Hello. I’m using Blender for about a year. Now I’m working on a hunting rifle model from an old CG Masters tutorial. My high-poly mesh is finished and now I need to create a low-poly version of it, so I can bake normals in Substance Painter. I’ve heard about some LOD techniques which can lower the geometry if the camera is far from this mesh, so the object looks good, but renders faster. Of course, it doesn’t work in the opposite direction (making your mesh more high-poly), so if I’ll need a close-up shot of the low-poly model edges will be noticeable, like in this example
Wouldn’t it work to use the low-poly model, and subdivide it as needed on close-ups to smooth it out? Sure, you might get undesirably large displacements of the surface is some cases, but for the mesh in the image it shouldn’t be much of an issue. And setting crease also gives additional control to deal with sharp angles.
My thoughts:
If you want to use it as a first person weapon mainly spare yourself the pain of creating LODS (most unpleasant aspect of gameart imo).
Second:
Goal of low poly is to capture curvature. Your mesh on the the left has unnessecary loops for game mesh but captures curvature badly. So it is not good at all for game usage. Understand and respect principles for game art creation, but dont be too stingy with polygons. It is 2019 after all.
Another thing: You would most likley get a better result for game use from blasting the hipoly through ZRemesher than continue with your current approach.
