Hi everyone.... This is something I just do not no how to do and it would help so much if someone could point the way. Nearly all tracks have these texture files, where all the images on one 3d object are combined into a single image and you may have 10, 20 or more images for the object in one dds file. I would so greatly appreciate if someone could point me in the right direction. Here is an example below....thanks in advance. Mike
If you use 3DSMax, search for autodesk uvw mapping tutorial videos. It's just about how (into which position) you map the objects polys to the texture.
@Mike Cantwell I'll admit I just shared something I thought might be useful. I can see what he's doing (in the second half of the video) doesn't exactly apply to your question, but I suspect it's along similar lines. And I would say you'd start by making the image you're talking about, and then when you.... er… apply the mapping to the object, you reference each sub-texture as you do it. So I think the process is the reverse of what you're thinking. But I could absolutely be wrong
most broadly, i think you're asking about texture atlasing. https://en.wikipedia.org/wiki/Texture_atlas Atlasing enables batching, which in the below example will allow 4 different meshes to become 1 mesh, which will reduce draw calls, which is a good thing there might be a few ways to go about it. Here is one. Unrwapping your single object into a 0-1 texture space, as usual. Then a few of those from different objects can be stuck into one bigger texture. Say if that now has the textures for four different objects, and each of those was a square texture, you now have one bigger texture, with one quarter of the space for each object. If you're in 3dsMax, you can use uv xform to adjust each object's coordinates. They'd need some combination of offset and scale. Either 2x scale, or 0.5 scale (i forget which way it goes), and some combination of 0 and 0.5 offset depending on which quarter of the big texture each object related to.
I usually map the objects 0-1 texture space, then I choose objects that can have same shared texture, and perform similar operation to creating AO: first flattening the mapping on a different map channel, then manually improving the result if needed, to make least amount of empty space and to make every face/poly covering big enough area (more important/visible parts to cover bigger area). Then I render to texture my map channel 1 to this new mapped channel. Obviously you need to have just 1 of each objects to avoid duplicate items occupying precious texture space. Also you can render AO onto the same channel, in case if your objects need some AO/shading, you can then mix it with albedo texture, as they will occupy the same space. Then, repeat it for normal and spec map - assign the map to the original channel and render to texture to the newly mapped channel. As soon as all your images are rendered, go to channel info and cut the mapping of new channel and paste to ch1. But.. all of this is only if the objects are complicated. In other cases I just combine textures in Photoshop, minding the dividing proportions, and then use UVW xForm in Max.
