Hello there,
I’ve been studying models and how they’re constructed lately. In this instance I downloaded a model from Polyhaven. I’ve selected the model in edit mode and separated it by it’s loose parts. Would it be safe to say that when this model was made, this was the approach they took to make it? (model each separate part by itself for the sake of simplicity?) Here’s what I mean:
Thank you for taking the time!
Here’s my take on this -
I would model everything as a separate mesh object where it would make sense to do so. The mounting bracket and screws could be 1 mesh object, but I would make them as 2 mesh objects, because each would have a different material on them.
The bracket that holds the camera on the mounting bracket could be 1 mesh object or 2, depends upon what you want to do with it. If you’re doing still images, then 1 mesh object. If you’re doing an animation, then 2 mesh objects, so the camera can rotate and pan up and down.
The front of the camera case, the camera itself and it’s backing box I’d do as separate mesh objects because each of them will have a different material on it. Also, the front of the case you may want to uv map to add image textures to for dirt/scratches. No sense in uv unwrapping more than you need to.
The hood of the camera (top of the case) would definitely be 1 mesh object because it has an image texture mapped to it. The main case, below the hood, again 1 mesh object because you may want to add an image texture to it.
I think it is easier to keep most stuff as a separate mesh object. When uv unwrapping, why deal with more stuff than the mesh you want to add images to?
Keeping things as separate mesh objects also makes it easier to hide parts you don’t need to see as you work on something. Just turn off the visibility in the outliner.
As a single mesh object, if I wanted to work on the hood of the camera and hide everything else, I’d have to select a vert from the hood. Do a ctrl-L to select linked verts, then a ctrl-I to select the inverse verts of my current selection. then H-key to hide them.
Just my opinion…
Randy
In the CAD area you model everthing.
For VFX you model every detail that could be seen in detail.
For the rest: what you can’t see… who the hack cares ?
So for example the top part of the camera housing might be not needed when it is covered by the extra sun/rain shield… the inner part between the “round camera window” and the actual camera might also be just a simple black plane with a hole and some image texture (depending on the detail you want; for example: is there a hero shot zoom onto this part; story telling-wise… there is someone watching ) ??
Man, thank you so so much, this makes much more sense now 
It does seem like dealing with just one mesh would be a complete nightmare when it comes to editing it and adding textures to it.
I really appreciate you taking the time and replying to my post!
Happy Sunday.
Perfect, not only did you make it clear but you went ahead and explained the different uses of modeling depending on the industry.
I truly appreciate you taking the time and replying!
Happy Sunday!
I also should have mentioned, after modelling everything as separate mesh objects, you then need to join them back together again via parenting. For example…
The camera mount and it’s mounting screws, I’d make the mounting screws be a child of the mount. For the camera itself, some part would need to be the parent of all the other parts that make up the camera. I’d choose the top cover to be the parent.
The pivot between the camera and the mount could be a deep discussion, if you plan on animating this thing. This advice is coming from someone who thinks animation.
For still shots, everything could be a child of the mount. Position the mount where needed and take the shot.
Parenting will cut down the clutter in the outliner as the child objects will be under the parent object and can be collapsed/opened as needed.
Randy
