semantics of 3d form

Assignment Objectives:

  • To be able to observe, translate, and recreate forms through different mediums (additive and subtractive materials).
  • Understand and explore the different levels of abstraction.
  • Be able to demonstrate a good understanding of the translated form. (If abstracted, is it too abstracted?)

PT I. Natural Form: Shell

For my shell, I decided to use the materials clay (additive) and low density foam (subtractive) to create my models.

Understanding the form of the shell:

Before I started sculpting or carving anything, I first did a few sketches to understand the form of my conch shell. This would then help me see the visual hierarchy of the components of my shell, which would then help me as I construct my shell with two different materials.

(above) preliminary sketches, looking at basic shapes and curves
(above) more explorations of the spiral and form of the shell

My Shell:

Iteration 1: Clay

Tools used:

  • Small Olfa Knife
  • Bone Folder


For my clay model, I decided to approach this by separating the form of my shell into three parts: the top part of the body, the end (tail), and concaved part wrapping around the shell (aperture).

To create the top part of the shell, I rolled out a piece of clay and wrapped it around a ball of clay. I rolled it so that it became a conical shape and added more clay on top when needed for the tip of the shell.

(above) I rolled a ball of clay and wrapped it in a sheet of clay to create the body)

I then rolled out another piece of clay and rolled that into a longer conical shape to add to the part I just made. After using my fingers to push and flatten the clay together, I rolled yet another piece of clay and cut out the appropriate shape for the aperture.

(above) refinement

For the details, I rolled out smaller cone shaped pieces of clay, cut them, and added them to the shell model. I only added the more prominent spikes from the center of the shell body and decided to simplify the smaller spikes on the tip. I also used the bone folder to help me smooth out the lines of the whorl.

Finished Iteration:

What to work on next time:

  • Be mindful of the whorl on the tip of the shell: each lower level of the spiral increases in distance.
  • Be careful of the proportion of the shell.
  • Observe and be mindful of how the spikes are placed and what direction they whorl.

Iteration 1: Foam


  • Large Olfa
  • Small Olfa
  • Xacto Knife
  • Marker

For my first prototype, I first cut out and drew the basic shape of my shell on on of the surfaces. This is so that I could get a sense of how big I want my shell to be and make sure I leave enough foam to carve out details. I then used my large Olfa knife to quickly get rid of the excess foam on the corners and get it down to a very general ovular shape.

(left) marked out the shape of my shell with a marker; (middle + right) used a large Olfa to get the general shape first

From there, I switched to my smaller Olfa knife as I began carving for the general shape of the shell’s body. After I got it smooth enough, I then used my Xacto blad to carve out the general shape of the top spiral and carved out the aperture the shells side.

(above) then used the smaller Olfa and Xacto to smooth the general shape out

Finished Iteration:

What to work on next time:

  • Don’t be afraid to subtract too much, just be careful of what you carve.
  • Technique: To get rid of unnecessary jagged cuts, cut as if you’re peeling a fruit.
  • Add more details, too simplified : figure out what gives your shell its own character and add that to the foam model.

Iteration 2: Clay

In my first iteration, the proportions of the shell were a bit on the flatter/narrow side overall so I focused on adding more volume in the second iteration. To do so, I took a piece of clay and added it to the side of my shell, then using my fingers to smooth it out.

(above) adding volume to my shell

As for the details on the top of the shell, my first iteration looked normal on the side but looked mismatched and off from the top view. To fix this, I made sure to change the angle of the shell I was observing often and looked at that for reference more often to make sure the proportions were better. In doing so, I made it so that the spikes were spiraling in one consistent direction and smoothed out the edges so that they gradually attached to the shell than jut out.

(left) changing the direction of the spikes; (right) close up shot of detail
(above) looking at how the end of the tail ends and curls into itself

Finished Iteration:

(above) proportion comparison from model to shell

What to do next:

  • Focus more and fine tune proportion, the original shell is much narrower and have deeper curves than my model. Try using different tools.
  • Top whorl of the shell model is too abstracted and smoothed out, keep it consistent with the level of detail given to the spikes.
  • Be careful of craft: make sure fingerprint lines aren’t distracting from the lines meant to show the spiral on the tip of the shell.

Iteration 2: Foam

For my second iteration, I mostly focused on making it so that the shell overall had more detail as my first model was too simplified. By adding the necessary details, it gives my model the distinct characteristics that my original shell has. I also tried to fine tune the surface of the model by slanting my blade to create smoother cuts.

(above) trying a different technique to cut smoother pieces

Besides adding the main spikes along the sides of the body, I also tried to imply the spiked pattern on the tip of the shell as well. I used my Xacto blade and attempted to carve out the spiral and the details.

Finished Iteration:

What to do next:

  • Like the clay model, fine tune the model more to fit the proportions of the original shell: foam model is also not narrow enough from body to tail.
  • Be careful of how much you subtract, I left too little foam to add the spike details on half the body in this iteration.
  • Try to be more careful in cutting out the whorl on the tip of the foam: try to avoid jagged edges that could distract from the intended cuts.
(above) side by side of foam and clay models

Iteration 3: Foam

For the third iteration, I did the same as the last one but aimed to capture the proportion more accurately.

