Chapter 2 — Modeling Modes

DigitalClay3D supports four ways to build a silhouette. Pick the one that matches what you're trying to make; you can switch modes at any time but each mode owns its own state, so don't expect a polygon profile to follow you into Revolution mode.

You set the mode in the Shape tab via the Mode dropdown at the top.

Revolution mode

The default. The model is built by spinning a 2D profile around the Y axis. Edit the profile, the 3D shape follows immediately.

The Shape tab in Revolution mode breaks into two stacked sections — the Object Settings panel at the top (model name, mode dropdown, height & diameter, scale operations, profile shape) and the Revolution Profile + Wall/Floor + Base panels below it:

Panel	Screenshot
Object Settings
Profile + Wall/Floor + Base

The profile editor is the panel on the left. Each dot is a control point with X (radius fraction, 0..1) and Y (height fraction, 0..1). Drag a dot to reshape; click an empty area on the curve to add one. Right-click to delete.

Pop the editor out to a wider canvas next to the 3D view via the ⤢ Pop out button — useful when fine-tuning the curve while watching the result.

Profile control points in 3D — when Show Profile Points is on in the Dimensions overlay, the same control points render as draggable spheres on the 3D model itself, with their Y/X coords as tooltips. Lets you sculpt the profile directly on the 3D view.

Cross-section profile

By default the cross-section is a smooth circle. Switch the Profile Shape to Polygon to give the body a faceted feel (3-side triangle, 6-side hex, 12-side dodecagon, etc.). The Chamfer Radius rounds polygon edges so they don't print as hard lines.

Starter Shapes — silhouette

The Starter Shapes panel sits between Object Settings and the Revolution Profile editor. Each button replaces the current silhouette with a curated outline so you don't have to drag dots from scratch when starting a model.

When a parameterized shape is selected, a single editable field ("sides" / "tiers") appears at the top of the Starter Shapes panel — edit the value and the silhouette rebuilds live, no re-click needed. The active starter button is highlighted while it's the source of the current silhouette; manual drag-edits to the profile don't change the highlight, but loading a different starter will.

A Pick… link in the Starter Shapes header opens a checklist of every shape with previews — curate the panel down to the silhouettes you actually use.

Lofting — Single-Object

Lofting interpolates between user-drawn cross-sections at different heights. Useful when you need the silhouette to vary in shape — not just diameter — as it rises.

The Shape tab in Lofting mode shows three stacked panels — the layer list, the active layer's cross-section editor, and the Lofting Options group below it:

Panel	Screenshot
Layer list + add/remove
Active layer's cross-section editor
Lofting Options (rotate / random)

The Layers panel lists every cross-section by height percentage, top of the model first. Click a row to make it the active layer.

• + Add Layer drops a new layer at the top of the stack. Slide it down via that row's slider to position it between existing layers.
• Copy Active duplicates the selected layer and parks the copy directly above the original (between it and the next-higher layer), so related layers stay grouped.
• The slider on each row sets the layer's height (0% = base, 100% = top). Drag the slider to reorder — the list re-sorts itself as the height changes.
• The × removes a layer.

When a layer is selected, a yellow ring is drawn around the 3D model at that layer's exact perimeter, so you can see which slice you're editing without counting from the top.

The Cross-Section Editor below shows the active layer's polygon. Pick a symmetry mode at the top:

Below the mode row, the Transform strip operates on the layer's points as an "object":

• Rotate ↺ −15° / ↻ +15° — capped at one full revolution per layer so repeated clicks don't accumulate floating-point drift past 360°.
• Scale − 10% / + 10% — shrinks or grows the pattern inside the canvas. Once the pattern reaches the canvas edge, further + clicks expand the layer's physical diameter instead, up to the printer's inscribed plate diameter minus a 5 mm wall clearance. Only the active layer grows; layers above and below keep their own scale.
• Move ← ↑ ↓ → — translates the pattern by 5% of the canvas extent per click. ⊙ recenters the pattern's centroid on the spin axis.
• Zoom — three buttons in the upper-right corner of the canvas (+ / − / ⊙ reset) shrink or expand the visible viewBox. Useful for fine-tuning points on small shapes.

When the active layer is a parameterized starter (Polygon, Star, Cross, Flower, Gear, etc.), a single editable field for that parameter sits above the canvas — sides, points, arms, petals, teeth, etc. Edit the value and the loaded shape rebuilds live without re-clicking.

