Blender Underground

Nope, neither accept shadows.
The difference is that Nor maps use RGB to map more 3-dimensionally, but otherwise it's the same.
The setting for bump maps is to turn off the RGB button, otherwise it's the same.
http://wiki.blender.org/index.php/Doc:Manual/Textures/Maps/Bump_and_Normal_Maps
You really need to make Nor maps in blender wheras Bump maps can be made in a 2D program.

How do I best uv-map a sphere? Boxy objects like cubes are simple, but do I do this with e.g. a football?

Well, it depends on how much distortion you're willing to work with...

If you want very little, I'd cut it up into slices, either length-wise, or around the middle. Or maybe some of both. Think of the different types of maps; you know, mercator projection, and so forth.

Err, no, I don't know

Slices... hmm... You mean marking seams from bottom to top and then aligning them like a plane? I tried to lay them out like a 2D-world-map. Well that seemed rather difficult. I think I'll try your method. Thanks!

Oh...like, on maps of the world. Some of them are square, and some of them are all bumpy along the top and the bottom; that's to minimize distortion. Because a map of the world is basically an unwrapped sphere.

So...a pictures worth a thousand words, right? well, here's how I'd unwrap a sphere.



Only one of those cuts goes all the way to the bottom of the sphere. The rest stop for two faces in the middle. The easy way to do this would be to select seams with the loop select from the top to bottom, mark them, and then select two rings of faces, and remove the seams from them. Then add one that goes from top to bottom.

This will be fairly distortion free; but it will also be fairly difficult to map an image to, if you want to put something in particular on it. You might want to do it in fewer slices, for ease of use.

If you really don't care about distortion, it might be worth simply projecting each half as a circle,and working with that. It'll give you the most distortion, but it will also be easy to draw on. Soo...pick a balance. Also, this projection will tend to distort around the sphere. projecting from one end or the other will tend to distort along the sphere. Here's a projection that might be easier to work with, once you stitch it up.



This will have more overall distortion, probably, but it's a bit easier to draw on. Or you could do half of one, half of the other, and do the middle like the top image and then ends like this image....really, it depends on what you're doing.

Thinking about distortion, imagine looking at the object. Or really, just look at it. Pick an angle. the bits that look flat from this angle will have minimal distortion; the bits that are steeply sloped will have a lot. The top projection looks at the sphere from the equator; the bottom looks at it from the poles. So you really need to pick where you're going to need the most detail.

Hopefully this answered your question. Hmm, there's some good stuff here. I might need to add some of this to the tutorial.

Thank you for the answer. I don't want to map an image, I want to bake a normal map to an uv-map. I read that the better I arranged my uv-layout, the better the map will look. I don't know if your trick will help a lot there...

Actually, it probably has more to do with distortion than the arrangement.

Just cut it up into lots of chunks, and you ought to be fine.

My apologies for not commenting sooner, I'm neck deep (maybe nose deep) into my math studies, and when I manage to pop in here, I've usually been in the Questions and Help forum.

This is an excellent tutorial, Pyro. Thanks for that. I was pleased to find the "Sync UV and Mesh Selection" feature, which is a very welcome addition, and one I didn't discover until skimming the tutorial.

I wanted to comment on something that might change the way that pinning and unwrapping is viewed. First let me say that pinning a vertex (P key, UV editor) keeps the Unwrap command from moving it. However its usefulness is only fully realized when "Live Unwrap Transform" is activated (apologies if this has been mentioned already and I missed it).

Live Unwrap Transform (UVs menu) constantly reevaluates the Unwrap command; that is, it continuously executes an Unwrap anytime a UV vertex is moved. However it is impossible to see the results unless you move a pinned vertex. This seems counterintuitive, but since Live Unwrap Transform reevaluates the UV mapping around the pinned vertex, moving a pinned vertex causes all the non-pinned UV's to change. Try it, and you might be surprised.

1) Throw down a plane and subdivide it a few times.

2) TAB into edit mode and Unwrap it (U Key).

3) Open the UV editor and pin a few vertices, maybe 3 or 4 verts from different parts of the mesh (P Key).

4) Activate Live Unwrap Transform (UVs menu). The Sync UV and Mesh function must be off.

5) Move one of the pinned vertices and note the response.

The benefit is more easily realized with a complex mesh, but if your reaction is anything like mine, you'll say something like, "Wow."

With Live Unwrap Transform, the pinned UVs move like normal vertices, and all the other UVs are transformed by Unwrap to accommodate. So while pinning a vertex might be useful for just what it implies, that is keeping UV vertices from moving during successive Unwrap commands, it can also be an invaluable tool for the mapping process when used with Live Unwrap Transform.

EDIT: It's important to have at least three pinned vertices, even four, to see the impact of pinned vertices with Live Unwrap Transform.

EDIT2: For the test case above, it seems useful to pin the corners of the plane, then pin a few vertices inside and move them around.

Ah! Thank you. I will add this to the tutorial when I have time.

(Probablly in a few months.)

Well a few months has gone by, since the last post. I found this tutorial useful, I got a slightly better grip on the UV tool now, so thanks.

However, there are a few comments I like to add as a complete newbie.

I now using Blender 2.48a. The U key no longer works as expected. I expected to see some kind of 'T' shape, but all I saw was one square. (This was for a cube.)

You talk about the UVtex, yet my UV picture will work with that text enter or that text box completely blank. So I don't see the point of this yet.

Also, one thing that I found odd. Was when you apply UV, I selected the map to 'cube' option. My picture wasn't working as expected. Until I selected the map to 'flat' option. This to me at least seems a bit strange. I know the UV map is flat, buy my cube is not. Flat option worked, I got the picture I expected, but not with the tool I was expecting to work.

Also your discussion about the different tools, while useful information. I didn't use at all. As there was no examples of them at work. I recommend perhaps doing a mini series of UV tutorials.

1) Cube, painting a face with a different colour. Showing that and how to get the face colour you want at the front.
2) More organic model, with a demo of some of those tools you mentioned and how each one benefits you, in that situation.

Thanks.

many months ?

was only posted this morning lol



PYRO ISN'T A MACHINE !!



...not yet though hehehhh



(Grumbling evils plans into head.... getting bored gone work on Strub's player)

Only HALF machine.




You probablly had all six faces on top of each other. This is determined by where you put your seams; it will unwrap differently.



This will come into play if you have more than one UV face layout in a file. No, it's not a problem if you have only one.



This is really out of my range of experience; I'm not sure how all of the "map to" buttons work. When I have time, I'll try an experiment with this.



Yeah, this tutorial could definitely use more examples! But it was a fairly rushed job...hopefully I'll get around to re-writing it one of these days. I should really pick a fairly complex example, like a human head, and go through the whole texturing process. I'm sure I'd learn a lot, too!

Anyways, thanks for the feedback.

That would be awesome. A human head is a really good example.