Unwrapping a wall

I am am trying to unwrap a wall mesh with butresses onto a concrete texture with staining at the top and bottom to look like old oily garage wall. If I smart unwrap and UV, it, it puts the areas of the wall all over the graphic. I need to unwrap in a strip that is as high as the graphic and that the sections fit in horizontely IYSWIM.
Thanks

Dicon

Depending on how your model is made, either add edge seams (Ctrl+E) and unwrap or use ‘project from view’. For hard edge objects it is usually best to change the uv unwrap method from the default ‘edge based’ to ‘conformal’
Without seeing what you are trying to unwrap and to what texture, can’t give you a definitive answer.

Ya smart unwrap can be difficult to work with, but it is fast no matter how complex the object is. So to use it, find a way to adapt. I am still learning and testing Blender myself!

An idea came to me so I tested it here. To make the problem a little more complex, more geometry was added to buttress on the wall. But not to worry smart unwrap can unwrap any thing! Next the wall was baked with AO. And then I simply texture painted wall texture over my AO map, fast and dirty.

http://i1135.photobucket.com/albums/m626/cabby24/Blender%20Pics%202011/wall_build.png

Thanks, it seem to be an awkward way of doing it, but will try that.
Dicon

I would consider using several textures here: one for the concrete; the others for the oily highlights, graffiti and what-not. With UV-maps as needed for each one.

The base concrete might work best as a procedural texture because of the topology of the piece. (If you pressed a tape-measure against the wall at the bottom, it would be quite a bit longer than a tape-measure near the top or in the middle.) You can easily wind up with disjointed textures at those boundaries … and your eye would instantly notice the presence of those “boundaries” because they don’t belong there.

You might want to use, say, “cylindrical projection” as the source of those coordinates, i.e. not using UV maps at all for the concrete. (Since it’s basically a uniform texture over what was, after all, a thing made of poured stone, it hardly seems necessary to use a map for it…)

For the oily streaks, first of all, you’d only need to be concerned about mapping those places where the streaks actually will be placed. And you’d probably position those maps yourself (remember: they can overlap) to add semi-random variety to the oily mess. If you wanted a graffiti to wrap around an edge, you’d map those areas such that the map-segments are contiguous. Such a map might look wrong for, say, concrete, or painted lettering or what have you, but be perfect for the graffiti that was sprayed onto it. Which would be fine, because that map would be for that purpose. A map for graffiti; a map for oily streaks.

When the real-world effect on a surface is, in the real world, composed of several different things, I suggest approaching each one individually. Different textures, and different maps. Then compositing. (This notion really pays off when you want to ‘tweak’ something to get it just-right, and you know that you can actually do that without waiting another X hours to see the results… which is basically why I try to do everything this way.)

Brilliant, thanks Sundialsvc4, will try that. Sometimes the obvious stares you in the face and you don’t see it.
Dicon