hi everyone.
Can someone explain the difference between texturing for an animation and for a game
is there any other difference then the texture quality?
You have pretty much answered your question. The primary difference between texturing for an animation and texturing for a game is basically level of detail. However, if you make a GLSL game, the detail is almost the same as the detail in an animation.
Blender has three rendering options for the GE (in order of detail, bottom to top): texture face, multitexture, and GLSL.
Pretty much everyone can run games in texture face. While it is the fastest mode, it also has the least quality. Uses UV textures for detail.
Multitexture is the most commonly used of the three. Most people can run games made in this mode. Also uses UV textures for detail.
GLSL - If you want to make current-gen games, use this mode. Objects are textured via Blender’s on-board materials, so you can add specularity (shininess), displacement, etc. This mode is almost the same as modelling and texturing for an animated movie. However, there are several downsides: This is the most graphically demanding mode and consequently requires a much faster computer. Also, some computers do not have the capability on their graphics card at all to run GLSL. If you make a GLSL game, you are limiting your user-base.
It is important to think about the amount and kinds of users you are reaching out to. If you have time, for instance, you could make two versions of a game, a GLSL version and a Multitexture version. Or you could just stick with Multitexture, which is generally supported.
I hope this answers your question and provides information for you.
Blender_p_user’s response is correct, but I’ll add a little more. Generally game engines have a limited form of lighting, unlike animation and raytraced images, a lot of the light in a game engine has to be faked (SSAO, Lightmapping, etc.) even in current-gen textures for games, there is a certain amount of lighting information painted into the diffuse and specular maps, usually artists pick a general light source for a model and use it as a guide line. When I’m painting a texture I tend to use an AO bake and paint in details as if it was lit from a top-down spot light.
This explains it, theres two articles about it here: http://www.poopinmymouth.com/tutorial/tex_theory.htm
Thanks, Oki , for adding the note about lighting, which I had forgotten.
Yes, DarkWaysLight, lighting, generally, is faked as real-time lighting is very resource hungry. Only the most graphics-intensive games, such as Crysis (which still can bring modern computers to their knees, metaphorically speaking), use real-time lighting. Faking lights will definitely give your game a speed boost.
If you’re making game models, definately look into high poly to low poly baking of materials, textures, normals, and even lighting. especially normals baking. It’s a way to make a really low poly model look like a really high poly model.