Preparing Meshes in Rhino 8: A Practical Guide to Mesh and ShrinkWrap

Rhino 8’s official documentation describes the Mesh and ShrinkWrap commands as two ways to convert architectural models and geometry from different sources into meshes. Mesh works with NURBS surfaces, polysurfaces, and SubD objects, while ShrinkWrap can also create a mesh around mesh, point cloud, and point objects. This guide explains which methods and settings to consider in Rhino 8 before exporting, visualizing, or working with scan data.

1. Decide what the mesh is for

A mesh is a surface representation made up of vertices and polygons. Rhino meshes contain triangles and quadrilaterals. Exporting a NURBS model to another format, representing a surface with denser geometry, and converting scan data into a workable form may not call for the same mesh settings. Before running a command, decide whether your primary goal is export, visual representation, or combining different geometries into a single closed mesh.

In Rhino 8, the Mesh command creates meshes from NURBS surfaces, polysurfaces, and SubD objects. The resulting mesh is separate from the source object and can be edited. According to Rhino’s documentation, a mesh created from a solid can be seamless—that is, closed—which is important when exporting to rapid-prototyping formats such as STL. Render meshes used for viewport shading are different: they are not created as editable meshes and can be extracted with the ExtractRenderMesh command.

2. Set Mesh density to suit your needs

After starting the Mesh command, enable the preview and check the settings. The density slider in the simple controls affects the number of polygons created across the surface. Rhino 8 documentation gives Density a default value of 0.5 and an accepted range of 0 to 1. A higher value creates more polygons, so consider the intended use of the file rather than judging by visible quality alone.

Detailed controls let you choose which criteria matter for different models. For example, Maximum distance, edge to surface limits how far mesh edges can deviate from the surface. Lowering this value produces a more accurate result, but also increases the polygon count and slows processing. Maximum edge length encourages the creation of smaller edges; Rhino’s documentation describes this option as useful when polygons of roughly equal size are needed. Both settings depend on the scale of the model.

The Refine mesh option recursively subdivides faces according to the specified angle, edge length, and distance-from-surface criteria. Turning it off makes the mesh faster to generate and reduces its polygon count, at the cost of accuracy. The documentation notes that this option is disabled for extrusions and ruled surfaces. Comparing the preview and resulting mesh on the same model can help you assess the settings’ effects, rather than testing them on just one object.

3. Check seams and planes on joined surfaces

Pay attention to the Jagged seams option if mesh edges need to connect along the boundaries of surfaces in a polysurface. When enabled, surfaces are meshed separately and their edges are not stitched together. This can make processing faster and reduce the polygon count, but gaps may appear at joins and the mesh may not be closed. When the option is off, Rhino can join the mesh edges of connected surfaces to create a closed mesh.

Simple planes meshes the edges of planar surfaces and fills the interior with triangles. This method can create more polygons and slow processing, especially on planes with complex boundaries. When the option is enabled, all other settings except Jagged seams are ignored for planar surfaces; Rhino meshes these surfaces using as few polygons as possible. Use the preview to check whether the settings produce the result you want. Also note that the render mesh wireframe preview is not shown in Rendered and Raytraced display modes.

4. Consider ShrinkWrap for mixed geometry and scan data

ShrinkWrap creates a new mesh around the selected geometry. NURBS surfaces, SubD, mesh, point cloud, and point objects can be used as inputs. Uses documented by McNeel include creating a unified solid mesh from multiple objects, combining pieces of scan data into one mesh, generating a closed mesh from damaged geometry, and supporting reverse-engineering workflows.

In the command’s basic settings, Target Edge Length sets the approximate length of edges in the output mesh; the default value is calculated automatically. Offset is applied in model units: a positive value expands the mesh outward and increases the face count, while a negative value shrinks the mesh and reduces the face count. Smoothing iterations determines the strength of smoothing. A value of zero disables smoothing; as the value rises, the mesh may lose more faces and shrink further.

Polygon Optimization can be set from 0% to 100%. As the percentage increases, more faces are removed, while faces along sharp edges are preserved. However, this can also cause the output mesh’s edge lengths to vary more. Inflate Vertices and Points creates a mesh from every vertex or point in a mesh or point cloud. It is on by default for points and point clouds, and off for other object types. Intersecting inflated mesh pieces can merge into a larger mesh. Fill holes in input objects, which closes holes in the input before processing, is useful in many cases, but can also close small openings that are intentional parts of the design.

What does this mean for architectural and visualization workflows?

The Mesh command gives architects and visualization teams a way to control settings when preparing NURBS- or SubD-based models for export or another stage that requires a mesh. ShrinkWrap is an option for teams working with point cloud fragments or geometry that is difficult to repair. In both cases, bear in mind that a more detailed mesh can increase processing time and polygon costs, while lower density can reduce the accuracy of the surface representation.

Common mistakes

• Assuming higher mesh density always means a better result: the polygon count also rises with density.

• Expecting a closed mesh with Jagged seams enabled: this setting meshes surfaces separately and can leave gaps between them.

• Treating positive and negative Offset values in ShrinkWrap as if they work in the same direction: positive expands the mesh; negative shrinks it.

• Assuming that automatically filling holes is suitable for every model: small openings left intentionally as part of the design may be closed.

• Looking for the render mesh wireframe preview in Rendered or Raytraced mode: the documentation says the preview is not shown in these modes.