One thing has always bothered me and I want to get others thoughts on it. In Edit Mode, I’m always hesitant to add new pieces of separate geometry, instead always trying to extrude from my main “object”. Is there a rule or guideline here? I feel like this hesitation is really hurting my ability to model complex objects, and I can’t seem to reason why I feel like adding separate geometry feels so wrong to me.
For example, on a gun I was modeling, I had a difficult time modeling the “details” like the safety button and pistol grip, because I tried to do it from my main geometry. Should these pieces be separate? Does it matter?
Is this the correct feeling to have, or should I just forget this concept and do whatever I need to so it works?
Yeah that is the basic rule of thumb. And further it gives you the needed control and creative choices that you would otherwise not have.
Just don’t over do it in the sense that you have lots of intersecting parts and hidden polygons. I have seen modelers do this and it can get to be a bad nasty habit. Not only is it a mess and hard to edit, it wastes texture space when it comes time to UV map.
So, just take the other rule of thumb into account which is only model what you will see.
As for final details it will depend on your workflow. If you are going to use this for a game or if high polygon count matters to you, you might want to create a high poly version with separate pieces, say for the seams of panels and parts that are tacked on for example, and a low poly version that is basically one piece that gets maps baked onto it from the high poly version for adding back the details with normal and AO maps.