Latest news
Using Gaussian Splat in V-Ray: a step-by-step guide for MayaNatural Vegetation Distribution with Chaos Scatter: 3ds Max and Vantage GuideCorona 16 Gaussian Splatting guide: lighting real spaces and Frame Buffer tipsWorld Architecture Festival 2026: Day One Winners Announced, Foster + Partners and KPF ShineChaos Corona 16 announced: Gaussian Splat relighting and parallax interior support arriveTurning Architectural Renders into Video with Wan 2.2: A Step-by-Step GuideShigeru Ban unveils cross-shaped timber-roofed church design for RomeChaos Cosmos Browser guide: download ready-made 3D models for V-Ray and CoronaSetting Up Unreal Engine Multi-User Editing: a guide to team scene collaborationWaugh Thistleton, Margate'deki Westbrook Pavyonu'nu Tracey Emin için dönüştürecek
Tips & GuidesRhinoGrasshopperNVIDIA Omniverse

Grasshopper Data Trees: A Guide to Matching and Clean Output

In Grasshopper, component results can vary even when the same items are stored in different branch and path arrangements. This guide explains how to inspect data trees, check matching, and organize structures in Rhino and Omniverse workflows.

Contemporary pavilion façade with repeating modules that evoke data branchesAI image
Representative image, generated with AI.Image: 3dsınıfı / FCA AI

In brief

  1. In Grasshopper, a single item, a list, and more complex structures are different forms of data trees.
  2. Parameter Viewer and Panel can be used to inspect branches, paths, and item counts.
  3. Flatten gathers all items into a single list, while Graft moves each item to a separate branch.
  4. For export to Omniverse, geometry must be connected to the OmniGrasshopper component through OmniLayer; the geometry input may need to be flattened.

In Grasshopper, a component’s correct output depends not only on the values being used, but also on how those values are arranged in the data tree. Rhino’s developer documentation covers tree structure and matching; the Grasshopper 2 documentation describes data modifiers in parameters; and the NVIDIA Omniverse Connector documentation explains how to publish Grasshopper data. The workflow below can be used to check data structures in architectural modeling and visualization projects.

Version scope: The K1 guide to data trees was updated on 15 July 2024 and does not specify a particular Grasshopper version. The K2 document is for Grasshopper 2. K3 describes using Grasshopper, which is included with Rhino, through the Omniverse Connector; it states that Grasshopper began shipping with Rhino 6. Modifier names and behavior should therefore be checked for the version in use.

1. First, identify what the tree represents

Grasshopper stores both a single item and a list of items as data trees. A single item can be thought of as one branch containing one element; a list is multiple elements within one branch. When the same values are distributed across multiple branches, the structure changes, which can affect how subsequent components match items.

A branch’s path is shown in curly braces, while an item’s index within the branch is shown in square brackets. For example, in {0;3;0}[1], the first part indicates the branch path, and [1] indicates the position of the item within that branch. In a model, keeping surfaces, curves, or point groups in separate branches can determine how those groups match in subsequent operations. So before looking at the values, define the goal of the structure: should the items be processed in the same list, or remain as separate groups?

2. Inspect the structure before adding a component

  1. Find the parameter or component output where the data is connected.

  2. Use Parameter Viewer to check branch paths and the number of items in each branch. The visual display helps compare how branches are distributed, while the text display helps read the paths.

  3. Use Panel to view the values. Check that the expected items are present and confirm their order.

  4. If needed, use TStats to extract paths, the number of items per branch, and the total number of branches.

This check helps you spot cases where two inputs contain the same number of items but are divided among different branches. If a component’s output is connected to another component, check the branch structure as well as the values. The core principle stated in K1 is to pay attention to how the data structure changes between operations; even a small structural difference can change the result.

3. Create the desired structure and verify matching

Components such as Grid and DivideSrf in Grasshopper can produce data trees; Entwine can combine lists into a tree. For example, to turn the X, Y, and Z coordinates of points into separate lists, you can deconstruct the points with Deconstruct or pDecon and combine the lists with Entwine.

When trees are matched, the last item in a shorter branch or list may be repeated to complete the match. A tree with fewer branches can also be expanded by repeating its last branch. This behavior matters when applying a single number to an entire tree or connecting lists of different lengths to a component. If the result is unexpected, compare the number of branches and the number of items per branch for each input before changing the geometry itself. Then check the matching result again with a Panel or Parameter Viewer.

4. Choose Flatten, Graft, and other modifiers for the task

In Grasshopper 2, data modifiers are available from the parameter menu. This lets you apply some operations without setting them up as separate components. The standard modifiers in the K2 document can be considered in four groups:

  • Structure: Flatten gathers items from all branches into a single list. Graft moves each item to its own branch containing one item. Simplify removes numbers shared across all paths; Renumber rearranges paths in the order they appear and removes deeper structure.

