2
$\begingroup$

I have a human model. It has 1 mesh for the body and legs. 2 for the feet, 2 for the hands, 1 for the hair and 1 for the head. Should I make them into 1 model? I want to use this in a game.

Whenever I rig, sometimes, say, the hand clips through the arm, or is completely detatched from it (Imagine it like a hand that's been only been sawed off about 75%, with a little piece of flesh keeping it attacked to the body. Twisted example, I know)

Also, if I want to make a background environment, should I make it all one whole, or multiple parts, such as say, one mesh for the ground, and multiple meshes for the trees?

$\endgroup$
  • 2
    $\begingroup$ A user wanting to provide a good answer to the questions, will, in my opinion, require information you have not provided. For example, what purpose do you intend to use the human model for? The answer will be different for a human model being placed upon a park bench as an adjunct to a scene, than if the same model is the focal point for which the scene is providing support. One change I would suggest, though, is to move the legs from the mesh with the body, to one leg each with one of the feet. $\endgroup$ – brasshat Mar 16 '16 at 3:40
  • $\begingroup$ Probably the most fundamental rule of thumb in modeling, at least in my experience, is that if one thing is to be moved relative to another thing, the two need to be at least separate meshes, and usually separate objects. Yes, you can have one mesh object for the ground, but each tree should be it's own mesh object. . $\endgroup$ – brasshat Mar 16 '16 at 3:42
  • $\begingroup$ Despite my comments, I consider this question to be out of scope because it is too broad, and is largely a matter of opinion. It would be much closer to being in scope if you were to provide a screenshot of the proposed background, a screenshot of the human model, and a clearer example of exactly what you are hoping to achieve. It might still be out of scope as too opinion based, though, as you can show the same material to a dozen blender heads, and get three and a half dozen opinions on the best way to achieve what you want. $\endgroup$ – brasshat Mar 16 '16 at 3:49
1
$\begingroup$

I would say model all separate objects as separate meshes and group all objects that form a single continuous surface.

In the case of the human model it should probably be a single mesh object, since it is a "continuous single skin surface", that way you won't get discontinuities, artifacts or seams while rendering.

Edit: I am not into character modeling so my view here is bound to be inaccurate, as suggested below, a clothed character can optimally be subdivided into the several visible patches of skin, like say head, hands, legs, etc.

As for the background scene divide it sensibly as much as possible. Take advantage of cloning and instances whenever possible for scene optimization. It can save a lot in file size, memory management while rendering and a lot of manual work updating different objects. Terrain can be one single big mesh, each individual tree and vegetation should be single objects, etc.

$\endgroup$
  • 1
    $\begingroup$ I concur with Duarte's suggestions, except for the second paragraph. If you are intending your human model to be unclothed, then it is a "continuous single skin surface"; however, if the human figure is clothed, then it could be broken into sub-objects, for example, a mesh above the waist, except for hands, the head, the mesh below the waist, including legs, and a mesh for the feet. $\endgroup$ – brasshat Mar 16 '16 at 3:49
  • $\begingroup$ True, I'm not much into character modeling, so I admit my view there is not a very informed one. Thanks for the feedback, updating answer $\endgroup$ – Duarte Farrajota Ramos Mar 16 '16 at 12:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.