Making the Base Mesh in Maya
First of all, I started making the base mesh, this is essentially the low poly model and will most likely be the mesh I will use when I export it into UE4 or Marmoset (using photoshop for the textures and maps) to render it. The Base Mesh shouldn't have any fine details like scratches for example, these will be created in ZBrush on the high poly and will be baked on which I will explain later. The Base Mesh is kind of a 3D silhouette of what you're making, once the main shape is complete its easier to apply extra details in other stages of development.
Exporting the Base Mesh Into Zbrush to create the High Poly
Once the Base Mesh is complete, I exported it into ZBrush so I can add all the details I want to, I also used this take a big chunk out of the base mesh (for the damaged pillar) which you can do, it just means you'll need to cut that part into your base mesh. Once you're in ZBrush you can paint details on with a range of different brushes to get the effect you would like, scratches, scuff marks etc. We create a high poly so we can create all this detail so that we can apply it to the low poly model through baking. Like I mentioned briefly before, if you change the mesh to the point where the silhouette is different to the shape of the Base Mesh, you will need to decimate the high poly model so its polycount is reduced so you can then transfer it to maya and cut in that detail onto the Base Mesh.
UV Mapping the Low Poly
Now the Low Poly is done, it now needs to be UV Mapped so we can apply more detailed textures (than Blinn and Lambert) that fit the model better. You unwrap the Low Poly through the UV editor and create different UV Islands of your model. Essentially its like unwrapping a cube into its net form, but with more complicated shapes. Once you've unwrapped, stitched and projected etc and the UV is complete, you are now ready to bake the High Poly details onto the Low Poly and also texture it in UE4 or Marmoset (using photoshop).
Baking
Baking is the process of putting all the high poly details and information onto the Low Poly through a bake. What is the benefit of doing this? why not just use the High Poly? The reason we do this is because if you are creating an asset for a game, and we used the high poly model, it would take too much for it to load because of the sheer number of poly and would reduce the render quality of the game as well as its performance. Through baking you are putting all that information onto a lower poly mesh meaning it has all the details of the High Poly but will load and render faster (as well as saving more memory) because of the lower poly count from the Low Poly. To start off you need to create a cage for your bake, the cage is a bigger version of your Low Poly and what it does is detect any information between the Low Poly and the Cage and then project it onto the Low Poly. I baked my Pillar and Time Machine using Xnormal. You upload the UV'd low poly model and the high poly model and wait for it to create your bake, the bake is then applied in Marmoset or UE4.
Types of Texture Maps
- Albedo: The base colour map for your model
- Normal: The map you get from your bake, just depth and no shading (from the high poly)
- Metallic: The map that changes how metallic your model looks (grayscale, 0-1)
- Roughness: The map that changes how much light your model reflects
- Ambient Occlusion: The map that contains the shading information
- Emissive: The map that can create glowing effects and lighting
No comments:
Post a Comment