Project 1c - Mt Bearmore

After lots of experimentation this week with my object, I believe I finally have a surface that might turn out pretty good from the 3D printer.  The most interesting part of my experience with modeling this week was that by the end, I felt like the simpler surfaces ended up looking better as polysurfaces and ultimately seemed to be a better form when extruded.  Luckily, if I am wrong and my polysurface is too simple, I have the other ones saved and can quickly switch one out.  I just felt like when the surface folded back in on itself, it no longer seemed like a viable option.  Plus, the final surface that resulted not only has some interesting curvature, but also looks like Mt Rushmore if you squint really hard, therefore coining it the name Mt Bearmore.
   Taking you through my experience this week you will see how my surface didn't really want to be a surface most of the time.  I believe this has to do with the large number of control points that I had modeled my surface with, since when I would scale or rotate my surface in different ways throughout the matrix, there would be fairly significant impacts due to the number of control points and the different directions they wanted to go.  So as a learning experience, if I were to do this again I would try to create my surface with less control points.  Now to walk you through the process that I underwent this week.
   I actually first started this week by trying to model my object in Maya.  I wanted to try out Maya since I had been solely working in Rhino these past couple of weeks.  That went poorly so I didn't save pictures of my bear since it looked like a beach ball.  Just as clarification, it had nothing to do with Maya as a program, just my inexperience with it and the fact that it doesn't like Rhino commands.
   So back to Rhino it was.  If you remember from last week I wasn't thrilled about how the head came out in Rhino as my final surface.  Therefore, I decided to try a new method of modeling my object, one that was recommended in class.  Instead of modeling my object using a number of different surfaces for each feature of the bear's face, I modeled profile lines of the bear's head in hopes to create one overall surface.  This is shown in Image 1:

After creating the surface for it the first time I felt like it actually didn't come out that bad.  It kind of looked like a pig, but a lot of the smoother curves to the bear's face were coming through in the surface.  Plus with some tweaking I would be able to indent the eyes and fix the problem I was having last week.

As I continued to morph my surface, I became aware how sensitive the nose of the bear was to to the different number of control points used when creating the surface.  This, coupled with the fact that I wanted to be able to manipulate the area around the eyes, led me to create the surface with a lot of control points.  I also began to manipulate the surface of the bear's head to see what kind of range of motion it had.

It quickly seemed to me that the most interesting parts of my surface were the curves and lines around the bear's eyes.  This led me to project a number of lines onto that part of the surface, essentially creating a new surface.  This new surface would prove to be much easier to manipulate, while at the same time reacting the most to the manipulation.  At first I also tried to see what kind of difference the different number of control lines would have when lofting the curves in Rhino, and ultimately I chose a lofted surface with a relatively large number of control lines (around 20) just because I thought it looked better.

I started making the matrix of my morphing surface next.  In the pictures below, the bottom right hand corner is the original surface and the middle top edge and the top left corner are the most manipulated surfaces.  As I started going up the first column of my matrix, I was rotating the left side of the bear's face in a 2D plain (from the top view) to see if it had any effect on the right side of his face.  It did not but it did have some pull on the nose.  Not being deterred by the fact that the right side of his face was still in tact, I figured this would be a good place to begin the next set of morphings.  Going up column two (from the right), the right side of the bear's face was now being rotated, but not just in a 2D plain but actually in a 3D plain that I tried to keep as constant as possible.

Column three then continues with a rotation of the nose of the bear since it was the only really recognizable part left.  This rotation I did in more of a vertical plain to see what kind of effect it would have on the surface as it started to double back on itself.  It almost led me to my final polysurface, but I decided it was a little bit too distorted.

So I went back and copied column 1 to column 4 to start relatively fresh.  In this column I stretched and rotated both sides of the bear's face using control points.  It wasn't a very exact method but produced relatively interesting results.  In the final column I tried to make the surface double back on itself again, but less extreme this time.

I picked a couple of my favorite surfaces from the matrix and then started seeing if I could refine them and prepare them for extrusion.  What I found was that most of the one's that I had picked were actually too complex, and by having the surface double back on itself, there were very thin local areas created on the surface.  Since they looked like 3D infinity lines I still thought they were interesting but I ultimately chose a surface that didn't double back on itself.

I went ahead and extruded this surface, and after the first time I could still relatively make out some of the bear's features, such as the nose shown in the image below.  I was left with the final product after scaling it down to size, and actually I scaled it even more than necessary in the Y direction (height) because I felt like it produced a better model.