I mostly focused on capturing the narrowness and length of the shell, as in the previous iterations for both clay and foam, they too thick in some areas.

Finished/Final Iteration:

Iteration 3: Clay

For this iteration, I took my previous model and cut the aperture off to add more volume on the side of the shell. This is because on the side of the previous clay iteration, there was a gap where the aperture started that was missing.

(left) cut off aperture; (middle) adding more clay for volume; (right) reattaching the aperture

I also used the bottom side of my Xacto blade as a mini rolling pin to roll out areas of my model. This then made the top part especially smooth, without the lines from my fingers.

(left) backside of attached aperture; (right) mini rolling pin! for spiral and for smoothing

Finished/Final Iteration:

1 Min. Video Checkpoint:

PT II. Mouse Form:

For my mouse form, I first found a 3D rendered gaming mouse and then later found one in person.

Actual Mouse:

The mouse I have with me in person is very smooth overall, there are no curves for where the finger or palm rests. As such, this mouse is very plain.

3D Rendered Mouse: FNATIC — Flick Gaming Mouse:

In contrast, the one I found online is less plain than the mouse I found in person. It definitely has more ergonomic elements that fit one’s hand, such as curves on the two sides and subtle pressed in pads for your fingers. Even the scroll is covered in ribbed rubber for easier grip. As such, I decided to use the 3D rendered mouse as a reference for my mouse form models, and the mouse I have in person as a model for hand to mouse reference.

Link can be found here:

To start, I first did a few sketches on the mouse I picked. This is to familiar myself with the object and understand its form from various angles. This was also an opportunity for me to explore the details of the mouse, such as how much the scroll peeks through on the side view or how curved it is from the back view.

(above) sketches looking at the form of the mouse

Iteration 1: Paper

Tools used:

  • Canson Watercolor Paper
  • Xacto Knife

For this iteration, I approached it by creating the base first, and then adding vertical pieces that hold up the structure.

(left) base shape; (right) length of base measurement for vertical pieces
(above) center vertical piece with the mouse scroll
(above) connection of the middle vertical piece
(left) fitting the far right and left side of the mouse; (right) all three vertical pieces fit together
(above) top hood of the mouse

One issue I found, however, was that there was no support on the two sides that held the left and right vertical sides together. This made the structural integrity of the mouse very weak, making it easy for it to sway or slant to one side. A compromise I made for this iteration was scoring the paper on both sides of the hood of the mouse so that the vertical sides would stay upright. This didn’t fix the stability issue though.

Finished Iteration:

Something I want to explore next:

(above) experimentation of paper scoring, indenting

What to do next/Critique:

  • The mouse scroll for this iteration is too flat, how to integrate it better? More 3D or implied (negative space)?
  • Try a different way of construction (stability).
  • Mouse is visually too vertically dominant, try gridding more: look at both x, y, z axis.
  • Top (hood) of mouse: try scoring paper to fit main curve of mouse.
  • Back end of the mouse: paper intersection should be flush.

Iteration 2: Paper

In this iteration, I mostly focused on using the ‘waffle’ technique more effectively to portray volume. I first created the base of the mouse, and cut slits in the back half of it. I then cut slits in the front half of the vertical pieces so that I could interlock the planes of paper.

(above) constructing vertical support

After I did that, I needed to ‘waffle’ in another horizontal piece of paper to create stability (which was what my last iteration mostly lacked). After I cut the slits for the second horizontal plane, I also trimmed the ends of it to make sure the edges were flush with the vertical planes.

(left) horizontal plane cut to match the vertical planes; (right) trimmed off part to make the edges flush
(above) main structure put in place

After creating the main structure of my mouse, I then tried two different shapes for the top of the mouse. I scored the back end so that it bend slightly to the form of the mouse curve.

(left) I ended up picking the second mouse top as the edges bowed outwards; (right) what the mouse’s underside looks like

Finished/Final Iteration:

Iteration 1: Foam


  • Large Olfa
  • Small Olfa
  • Xacto Knife
  • Marker

Like for the conch shell, I used three different types of blades for my foam model. Using the large Olfa for general shaving down and small Olfa + Xacto for detailing, I chipped away at the foam after having the general shape drawn on two sides.

(left) front view; (right) side view
(above) cutting down to a very general shape

Every once in a while, I would also redraw areas that I need reference to. This is so that I wouldn’t lose track of what shape I’m aiming for. I also slanted my Xacto blade for smoother cuts.

(left) using a micron pen, I dotted where I wanted the curve; (right) slanted blade technique
(above) almost done!

After refining with my Xacto blade some more, I used my bone folder to go over all the rough ridges around the mouse. This then made the foam overall appear smoother (as inspired by Michael, who mentioned he used his hands to rub over the surface last crit).

(above) smoothing foam out with bone folder

Finished Iteration:

What to do next/Critique:

  • Edge on the back end ends sharply, there isn’t angled cut.
  • Mouse in general is very long.
  • Keep in mind of the subtle angled cuts at the bottom of the mouse.
  • Mouse scroll: corners should be more rounded

Iteration 2: Foam

For this iteration, I redid the mouse again but focused more on proportion.

