Say for example that someone is modeling a sword asset. A sword has many pieces -- the blade, guard, hilt, pommel, etc. -- so should this be modeled as a single mesh (or as multiple meshes which then get boolean'ed, joined, etc. into a single mesh), or should each piece be modeled separately and then get constructed and glued together later in a game engine environment?

  • 1
    \$\begingroup\$ It all depends on your target game/engine requirements \$\endgroup\$
    – Kromster
    Sep 29, 2021 at 19:01
  • \$\begingroup\$ @Kromster How do you mean? Could you elaborate on that? Do I need to provide better examples? \$\endgroup\$
    – NJJ_002
    Sep 29, 2021 at 19:09
  • \$\begingroup\$ If it is intended to be used as one piece, then it should be modelled as one piece. If it needs to be used as multiple pieces, then you will have to model it as such. Simple. \$\endgroup\$
    – Engineer
    Sep 30, 2021 at 12:30

2 Answers 2


Some games/engines might be happy with a single mesh, a generic material and a texture, so that they could put it into the atlas with e.g. other weapons.

Some games/engines might need a standalone model with custom material and textures and masks denoting material properties (diffuse, roughness, specular, metallic, reflectance, etc)

Some games/engines might benefit from or need a multi-mesh and multi-LOD model with custom materials and textures and masks denoting material properties (diffuse, roughness, specular, metallic, reflectance, etc) for each separate mesh.

It all depends on your target game/engine requirements.

As an approximate guideline - the more highpoly/detailed the model is - the more effort might need to go into structuring and organizing it (masks, materials, textures, layers, meshes, LODs). If anything, you can include several levels of detalization in the pack.


It depends on how you plan to use those meshes.

On your example, if the sword and the hilt and all the other components are meant to be together, and are never meant to be separate, then there's pretty much no reason to have them as separate meshes during loading. It just adds complexity, and could potentially make drawing that asset slower (if not optimised right).

Having said that, both ways would work. A mesh would mostly be divided into separate meshes if it needs to be this way, either because it can be separated at some point during the game (a sword that breaks into pieces?) or if parts of a mesh are reused in other areas (some hilts can be found without a blade in the game).


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .