I know but using static_cast<bNodeSocket *>(BLI_findlink()) seems to be most common for the math node for some reason. See:
https://projects.blender.org/blender/blender/src/commit/0edd60b4c77f58…
I'll take a look at it after merging the Noise and Musgrave Textures. Is this issue also present with Voronoi?
Right, these "artifacts" were also mentioned by some artists in https://blenderartists.org/t/thanks-to-your-help-fractal-voronoi-noise-was-added-into-blender/1448508/26?u=hoshinova Some Textures…
@JacquesLucke It's also very uncommon to have such small Lacunarity values in the first place.
Since 4.0 and therefore Fractal Voronoi hasn't been officially released yet I want to fix it…
If you can't think of anything immediately it's also fine to leave it for now. I'll think about an appropriate function in the meantime.
There's almost no difference. It only occurs when the Lacunarity goes from a very small value to exactly 0.0.
If you take a look at the test .blend files I've provided you'd also find that…
I support this! Though I'd just call it "Grid Texture node" instead of "Hexagon Grid Texture node" to make it compatible for when CharlieJolly or I add square or triangular grids in the future.
Honestly, I'm also not entirely convinced by this solution mainly because it loses all the detail in the highs and lows. I thought of a solution which involves using a piecewise function which…
Test files for checking the implementations if necessary: Fractal Voronoi: Fractal_Voronoi_Test_File.blend
Voronoi: [Voronoi_Test_File.ble…
A .blend file for testing the whether all shader node implementations are equivalent is here:
