The Semantics of 3D Form

PROJECT INTRO

Project goals:

● Able to see and translate natural and
human-made forms through various
mediums
● Able to translate how individual elements
constitute a larger form
● Develop 3D forms which have meaning and
are aesthetically balanced and composed
● Able to understand and speak to 3D form
aesthetics, semantics, and their inherent
meanings

This project is broken down into 3 portions: two models based on a natural form (seashell), two models based on an artificial form (computer mouse), and a final hybrid model that incorporates formal characteristics of both.

NATURAL FORM: SEASHELL

Sketching and First Iterations:

Although I have my physical shell with me, I thought it would be important to have a 3D scan of it so that I could manipulate it digitally and so my peers and instructors can have a better idea of what I’m working from, instead of pictures. I also thought I would learn about the form and the hierarchy of forms needed in my recreations by transferring it to a digital model. I took around 70 photos of the shell and processed it using Meshroom, and hosted the model online using Altizure.

Source images for my model

Modeling process in Meshroom

Form | Framework | Texture

The interactive 3D model of my shell can be found here.

The model is very accurate to the shell for the top portion, but the photogrammetry software isn’t detailed enough to cleanly separate where the shadow meets the form and creates an inaccurate subform near the base.

Form analysis and notes on features

After learning more about the form of the shell, I went around and looked for tools for carving and sculpting my forms. I found some that will be too textured for what I’ll be doing for this project, but I wanted to try out a wide array of tools to find the ones best fit for translating these forms.

Sculpting and Carving toolset

SHELL PROTOTYPE 0

I wanted to understand the material and tooling limitations before diving into my first prototype, so I decided to make some very rough ‘first-steps’ prototypes in foam and clay, to warm me up for the first prototypes.

Clay Prototype 0

This exploration was made almost absentmindedly, and I began by just making the simple two-cone main form of the shell. From this main form, I worked subtractively for the rest, using a wooden cuticle stick to define the part where the spiral ends in a leaf-like manner. I also imprinted the spiral moving up to the point. This is very unrefined but the goal was to just get my mind thinking about how I will physically replicate the shell in this medium. Moving forward, I will likely use an additive technique to create that final layer of the spiral that separates from the primary form, and wrap it around a center piece.

Foam Prototype 0

The foam was more difficult to work with than I had imagined, I have experience with floral foam that is much more forgiving as far as fine texture changes are concerned. With this foam, each cut has to have a very finely defined entrance and exit, otherwise, the horizontal line where the blade exits the surface leaves a distinct line that interrupts the smooth surface. For this prototype, I cut a rectangular prism from my larger block of foam and carved off pieces from that block. This led to some of the fineness and proportions of the form to be lost. I think for my next prototype, I will draw the three outline planes with a sharpie and cut all of the way through the object, then refine it from there.

Technique for prototype 1 for carving complex forms

After learning some of the attributes and limitations of each material, I’ve decided to go back to doing some more drawing to abstract and translate the form before I make my first prototypes.

SHELL PROTOTYPE 1

Clay Prototype 1

3D scan here

I forgot to take process photos during the construction of this prototype, but I essentially made two clay cones (for the dominant and subdominant forms) and then rolled out a flat piece of clay for the final separated sheath. The spiral on the tip was carved in using a cuticle tool and sculpted round to replicate the shell. For the engineering of the sheath, the clay is much more responsive and easy to use than the foam, because with the foam I have to use a two-dimensional blade to carve out the inside of a complex bowl form, while not interrupting the dominant form on the opposing face of this channel.

Foam Prototype 1

3D scan here

For my first foam prototype, I began by drawing the outline of one side of the shell onto the face of the foam, and cut all the way through, creating an extrusion of that outline. I then rotated it horizontally and did the same again, leaving me with a very cubic representation of the shell. From this, I used a smaller blade to cut off the corners, making the cones of the shape have 8 facets each, and then I refined the curves from there.

While I feel like this is a major improvement from my first stab at it, there is a lot of refinement still needed and some more decisions that I have to make. These include-

• Do I render the spiral at the tip? Or is that considered texture and I should just make a conic form?
• And how do I deal with the last sheath that separates from the major form? Does this need to be included?
• How can I carve out the inside channel with something like foam?

I need to consider these things moving forward.

During our class crit, I learned that my struggle with carving the inside of the channel of the final sheath wasn’t a unique struggle, and there isn’t a good way of fixing it aside from looped carving tools, which I don’t have access to. I learned that heat can also be used to smooth both the clay and foam, which I may try in the future, but I’ll focus on using the carving and sculpting tools I developed from the first prototype for my next steps.

SHELL PROTOTYPE 2

With this round of prototypes, I zoned in on some of the finer details and subtleties of the shell. I received a lot of positive feedback from our class crit session and used the particular notes to help me advance my existing models. Due to the amount of time and limited materials, I decided to modify my existing models for this round instead of making all new models.

Clay Prototype 2

Foam Prototype 2

I defeatured the tip of the foam shell, removing the spiraled grooves because it’s more difficult to pick up on the small form change from the groove in the foam. I decided to attempt to capture the overall form of the point, instead of having the texturing of the spiral get in my way.

Both Prototype 2s referenced next to the shell

Having the side-by-side photographs of both models next to the actual shell made it very easy to critique the form differences, as it was almost a ‘spot the difference’ game.

Some feedback-

Foam

• Thinner tip
• Get rid of the deep curve, smooth it out
• Make point more gradual
• Open the fold/sheath a bit more, larger on the shell
• The taper should start earlier

Clay

• Rounder spiral
• The ‘facet’ or where the conic shapes meet isn’t circular, mine needs to incorporate more of the curve

In general, I need to focus on the points of connection for my models and make sure that I nail the transitions and curve shifts. My form is a little more simple than the very dynamic starfish or some of the other spiraling shells, so the subtleties on the large smooth forms are much more apparent. I plan on revising my clay model and making a new foam model for my final.

VIDEO RESPONSE 1

Although I had said that I would modify my clay model for my final, I decided to completely start over as I wanted to have a blank slate with which to build proper proportions.

The very beginning formation of my final clay model

I also restarted my foam model, using proportionally correct photographs to identify the outer bounds of two of the dimensions.

Prepping my final foam model

I refined both my foam and clay models in a similar fashion to how I created my first two prototypes of both, so the method and thinking were the same, but I incorporated what I learned from the class critiques and further looking at my form.

SHELL FINAL

INDUSTRIAL FORM: COMPUTER MOUSE

Sketching and First Iterations:

I began to draw and map out my computer mouse in order to understand the form and contours. I learned a lot from my progress on the shell, which guided me to do some more effective drawings for the purpose of these models.

I also decided to ditch making the 3D models of my references and prototypes, as these were time-consuming and less effective than simple side-by-side photo comparisons.

I’m using a Logitech M705 for my reference mouse.

My drawings

Reference images from https://www.cgtrader.com/3d-models/electronics/computer/marathon-mouse-m705-by-logitech

For right now, I think that the thumb buttons on the side won’t be included because of how little they interrupt the overall form, but I’m still going to try to capture the scroll wheel and top button, even if they are subordinate forms.

MOUSE PROTOTYPE 1

Paper Prototype 0

I found a very useful method for the planar breakdown of complex forms. I used picture hanging wire to wrap a line around my mouse reference, then slid the formed wire off the reference to have a 1:1 accurate cross-section of the contours of that plane. I also tested tin foil, but the resolution of the final cross-section was much worse, so I used the picture hanging wire to create templates for my first exploration.

This exploration was a very ‘to the point’ and accurate mapping of the form, but not a very effective method for communicating. While the planes are geometrically and proportionally correct, the location, spacing, and angle of each plane doesn’t help to communicate the form of this specific mouse that well.

Paper Prototype 1

I used tape and sketchbook paper to create a maquette for my prototype

Based on my first exploration, I decided to simplify the model and use the existing planes on the actual mouse and translate them into a mix of flat and curved planes on the model. The model consists of only 3 pieces with no adhesive, and I think is much more successful than the first exploration.

The only thick stock of paper I have is from file folders, which are unfortunately blue and not very thick. While I think these will work ‘fine’ for what I need them to do, a less flexible thicker stock would be more effective, so I’ll keep looking for it for my second model.

Paper Prototype 0 and 1 with Mouse Reference

Foam Prototype 1

I used a very similar technique that I had previously used to carve my shell to carve my mouse to of foam.

MOUSE PROTOTYPE 2

Paper Prototype 1.5

In order to get an accurate 1:1 model of all of the curves of my mouse, I created a mold out of tin foil and cast it with hot glue and tin foil. From this casting, I drew the lines of where the planes cut the form and drew them on using a sharpie. I cut the casting along these lines so I had the accurate planes at these points along the form.

I then traced these slices to have the paper cutouts match the mouse, and created a lattice that matched the mouse. This lattice had 5 planes running on the minor axis and two running along the major axis, and then I added one piece that wrapped around the sides for where the grips are. I was unsure of how to close off the top, so I made a very rough mockup out of copy paper and used that to approximate the top, but I didn’t add anything because the subtleties of the curves were lost with that top piece.

Foam Prototype 2

Feedback:

Q’s feedback on improved curves and proportions, but still some work to be done.

Left: The top sheath mockup I had made covers up the fine details of the top of the mouse, so a more minimal plane or no plane could be used. Right: The expression of the side curves of the paper isn’t fine enough, their too flat and don’t capture the curves

I have a lot to still refine for my paper form and have to consider a lot for my final version, but my foam version is very close and I think that a few small tweaks should have me pretty close.

I knew that I had to maintain some form of lattice for the paper form to create some structure using the flimsy file folders, so I created a different grid using the measurements I achieved from my foil mold.

The above image shows some of my initial cutouts for how to create a 3x5 sheet lattice, but this doesn’t incorporate the subtle curves that need to be accentuated that we discussed in our class critique. Based on the lattice model I added panels that connected in to create a communicative but expressive model, a step forward from my strictly accurate model 1.5.

The final piece layout

For my foam model, I simply made minor tweaks and refinements to my existing model, shaving some small pieces off to make a more similar outer edge.

MOUSE FINAL

HYBRID FORM

Word Association Blender

I used a word association exercise to create a set of terms that described my forms to guide my hybrid form, instead of making a direct visual hybrid. This exercise should distill form qualities from both objects, making the hybrid form more expressive and less replicative.

SHELL: Sheath Spiral Radial

MOUSE: Round Pointed Arch

HYBRID: Spiral Sheath Pointed Arch

Spiral is the uncommon word here, coming from the mouse, and the rest are all shared in both forms.

This generative chair was designed using AI for Kartell, and the iterative process shown in the video below was interesting in how I can use generative design for my hybrid form.

HYBRID PROTOTYPE 1

I began by drawing some simple proposed ideas but I knew that a lot of my thinking would occur when I began working with the clay.

I created very simplified clay models as a warmup to help get my hands thinking about the mouse and shell forms.

First prototype

My first hybrid prototype was a little too focused on combining the forms of my two objects and not enough on exemplifying the form descriptions that I had come up with. I decided to eliminate the word Sheath from my word set in order to clearly communicate the other 3 terms.

I prototyped multiple ideas using the clay that followed these terms, eventually landing on my final form and method.

The final process for my final model.

I created a DIY products light table to photograph my hybrid for the class presentation, using tracing paper, a glass table, and a reading light.

HYBRID FINAL

REFLECTION

It was difficult for me at the start of the project to know when a model was finished or accurate enough because when replicating a form the level of detail and proportional correctness could be chased forever. Luckily, many constraints such as time, sanity, limited tools and resources from the quarantine, and specific feedback helped me meter what was important for me to learn from making these models. Beyond learning the constraints and abilities of each of the materials, I was forced to really understand all of the subtleties in form of my objects and learned how to breakdown and analyze the forms into more comprehensive models and abstractions. The hybrid model was the step into completely designing a form, which I really appreciated after two weeks of replicating other forms. I learned how important it was to gain form descriptor influences from objects without directly replicating the objects. I’m happy with how all of my models came out, and I think they’re successful in different ways of communicating different aspects of the forms through their different materials.

ALL MODEL PHOTOS