Hello there,
I am a beginner in this entire 3d world so the question is probably stupid but I couldn't find an answer online that felt right so here It goes.
For simplicity, let's pretend I am trying to make a wall that has several assets against it, for example, a closet and maybe a whiteboard.
In such cases, I do each asset separately and then position them, meaning I have parts of the wall that are not being seen and the same goes for the closet and the whiteboard.
Should I even bother wasting time with those parts of the "scene"? I was thinking of deleting them all along.
From podcasts and blogs, I got the idea that usually, games do something like this: (Correct me if I am wrong please) They have libraries of assets where they then create the composition of the scene, and somewhat in the pipeline they delete all the geometry that has no purpose.
My question is if I should do this? Is this a good practice that game scenes require?
I only ask this because it seems to me like a waste of UV space, time, memory, and resources. If I just delete that geometry I don't have to UV it and I gain all that UV space back.
I know that in this make-believe scene I created to make my question it seems like an unnecessary effort but I am currently learning Maya a bit more in-depth and I am doing so by trying to make a complex hard surface asset where I use the exact same process. I separate it in many pieces that then I fit together to make the final asset. The problem is that at least 50% of my UV space is hidden faces and for what I hear game assets don't use UDIMs so I am looking at a lot of wasted UV space which results in poor quality texture or me having to crank up the resolution.
So I just want to know what is the correct thing to do in these cases.
Am I seeing something or is my inexperience just talking louder :')?
Replies
But i fear in a real production you have to move them several times and you would be save not deleting them. In case of the Wall.
For other assets everything you dont see delete it.
Since these assets are all the same and can be placed at any angle you model all of it - nothing is deleted and occlusion / culling at runtime takes care of not rendering bits you don't need.
For unique or very specifically placed assets we don't make geometry you can't see in the first place.