The Semantics of 3D Form

Chelsea Tang
10 min readApr 22, 2021

Day 1

April 21, 2020

Task: create the form of a shell and a mouse using two mediums

Sketches:

Shell drawings

Clay: additive and subtractive

Soap: subtractive

Paper: additive and subtractive but more difficult

I first decided to do a clay shell to explore the form in a medium that is forgiving. The clay allowed me to add and subtract without consequences.

Building the base:

First pass

I then added spikes by just creating a ball of clay and smoothening the edges onto the shell form.

Added detail

Prototype 1

Prototype 1

Issues:

  • back slant not steep enough
  • ridge markings on skirt too detailed
  • edge lip too prominant

Day 2

April 27, 2021

Feedback:

  • shell back is not drastic enough
  • shell curvature is a C curve and not a S curve
  • get the ridges correct (right placement and height of valley)

Then I started to work on the paper shell.

Sketches

I decided to go with an approach with an internal skeleton and outer cover for the shell.

Attempting outer shell using an outer cover

During the process of making this, I realized a lot of the things I envisioned in my mind would not be able to actually execute in real life. Having multiple pieces of paper in a spoke manner was challenging since I had to think of a way for the papers to slot together.

Making a more accurate version.

More accurate

To create the outer cover, I tried this bisection method where the form is cut and another piece of paper is inserted.

Doing this was really difficult since the spokes would have to be cut on an angle since the shell’s curvature is slanted. I realized this would not be a suitable solution.

Putting planes between the spokes

Instead of creating a bisection, I added planes of paper in between each spoke and attached them using slits and tabs. This was the most reasonable solution to create the outer cover; however, since I did individual planes, it was very time intensive.

Final
Paper model

Moving forward:

  • Make the spiral area more cohesive

Day 3

April 29, 2021

Feedback:

  • Paper model — create a cut out for the inner concavity
  • Good conservation of plane/volume
Critique from Wayne

I then started my mouse form by sketching out the proportions. I realized that my mouse was uniquely shaped and had many non-symmetric curves and angles.

Drawing proportions

This exercise helped me understand the curves of the mouse and place them proportionately. I started out by creating the general shape in soap and then going back and refining the form.

Work in progress
Bottom contrast

The bottom of the soap I was using had a curved bottom, which made it harder to create a flat bottom that is seen in the mouse. I decided to not make the bottom flat because I would have to scale the mouse down significantly (and perhaps sacrifice the top curvature of the mouse) because the bottom curve takes up a large part of the soap’s volume.

Final soap

Moving forward:

  • Maybe change scale to achieve a flat bottom

Day 4

March 4, 2021

Feedback from Daphne:

  • Rounded mouse bottom is fine, limitations of the medium
  • Make the top edge of the mouse buttons more angled
  • Reduce the prominence of the mouse wheel
First version

I wanted to analyze the mouse better so I decided to take the pictures and mark them up in procreate to compare the proportions and details.

Here are the things that needed to be changed:

Changed version:

Clay mouse

Moving forward:

  • Make the mouse smoother
  • Further adjust proportions

Day 5

May 4, 2021

Feedback from Q:

  • Make some adjustments to the shape and form
  • Adjust the concavity on the left side of the mouse to be smaller
  • Check corners for discrepancy
  • Adjust shell inner concavity to be larger
  • Add some volume and detract volume from the shell body (areas shown in the last picture)

There were a lot of changes that I had to make. I would say the paper was also pretty easy to make adjustments since I could swap out some of the certain pieces for new ones. The soap, however, was very challenging since I could not add to the form and would have to restart.

Editing the paper shell

I basically just cut the paper at the pencil line so that it could accurately hit the nubs on the shell to convey the highest points of the curve.

Paper shell underside, before (left) and after (right)
Paper shell, Before (left) and after (right)

I basically just cut the pieces in between the shell to make it line up with the nubs, cut the inner concavity to be larger, and added a plane behind the concavity.

I added the plane because I thought that in other views besides the underside (like the side view), the gaping hole was distracting and not accurate to the form.

Large hole that should not be there in this view

I then moved onto changing my soap.

Flattening the mouse

I flattened my mouse to see if it would be ok. I think it worked pretty well but I still had issues with additions to the form. I made some changes to the form to adjust it to the critiques.

I also made changes to my clay mouse and shell, but I think that my clay mouse still has many issues.

After the class, I made some adjustments to the clay mouse. I messed with some of the curves and the middle section.

Before (left) and after adjustments (right)

New final photos for the clay mouse:

Final photos for the soap mouse:

Final photos for the shell:

This project really made me consider form in an analytical and careful manner. Although I originally thought I had a good eye for detail, I realized that I was creating my models based on how I perceived the objects rather than their actual form. For example, I would make certain features larger on account for it being more memorable in my mind when the scale is actually smaller. I tried to combat this by comparing the silhouettes of my models and objects without thinking about what the object is. This helped a lot with seeing the problems, but it was difficult for me to make all the changes.

Since the model is a 3D object, it was difficult to change the form of something in one view without changing the form in another view. Sometimes when I made something look pretty accurate in one view, it would make the other views messed up. Going back and forth between the views and adjusting accordingly helped, but it would never be perfect. I think that if I had more time, I would just work more on this aspect.

--

Chelsea Tang

Design and CS student at Carnegie Mellon University