Archive for February, 2010

UV mapping…

Ok, for those who may be reading this and unaware of the term, UV Mapping is the process of unwrapping your 3D model into a 2D map that can be colored and used to texture your object.

It’s a pain!

Ok … actually, for most of the parts of Redeemer, it’s not so bad. The handle is a cylinder, the guard a box, and the sword is planar. Most UV tools have those types of projections (as they’re called) built-in.

But those quillons….. Why oh why did I decide to twist them like that?

But, I got a good tip from my friends at Aery Soul. Since UV mapping maps the texture relative to the object, not to absolute coordinates, it would actually be possible to map a straight square column of the same dimensions as the quillons, and then align that map to the twisted one.

So, that’s my goal for the weekend. Even though the Colts are playing in the Super Bowl on Sunday, I’d like to finish the weekend with the UV mapping done on Redeemer. I’ll let you know how things go next week!


Redeemer – the Guard

The last piece that I reworked in the sword is the guard. Once again, my buddy As Shanim provided the insight on how to improve it.

The first tip he shared was that mine had way too many polys in the low-poly version. Having more polygons to start with makes it a little harder to control how the smoothing affects the final mesh.

His other feedback was about the ridge in the middle. He showed me how using a flat planar surface instead of just raising an edge down the middle works much better. It allows for a cleaner profile, and it also allows me to make the ridge wider or thinner depending on exactly what effect I’m trying to achieve.

To talk through the modelling process a little…

I started with a plane with a single face. I decided not to work with the cube because tesselating a cube is harder than just getting the faces right on a plane.

I added the vertical edges first. I kept one edge near the back of the guard, since I knew I wanted to keep that part pretty flat. The others I spaced out as equal as I could going by eye. I could have used absolute coordinates and done the math to divide it all exactly, but I felt like that wasn’t necessary.

Next up were the horizontal edge. Here I had a challenge. I didn’t want a central edge, I wanted two parrallel edges close to the middle. But, splitting it down the middle is easier. Finally, I figured that I could go ahead and split the faces exactly in half, then extract two edges exactly parrallel to the center line. Then I decimated the centerline so that it kept the faces.

The other horizontal edges were added by tesselating the upper and lower rows of faces exactly in half, then in half again.

Now I moved the vertices of the corners of the edge faces in either X or Y to conform to the outline of the guard on the reference photo. I ended up with one quad in the corners which didn’t work; so I welded a couple vertices together and made a triangle instead.

I did the same thing to the inner faces that outlined the depressions I wanted in the top and bottom halves of the guard. Once the outline was what I needed, I did an axial extrusion of the depressed faces straight back on the Z axis.

I tested the look a couple of times by smoothing the plane. Once I was happy with what I had, I replicated the face for the other side of the guard, bridged them together, and added the necessary edges to keep the back square.

That last part was a bit of pain because you end up creating n-gons by accident. Thankfully my modelling tool will automatically highlight n-gons; so I could see where I needed to fix my edges pretty easily.

This is probably the last update on the modeling process. I did change the blade slightly, but just by adding another row of edges along the edge of the blade to keep the smoothing process from making it too dull.

Next up: UV-mapping the objects so that I can apply textures to them.

Nothing new

Just a quick note to say I haven’t abandoned this project or anything. Last few days were a bit hectic with a visit with my kids and stuff. I have reviewed As Shanim’s feedback on the guard portion of the sword and expect to be able to complete that in the next day or two.

It’s really great having an expert like him to review what I’ve been doing. The mechanics of modeling aren’t all that tough. But sometimes it’s hard to know what the optimal choices are. He’s helped me a lot.