Something you said in the last paragraph seems to have jarred some realization in my brain. Just to be sure, when you refer to a texel, you’re referring to a pixel in a texture bitmap, correct?
That would explain some of the other posts I’ve read. Some people refer to having a single terrain texture some 2 or 4 times the width and depth of their height map, meaning 4 or 16 texels map to a single square in the height map and those texels get “spread out” over the rendered quad (or 2 triangles). The texture map used is built using Java2D techniques and whatever algorthm I want as the scene is setup. It’s static during rendering.
Does that make sense? That’s my distillation of a few posts I’ve read that didn’t seem to spell the details out. It makes more sense now.
On to the detail map… I’m not clear how that works in practice. I understand the theory, but how does it get implemented? It sounds like a texture to each triangle problem again. Is this a form of “multi-texturing”? And when you say the detail map “modulates” the underlying texture, are you taliking about the rendering properties like BLEND and MODULATE? Can a detail map be used to modulate a simple color shaded triangle (i.e., a color assigned to each vertex and the engine interpolates the colors accross the face of the shape)?
I once took a look at FG, but I wasn’t looking for texturing techniques at the time. I’ll have another look.
-Tab