  • Sorting: Sort orders items according to the sorting rules for their type; Randomise shuffles the order; Reverse reverses it. Sorting may not be meaningful for every data type.

  • Cleanup: Null, Invalid, and Empty remove null items, invalid items, and empty branches, respectively. Deleting items from a branch can leave it empty; removing an empty branch can also create gaps in the paths.

  • Value operations: Meta and Expression are used to assign metadata to data or process each item individually. In an expression, the current value is represented by x.

Don’t enable a modifier simply to make the output look simpler: for example, Flatten removes the grouping relationship between separate branches, while Graft separates items from one another. In Grasshopper 2, the order of operations can affect the result when multiple modifiers are active at once. According to the documentation, expressions are applied first; null and invalid items are removed before empty branches. Also, if Sort is enabled, Randomise is skipped; Flatten can also make some structure modifiers ineffective when they are used together. So inspect the final output again.

5. Verify output in Rhino and Omniverse workflows

According to the Omniverse Connector documentation, geometry is not fed directly into the OmniGrasshopper component; it is connected through OmniLayer or OmniGroup, which converts geometry into a layer group. An OmniGrasshopper component takes OmniLayer groups or lights to be published, and only one component of this type can be used per canvas.

In the OmniGroup component, the geometry input must be flattened. The documentation recommends flattening the geometry input if OmniGrasshopper turns red and reports an error stating that the same key has already been added. To do this, right-click the Geometry input and select Flatten. Any geometry that is not connected to OmniGrasshopper through an OmniLayer node will not appear in the final output.

Before publishing, connect to a Nucleus server using the Server Sign In button in the OmniBrowser component; then use the Publish button in the OmniGrasshopper component. The documentation states that the model updates in real time when Live Sync is enabled. However, K3 reports that the USD Project function in the as project option is deprecated and will be removed in future versions; check the status of the project publishing option when setting up a new workflow.

What changes in architecture and visualization workflows?

When using Grasshopper to produce repeating façade elements, point arrays, or groups of different geometries, setting up the data tree correctly makes it more predictable which items components will match. In parametric variations, operations such as Graft and Flatten directly affect whether groups are preserved or gathered into a single list. For teams exporting to Omniverse, it is also important to connect geometry through OmniLayer and check whether flattening is required.

The documentation does not specify hardware requirements or licensing costs. Before deploying the workflow in an office, test the Grasshopper version in use, the Connector’s publishing options, and the Nucleus connection alongside the existing project setup. In particular, if an older workflow relies on the USD Project option, do not overlook the deprecation notice in the documentation.

Common mistakes

  • Checking only item values and overlooking branch paths and item counts per branch.

  • Using Flatten to gather all items into a single list when groups need to remain separate.

  • Failing to account for each item moving to its own branch after Graft.

  • Not checking how empty branches or changed paths affect the result after using cleanup modifiers.

  • Failing to connect geometry for Omniverse export through OmniLayer, or trying to publish OmniGroup geometry without flattening it.

Sources

3 sources
D
developer.rhino3d.comdeveloper.rhino3d.com/guides/grasshopper/gh-algorithms-and-data-structures/advanced-data-structures
Summary
R
rhino3d.comrhino3d.com/docs/grasshopper2/topics/DataModifiers.html
Summary
D
docs.omniverse.nvidia.comdocs.omniverse.nvidia.com/connect/latest/rhino/manual/grasshopper.html
Summary

Source texts are not republished; short quotes are marked, everything else is our own summary and commentary.

3dsınıfı’s take
3dEditor’s assessment

For architecture and visualization offices in Turkey, being able to read a data tree makes it easier to find the cause of a matching issue instead of changing connections through trial and error. The result is a more controlled modeling workflow, especially for parametric façades and repeating geometry.

For Omniverse export, checking flattening and the OmniLayer connection from the outset can reduce publishing errors. Since the documentation provides no information about hardware or licensing costs, teams should test the workflow with their own versions and server setup. The deprecation notice for the USD Project function should also be considered when setting up new project workflows.

Frequently asked questions

How do you view a Grasshopper data tree?

`Parameter Viewer` shows paths and item counts per branch; `Panel` can be used to inspect the data. `TStats` can extract paths, the number of branches, and item counts per branch.

What is the difference between Flatten and Graft in Grasshopper?

`Flatten` gathers items from all branches into a single list. `Graft` moves each item to its own separate branch containing one item.

How do you publish a Grasshopper model to Omniverse?

Connect geometry to the OmniGrasshopper component through OmniLayer or OmniGroup, then sign in to a Nucleus server and use `Publish`. The Connector documentation also recommends flattening the geometry input.

Comments and the forum are in Turkish.Join the discussion
+

Related news