Adjacent layers render as faint dashed outlines behind the active curve so successive cross-sections can be aligned visually. The layer-below ghost uses the accent colour; the layer-above ghost uses a muted grey. The ghosts scale by each neighbour's per-layer Scale ratio so a physically wider layer reads as a wider ghost.

Style is Smooth (Catmull-Rom curve through points) or Polygon (straight segments). Pop the editor out for a bigger canvas.

Starter Shapes — cross-section

Below the Style toolbar, the Starter Shapes grid replaces the active layer's points with a curated outline so you don't have to drag dots from a default circle every time:

After clicking a starter, the parameter input above the canvas drives a live rebuild. Manual edits to the points (drag handles, transform buttons) take effect immediately, and the layer's existing per-layer Scale carries through to the loaded shape.

As with Revolution's Starter Shapes panel, a Pick… link in the header opens a checklist for hiding cross-sections you never reach for.

Lofting Options

The mesh lofts (extrudes + interpolates) between consecutive layers, including per-layer scale so an upper layer with a larger physical diameter widens smoothly toward the top.

• Contour Rotate — twists each layer relative to the one below. 0° produces a clean stack; 180° spirals halfway up the model; 360° completes one full revolution. Capped at 360° because beyond that the geometry just repeats and the slicer struggles with self-intersecting toolpaths.
• Random Lofting — toggles a small radial perturbation on intermediate rings (the rings between user-drawn key layers). Key rings stay exact so your drawn cross-sections aren't distorted; only the in-between geometry gets the organic noise.

Lofting — Multi-Object (experimental)

A different lofting engine that lets you place up to 8 round or square primitives per layer. Adjacent layers' primitives merge automatically: three circles at the base smoothly become one circle at the neck — the classic octopus-base / branched-vase problem solved with a single watertight mesh.

Enabling it: Settings → Experimental → tick Multi-Object Lofting. The Lofting Mode dropdown will then appear at the top of the Shape tab's Lofting panel, alongside the Single-Object option.

The plan-view editor is a top-down 2D canvas:

• Click + Round or + Square to drop a primitive at the canvas centre. You can place up to 8 per layer.
• Drag the body of any primitive to reposition it.
• Click a primitive to select it; corner handles appear.
• Drag a corner to resize; the cursor follows even when the primitive is rotated.
• Numeric inputs below let you fine-tune Center X / Z, Width, Height, Rotation°.
• The previous and next layers' primitives render as faint dashed outlines so you can stack shapes vertically.

The mesh is generated using SDF + Marching Tetrahedra: each primitive's signed distance field is rasterized to a 2D grid, layers are linearly blended along Y, and the 3D iso-surface is extracted. The result is watertight and topology-changing, so 3-tubes-into-1-trunk is seamless.

Surface Smoothness slider — under the Lofting Mode dropdown. At 0 the SDF lerps linearly between layers (visible creases at every layer boundary). Crank it to 0.5–0.7 for a smooth body; 1.0 polishes everything to a near-perfect surface but loses some sharpness on tight curvature.

Textures and surface effects don't apply in Multi-Object mode — the SDF mesh has no axisymmetric vertex layout for those passes to target. The Texture and Effects tabs show a notice when you're in this mode.

Intelligent mode

You drop in a reference image (a photograph or sketch of a vase). The silhouette is auto-traced — the foreground/background separation uses alpha for transparent PNGs and a corner-colour distance threshold for opaque images. The result becomes a Revolution profile you can then edit.

Stage	Screenshot
Image picker + sample-count slider
Traced profile + 3D result

Workflow:

1. Switch to Intelligent mode.
2. Click Pick Image and select a PNG / JPEG.
3. The thumbnail appears in the panel with the trace status (number of points sampled, silhouette aspect ratio).
4. The model updates to match the traced silhouette. Aspect ratio is honoured automatically — a tall narrow image produces a tall narrow vase.
5. Optionally tweak the Sample Count to change profile resolution (more samples = more curve fidelity).
6. Switch back to Revolution mode to fine-tune the profile manually.

Unsupported inputs surface a friendly message: "Couldn't find a silhouette — is the background plain enough?"

What to read next

• Chapter 3 — Shape Tools for the pottery wheel, dimensions overlay, and build-plate management that apply to every modelling mode.
• Chapter 4 — Surface Decoration for textures and the Pattern Graph Editor.