Very interesting… So according to what you said, I can claim that the RVT does work, maybe, between two regular meshes, if one of them is a flat surface ! It’s a bad idea if you try to blend two meshes like these in the RONOWE’s picture (I’m talking about the rock and the sphere) but, let’s say we are trying to blend this (let’s say it’s a rock on top of, I don’t know, some kind of platform) :
](filedata/fetch?id=1837909&d=1606835282)
The surface that blends is flat (or let’s say almost). But in the same time, the assets are not just simple flat surfaces. I mean they are not 2D planes. So the question I’m asking is : What the RVT captures or sees exactly ? The entire texture or just the area where the textures of each assets cross together ? When I see how it works in videos, I guess it captures only where it cross but, I don’t know, I have a doubt… Well, in other words :
This way of blending with RVT (landscape with mesh) is OK, for sure :
](filedata/fetch?id=1837910&d=1606834403)
This way is bad, for sure too (let’s say it’s 2 rocks or kind of that) : ](filedata/fetch?id=1837911&d=1606836250)
But what about my first picture ? Is the RVT system will capture some duplicate Z points (I’m talking about the height coordinates) as the blue mesh has, in short, a face above and a face below ? And so this is certainly problematic… Or will it capture just the area where the textures of each assets cross together ?
If the second assumption is right, so it’s all good. If it’s the first that is, I guess we could easily fix this issue by cutting the blue mesh to make it look more like a real 2D plane, like so (like if it has a hole) :
Maybe it also needs to cut the blue mesh a little heighter because it is still a bit curved “toward the inside” so that’s why I colored the base in light blue (it’s a very simplified and schematic draw ; find the parts of the mesh that “curve toward the inside” would be more complex in reality, on a 3D asset, but I think you get what I mean…)
What do you think about this ?