I’ve begun learning to model and have a few, hopefully simple, questions regarding the modeling process/workflow as it pertains to Unity. Thank you in advance for any of these questions that I can get help on.
Should I be looking to model all of the meshes I hope to import into Unity as tri-sided polygons? From my understanding, when importing a model into Unity the mesh is automatically converted to tris, but is it better for me to do this within my modeling tool manually?
What unit of measurement for scale is best/most widely used, or does this change for each project/game?
If I’m making a static mesh, should this object be all one mesh, or can it be a single object made of multiple meshes. For example a car door with the handle being a separate mesh.
If 3, what should I do with the intersecting faces of the multiple meshes that cannot be seen, can I just delete them?
for me, I use a made-up unit, and set the grid snaps to be “as wide as the players box”… meaning, the smallest grid snap I use equals the narrowest passage the player can move through… so a person would be like 2 units tall and 1 unit wide, basically an “almost meter”
No need to usually, you can even import N-gons (more than 4 sided polygons) but which I would not reccomend. When you triangulate first then you know the exact outcome because the triangulation can be different. Here is one example of what errors can occur. While your 3D programm would make the right mesh Unity might import the left one. In such cases I use triangles.
But Quads are generally better as it is easier to edit the mesh file, you never know if you have to. Especially when you ship it to other people who don’t have access to the source file.
In general 1 unit = 1 m that seesm to be true for most Unity Assets. That is also how default physics (gravity) is set to or Audio doppler effects.
It can, you are free to define your workflow. All depends on your target hardware and other requirements like animation.
Yes, unless you run into issues with shadows. Sometimes you end up with situations where the backside doesn’t render shadow. There are different ways to solve this though.
No - that will cause headaches while modeling - especially if the model is over 5k polygons.
Best to keep with quads - especially if going to be exporting to a sculpting package.
If concerned about areas like McDev pointed at - test export, it takes like 3 seconds.
1unit = 1meter. Set up all 3D content creation tools to match this Unity default unit setting to save on headache medicine later on.
My rule of thumb for this is - if it will be or can be manipulated it will be a separate mesh, unless rigged with bones. Otherwise the only time I ever have separate meshes is when I want to simplify the shader/material. So instead of having a multi-material on one mesh, keep it as separate meshes which can have unique materials/shaders. However this can cause excessive overhead, more meshes=more textures/materials/shaders=more draw calls. This is also avoidable, but more complex in engine - imo.
Don’t worry about it until you encounter an issue. Unless dealing with transparent materials/shaders this shouldn’t become an issue. You might experience some unintended effects from certain mobile shaders, but these are rare - and are mostly related to sorting order (?) when working with transparent / semi-transparent shaders.