silently reshuffled by the engine at some point.
Here is an example of a very basic use case for importance:
This oval room is a bad fit for box and sphere projected reflections. Box projection can accommodate the uneven aspect ratio of the oval but not the curved shape, while sphere projection has the exact opposite problem.
Instead you would use three or more sphere projected reflections that blend together to approximate the shape of the oval, but that would require setting the importance of the central probe higher than the outer probes so the blending is even from the center. As you can see in the video that’s not what happens though, instead one of the outer probes has the highest hidden importance value which causes the blending to be uneven depending on which direction you move along the oval.
I have read all the pages related to reflection probes in the HDRP documentation to see if there was something I missed or if importance is hidden in the additional properties, but that is not the case. I can only imagine the importance was removed because somebody thought that the weight value serves the same purpose, which is wrong.
I can get somewhat even blending if I set it to the fraction corresponding the the amount of probes, in this case 0.34, but that will result in a significantly dimmer reflection because the influence volumes of the probes only intersect partially. The result is a comletely uneven reflection which is way too dim and then suddenly becomes much brighter at the narrow intersections of the influence volumes, which looks horrible and is not at all what you would get by blending all probes with a weight of 1 and manually configured importance.
