Texturing Techniques/Methods/Useage Question

Guy

Hey Polycount, I've been doing 3D modelling on and off for quite a few years now, I also like to rip models from games and the sort.

My question is, while ripping models, I've run into ones that use textures like these:

How/What are they used for?
Are they sculpted and baked like a normal map?
Are they something used generally for real-time models, or for cutscenes?

Normally I don't do a lot of higher poly modelling, but I'd like to and get myself up to date with newer methods.

I'd genuinely appreciate any light you guys could shed on this, thank you for your time.

musashidan

Could be a packed texture. Open it in PS(or similar) and look at each of the R, G, and B channels individually. It could be Gloss/SSS/Reflection maps for a skin shader or a mask for hair/eyebrows. A packed texture is just a method of saving draw calls/memory in-engine by packing multiple greyscale maps into unused channels of a texture.

Guy

There were 3 other textures along with this one:
- Diffuse (base colors/details)
- Specular Map (B/W image)
- Normal Map

I took a look at the channels, R and G have two different Greyscale textures, and Blue is Black, so it seems it's as you say.

I'm curious how this would work for masking, why not just utilize the Diffuse texture and the alpha channel to define what's seen and what isn't?

NoRank

Guy said:

There were 3 other textures along with this one:
- Diffuse (base colors/details)
- Specular Map (B/W image)
- Normal Map

I took a look at the channels, R and G have two different Greyscale textures, and Blue is Black, so it seems it's as you say.

I'm curious how this would work for masking, why not just utilize the Diffuse texture and the alpha channel to define what's seen and what isn't?

Well, because this is not the intention on making this packed textures probably. Packed textures are just what musashidan said, they are a "mix" of different maps that are packed in different channels and used on a specific shader built to identificate these specific channels and for what they are for. 

They are not just like opacity maps (even though you can pack one opacity map in them, it's just a matter of how your shader was built to identify these channels).

Idk if you got it wrong or I did interpret your question wrong lol. I'm pretty sure that on the wiki there is a good info about that, with some images showcasing how it actually works.

musashidan

Guy said:

I'm curious how this would work for masking, why not just utilize the Diffuse texture and the alpha channel to define what's seen and what isn't?

The mask, in this case, could be an SSS mask to dictate where the shader applies a sub-surface effect. Or a facial hair mask/etc. If the diffuse texture had an alpha channel(which it wouldn't for this purpose as it would destroy mipping) then it wouldn't be masking inside the bounds of the texture but, around it. If that makes sense. That is why a custom mask would be used separate from the diffuse texture itself.