Well Ive been messing around with nodes to see if i could come up with a convincing material for skin, that worked on bge. (since no sss is allowed )
I’ve developed 2 methods for such:
1st
This one used 3 distinct materials that mappped R G and B diffusion throughout the object, and used ramps to limit such dispersion. Then I combined RGB, added a second material only for spec, mixed it and voilá
2nd
Same as 1st, but this time i made the materials oren-nayar and made the Hard paramater different for each color (rather than ramps), AAnnnd i use the Z-depth from teh camera to map translucency. Though this method yelds results no prettier than the 1st, it’s physically more accurate.
BTW, can someone lend me a rigged female model thx haha
PS, the 2n method is attached, for some reason i can’t seem to attach the 1st
1st, the Material is all dark turquoise with dark-reddish Shade while all the Lights are just plain white. Not very »skinny«.
2nd, in the Game Engine your Transparency Thing leads to a screwed Z-Order, or maybe just on my GraphicsCard and not on Yours, but that doesn’t work either Way: Skin is not transparent in the Way the BGE handles Transparency. Skin is translucent, but not so much transparent.
I did not use transparency haha, rather the translucency parameter. Also for that weird color, try movinfg one rgb input color wheel, so the material updates. That happens a lot on the GLSL view haha.
Anyways to be honest I was kinda lazy yesterday and didn’t correctly make the material (z depth was kinda overlooked), today i made an effort and actually made it look nice, but sadly i can’t upload it properly, so i’m gonna describe it:
i take 2 rgb inputs, separate them and assign the values to reflection and color on 3 different materials (RGB) that only have diffuse. Each material has a color ramp in the diff shader, this ramp goes from white (maximum lvl) to black. This restricts the dispersion of rgb, and i’ve made the B ramp the shortest, and the R ramp the longest.
All diff shaders on the rgb materials are oren-nayar, with different hard parameters (R>G>B).
Then i take the color output and combine the RGB, set another material for the specularity (blinn spec shader), mix the color output of both (actually add, haha) and link it to the output material.
Now on the translucency part i gave it another go and this time i used a series of math operators to create a gaussian function that has a maximum value of 1.000 for 0.003 of z depth, and minimum of 0 for 0.4 z depth. Grabbed another RGB input, separated it, multiplied each channel eith the resulting function and assign to the tanslucency of each material. This part produces some errors on the material preview but on glsl it seems to work fine (though i can’t tell if it works or not)
)