Until Arnold updates their shader to work correctly with tangent space normal maps and hard edges that's really all you can do, it's very frustrating. To that last point have a look at Brent LeBlanc’s guide to texturing with substance painter and then bringing the model and shaders back to render with Arnold in Maya. To fix gradations add extra edge loops or bevels to the low poly. You need to have a way to export models out, UV them, texture them, and then possibly bring them back into Maya for rendering with Arnold. Triangulate low poly mesh, soften all edges, bake in Painter, preview in Arnold render with exact same triangulated mesh and ignore what you see in Painter. Ironically how you fix gradations is by hardening the low poly normals and re-baking, so you really only have one option here which is not a good one. The reason you don't see gradations in Painter is because they use Mikk tangents and their viewport shader is better than the Arnold aiStandard shader. Having any hard edges creates seams in the final render, but softening all the edges creates gradations like you have in your render.
Now for the bad news, as far as I can tell Arnold can't correctly render a tangent space normal map in Maya unless you soften all normals.
In Maya you must use that exact same mesh, the triangulated one to apply the normal map to so the triangle direction doesn't flip.
In substance make sure your mesh is triangulated before you bake the normal map. We'll show you how to get started with SimReady USD assets, and present the tools and techniques required to develop and test these assets.The issue you're facing is because Substance Painter uses the Mikk tangent space and Maya viewport and Arnold use something else. SimReady assets are more than just 3D objects-they encompass accurate physical properties, behavior, and connected data streams built on Universal Scene Description (USD). NVIDIA is introducing a new class of 3D assets called "SimReady" assets-the building blocks of virtual worlds.
Whether building digital twins or virtual worlds for training and testing autonomous vehicles or robots, 3D assets require many more technical properties, requiring a need to develop and adopt novel processes, techniques, and tools. Creating virtual worlds is hard, and today's existing universe of 3D assets is inadequate, representing just the visual representation of an object. The next wave of industries and AI requires us to build physically accurate virtual worlds indistinguishable from reality.
0 kommentar(er)
