Rock

[Alexn007]

wow! so convincing, I love your work!

[René]

Thank you.

[René]

The 'problem' continues to arise, even with Perlin billows. The warp seems to have a preference for the sides and top, or the sides and bottom, depending on whether the displacement is positive or negative. Although the result is quite attractive. Unclamped fractal gives a slightly better result.

[N-drju]

What an interesting study! And yes, you are right about the warp.

You may try to put desired nodes before the compute terrain node. At least this is what sometimes helped me to achieve effects on near-vertical surfaces. No need to use normal then. Things tend to, kind of, sort themselves out.

[Dune]

What if you warp by tiny fake stones? I don't know how big your rock is, but that may offset the straight edges somewhat.

[René]

Initially I used fake stones to warp voronoi because I thought they would deform in the same way everywhere because they are more or less the same size. Not so, the straight lines remain. Nor is it the voronoi shader with its tight angles; the same happens with Perlin.

[René]

What an interesting study! And yes, you are right about the warp.

You may try to put desired nodes before the compute terrain node. At least this is what sometimes helped me to achieve effects on near-vertical surfaces. No need to use normal then. Things tend to, kind of, sort themselves out.

I can try that although I'll probably have to adjust the whole shader tree.

Maybe instead of warping, the texture can be masked with a very light power fractal: white 1 and black 0.95 or something like that.

[N-drju]

You could indeed try to add some FSS to the surface (not necessarily tiny, could be bigger) but as an additional surface, not a warper, and see how that works.

I understand you have already tried the warp input with vector disp. and a warper?

[WAS]

When you warp with a general displacement, you are warping on X Y and Z with the same ferocity. Warping Voronoi or even other shaders on Y too extremely will cause that weird warping effect, as you're throwing the Y point up and down, and it may not align with the X and Z points.

No variationns I think really help with the 3D aspect of warping, additionally, other noise flavors besides perlin have pinches/seams in their noise which can effect warping; again, sending Y (or X/Z) out of alignment in relation to other directions. Even with warping on just X and Z (because out amplitude of this noise map as displacement will be our Y basically) you can create "lines" because you warp too far one direction, so the noise stretches.

Additionally, parts of the displacement not warping are likely cause by little/no displacement in the area of the map displacing that specific part of your voronoi or other displacement. You could probably solve this by using the noise as a mask for those little/no disp areas and multiplying the noise map/displacement. Or just using that mask to add another PF with disp in that area.

But in that topic you linked, you can get something similar in TG for sure, see example.

[WAS]

Here's an example of the same method of warping on voronoi. There are still "straights", but that's just how the voronoi shader is, it creates triangle-like filler, but they are not unaffected, and more natural. I also think there is an issue with certain noise and the type of "faces" it produces, where things may get less effected. With the classic erosion shader for example, lateral disp tears it up really bad for some reason, and I think it's the "quality" of the terrain it produces. Doesn't matter if it's low or high res, lateral work is really hard on it and usually requires extreme patch sizes in compute terrain to subtle out issues.