(left) I first carved out the sides and slowly carved the top part to match the curve; (right) I made sure to keep the back edge sharp
(left) front view; (middle) I left the scroll bit to trim for last; (right) I flattened the top curve even more

This time round, I made sure to keep the length and curves as accurate as possible, being extra careful around the scroll as well. I also made sure to change the angle of the reference model more frequently to prevent any accidental mishaps in cutting too much.

Finished/Final Iteration:

Once I finished carving, I cut out the middle strip surrounding the mouse scroll and used my bone folder to smooth out the entire model.

PT III. Hybrid Form:

Word Association:

Shell: Organic, Repetitive, Spiral, Twisted, Angular
Mouse: Symmetrical, Rounded, Smooth, Industrial, Long (for a mouse?)

Traits Picked:

  • Rounded
  • Smooth
  • Organic
  • Angular
  • Materials Used:
  • Small Olfa Knife
  • Clay
  • Glue stick as a rolling pin
  • Eraser for smoothing surfaces out

Rapid Ideation Sketches:

(above) quick sketches
(above) sketches done on my iPad

For my hybrid form, I wasn’t too sure as to what I wanted to do. I was really interested in the idea of having something rounded and smooth, but also angular—as inspired by the horns of my shell. I was interested in the trait ‘angular’ because putting ‘smooth’ and ‘round’ with ‘angular’ seems like they would clash with each other. But after thinking about it more, there definitely could be forms where its smooth and rounded but angular, such as a pointed object with a rounded top.

However, when I started sketching, I was then more intrigued by how the shell’s body was folded into itself, creating a twisted and curled shape.

As such, I first started by rolling out a block of clay. I then cut the clay in half and used the side of a large glue stick to roll it out flat.

(above) after rolling it out, I then cut it to the shape I want

I then used my Olfa knife to cut and trim to the shape I wanted.

After creating the strips of clay, I then curled it into a ovular shape, and connected another strip that wrapped around the first form.

I then stopped here as I was too uncertain as I wasn’t too happy with the direction I was taking. I was also unsure as to what to do next.

Finished Iteration (In Progress):


  • Needs to be more angular
  • Q suggested I could exaggerate the front and the back more (as seen in the picture)
  • Focus on the traits you picked out.

Iteration 1.5: Hybrid Form Clay

After taking the critique from the previous class, I then tried to shape my unfinished form to be more angular. I also tried using my knife to cut the rim of the clay for sharper and cleaner edges.

Iteration 2: Hybrid Clay
The words that I picked originally was rounded, smooth, organic, and angular.
But I changed it to smooth, symmetrical, angular because having both rounded and smooth was redundant and also due to all the traits were either heavily shell related or more seen in both shell and mouse. Symmetry then became my uncommon trait instead of angular because there were angular parts in both the mouse and shell, whereas symmetry is more distinct in the mouse.

(above) inspired shape from previous iteration

I ended up changing the design of it quite differently, and approached it differently. I still took inspiration of general shape from the first iteration but made more more block-like and sharper, angular edges.

I first rolled out a block of clay and used the side of my glue stick as a rolling pin to smooth out the surfaces.

I then used my Olfa knife to make a diagonal cut and then rolled the top and bottom edges.

I then used my Olfa knife to even out the bottom lip of the clay. I also used my fingers to dent and smooth the center and later used my eraser to further smooth the surface out by ‘rolling’ on it.

Finished/Final Iteration:

Prior to this project, I wasn’t too familiar with using additive and subtractive materials to create 3D models. Even though I have used clay before, this was the first time I had to see and interpret a real life object into my own abstracted form. I ended up enjoying foam the most, even though it was the hardest medium between the clay and paper. As such, one of the main things I’ve learned from this project is that it’s definitely easier to abstract from seeing a reference than creating something entirely new. The skills I learned from this project will definitely trickle into future projects.
As for my hybrid form in particular, I struggled to combine the traits I noticed into a hybrid form, even though I got pretty familiar with the natural and man-made forms. The final hybrid form I ended up creating was mostly inspired by the shape of the horns on the shell and the symmetry of the mouse. Even though aimed for it to look more sleek and symmetrical like the mouse rather than ‘organic’, but it ended up being a very simple and overly abstracted form. However, I’m still pretty satisfied with the direction I ended up taking with it as its definitely the direction I’m happier with than my first iteration. As such, there’s still a part of me that wish that I had more time to think my ideas through and showed that second iteration hybrid during critique instead…so that I could further the direction I liked more to an even better hybrid form.




Hey! I’m currently studying design @CMU with a focus on communications design + minor in HCI. You can find some of the projects I’ve worked on here.

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Sarah Xi

Sarah Xi

Hey! I’m currently studying design @CMU with a focus on communications design + minor in HCI. You can find some of the projects I’ve worked on here.

More from Medium

ESP32 with Web Server and LEDs

CS371p Spring 2022: Raju Kakarlapudi

Code Smells — VELV

Moonlight Express 面世了