Rhino 8 QuadRemesh Guide: Clean Quad Topology and a SubD Workflow

Rhino 8’s QuadRemesh command can be used to generate a quad mesh with editable topology from existing surface, mesh, or SubD geometry. The workflow described in McNeel’s Rhino 8 documentation involves setting a target edge length or face count, adjusting the level of detail, and, if needed, converting the result to SubD. The command is located under SubD / Mesh > Edit Tools > Quad Remesh. This guide focuses on understanding what each option changes when preparing models for architectural modeling and visualization.

1. Choose the input geometry and target

Before starting QuadRemesh, select the surface, mesh, or SubD object to convert. The command has two primary target options: Target Edge Length and Target Quad Count. The first sets the approximate edge length of the resulting mesh and depends on the model’s scale. The same target value does not guarantee the same face count for objects of different sizes; as the object gets larger, the resulting face count also increases. If the target edge length is set too large relative to the object, remeshing may fail.

Target Quad Count gives the algorithm an approximate face-count target and works independently of scale. The aim is to produce a similar number of faces even when the same input geometry is at different scales. So, use the first option when a specific edge length is the priority, and consider the second when you want to roughly limit the face count. Neither setting guarantees an exact result; each is a target the algorithm attempts to reach.

2. Control quad density and boundaries

Adaptive Size changes the balance between face size distribution and detail preservation. At 0, the goal is to produce fewer, more consistently sized quads. As the value increases, smaller faces may form, especially in areas with high curvature. Values above 30 reduce control over achieving a low quad count, while 100 aims to preserve more detail. Rather than treating a single value as the default for every model, check the preview against the desired surface characteristics.

If the input is a polysurface or extrusion, the Use Surface Edges options determine how polysurface boundaries are handled in the mesh. Off ignores these boundaries. Smart preserves them except where the algorithm considers them insignificant. Strict preserves all polysurface boundaries. Consider Smart or Strict when it is important to retain the boundaries between joined surface pieces in the topology; Off can be used when boundaries are not significant to the result.

3. Use symmetry and guide curves

If the model is symmetrical, you can enable symmetric remeshing along the X, Y, or Z axes at the center planes of the object’s bounding box; multiple axes can be enabled at once. This setting is only meaningful when the object is truly symmetrical and the correct symmetry plane is selected. Enabling it on an asymmetrical model may not produce the intended geometry.

Guide Curves can be used to direct quad flow or define more detail in specific areas. For the curves to have an effect, they must be projected onto the input object. There are three behaviors under Curve Influence:

1. Approximate slightly influences the overall flow direction of the quads.

2. Create Edge Ring attempts to position edges that cross the guide curves perpendicular to them; the effect is stronger, but the edge rings may not follow the curves exactly.

3. Create Edge Loop places edge loops along the guide curves and has the strongest effect.

When working on curved façade elements or surfaces with a specific flow direction, project the guide curves first, then compare these influence options. Check the result using Preview; the preview updates when settings change, and the average edge length and face count are shown at the bottom of the dialog. If a dense wireframe input object makes the preview hard to see, enable Hide Input Objects.

4. Hard edges and SubD conversion

Detect Hard Edges adds a hard-edge loop when the angle difference between adjacent faces is greater than 30 degrees. If the geometry comes from an irregular mesh or unwanted hard edges appear, consider turning this option off. QuadRemesh’s Convert to SubD option converts the generated quad mesh into a smoothed SubD surface. Hard edges are preserved where possible; the SubD Creases and SubD corners options carry crease edges and corner points from the mesh into SubD features, respectively.

When Interpolate SubD is enabled, the quad mesh vertices are used as SubD points, and the SubD stays on top of the quad mesh. When disabled, the quad mesh vertices are treated as control polygon points; depending on the input shape, the result may be slightly smaller or larger. Rhino’s documentation relates this option to UseMesh in the ToSubD command.

A separate ToSubD command can also convert meshes, extrusions, surfaces, and polysurfaces. With ControlPoints in the UseMesh setting, mesh vertices become SubD control points; with Location, they become SubD points. DeleteInput determines whether the original geometry is deleted; when experimenting, check the deletion behavior if you want to keep the source object. Surface trims may be lost during conversion, polysurface faces may be converted individually, and faces associated with non-manifold mesh edges may be removed.

Use in architectural and visualization workflows

QuadRemesh is documented as a way to create manageable quad meshes for rendering, animation, reverse engineering, CFD, and FEA. In architectural visualization, this offers a way to turn a dense or irregular input into a more controlled face structure, with the option to smooth it using SubD. However, the result depends on the target and the geometry; the same settings should not be assumed to work for every model.

For an office workflow, first decide whether the target face count or edge length matters more. Then inspect face distribution and boundaries in the preview; project guide curves if needed. Before moving to SubD, account for the possibility that trims may be lost and non-manifold areas may be removed during conversion. Since the documentation does not specify special hardware requirements or licensing conditions, no further assumptions should be made on these points.

Common mistakes

• Treating the scale-independent Target Quad Count and scale-dependent Target Edge Length settings as if they work the same way.

• Entering a target edge length that is too large for the object and overlooking the possibility that remeshing may fail.

• Expecting guide curves to have an effect without projecting them onto the input object.

• Enabling the symmetry option without checking that the object is symmetrical and that the correct center plane is selected.

• Failing to check for possible trim loss, individual conversion of polysurface faces, or removal of non-manifold areas before using ToSubD.