So I hit the end of 3d motive's tiling texture tutorial, only to discover that they didnt cover making the material in UDK. I copied some of what they did in the tutorial files, but I don't know what some of the stuff is actually doing.



I don't really understand what the vertext blend mask is actually doing (wtf is a divide???) nor do I get what the bump offset is doing. I know that it's possible to illustrate depth using the bump offset, but I haven't seen it used like this before.

Also, is there a better way to improve the normal map than what I have here? Can I use heightmap info rather than a constant3 to get more depth?

Im not good at udk materials but the vertex mask seems to make a mask out of the red channel with what i think is math nodes like divide and such to make a mask for the black to overlay over the specular. And bump offset is using the red channel of the diffuse as a heightmap i think. Someone can probably tell you for sure but this is what i think.

the bump offset node can only intake the value from one channel. The Bump offset itself is using a heightmap from the red channel and being used to offset all the maps applied to the shader(to unify the effect). You can increase the depth of the parallax effect if you click on the bump offset node and change the height ratio

http://udn.epicgames.com/Three/Mater...tml#BumpOffset

the vertex blend mask in your set up is for vert painting different specular color maps.
http://www.3dmotive.com/training/udk...aint-with-udk/

Ok I kinda get the bump offset a little better now. Does it make a difference which texture channel it uses? or is it good practice to use the red vs the blue or green?

And basically I don't need the Vertex Blend Mask since I'm not doing any vertex painting for this particular material right?

It does not matter what channel you use you can use the alpha channel, what it need is a black and white image which all channels are. (It "want" a heightmap so the color channels of a texture may not always work in the best way)

Then would it be more efficient to copy the heightmap into the Alpha Channel of my Diffuse or Normal? or would that just use more memory?

Well it would use a little more memory if you use an extra channel (alpha) but if it would make better result than i would recommend it.

Thanks I'll give it a shot