Skip to content

Introduction to node-based Build Assembly#

Clarisse offers a high level node-based assembly workflow which is a flow of atomic nodes describing operations specifically aimed at sequence or shot lighting. The resulting graph, also called build assembly or build, generally starts by bringing and merging assets together, continues with lookdev association and lighting and ends up by render layers definition, local edits and pre-comp.


All features described in this section and sub-sections are only available in Clarisse BUiLDER

graph TB

Schematic view of the node-based scene assembly workflow in Clarisse BUiLDER

New to Clarisse?

This section of the documentation is specially focused to scene assembly and lighting so if you are new to Clarisse, don’t worry: we’ve specifically created this type of short notes to quickly cover key Clarisse concepts targeted specially for you. These notes will also provide links to sections covering these concepts in greater details, if you wish to learn more about them. However, Clarisse being a powerful 3D DCC, we still recommend you to have a look at its feature set to totally leverage from the full power of Clarisse! If you are already pretty familiar with Clarisse key concepts such as Contexts, Groups, Rules, Material and Attribute Editor, Explorer and the Image View, you are going to pick up the specific feature set of the build assembly workflow very quickly!

Build Assembly Workflow#

This workflow is based on a set of nodes that are specialized to perform certain types of operations that can be grouped into logical categories:

  • Scene Assembly nodes to generate and manage scene contents.
  • Selection nodes to procedurally select nodes and inject them in your builds.
  • Rendering nodes to setup AOVs, render layer definition, setup render scenes...
  • Compositing nodes to perform compositing operations on images such as color correction, image merge or blur.
  • Build Layout nodes to arrange the layout of the build assembly through visual notes, backdrops or dots.
  • Process nodes to run processes such as render to disk for example.

The Build Assembly#

The build assembly itself is a flow of successive operations where you explicitly use nodes to perform specific operations at one point of your build. This workflow is very powerful as items and scenes can have multiple representations throughout the build assembly. Depending on how you design your build, it is possible to let the artist work on multiple shots or even sequences at the same time!

Memory Optimization#

Of course this wouldn’t work if all assets and their multiple representations actually lived in memory. To make this work, Clarisse evaluation engine goes a step further than before. Unlike previously where Clarisse was deduplicating data and loading geometry on demand, the engine is now able to load scenes partially on an as-needed basis so that it only uses what’s necessary while at the same time automatically deduplicating identical scene elements in memory.

Workflow Benefits#

One of the great benefits resulting from this workflow is builds reusability. Builds can be setup in such way that they can become asset-independent templates to be used as starting point for new shots and sequences. Another great benefit is that this workflow tremendously simplifies teamwork. Indeed, thanks to the bird’s eye view of the build assembly, you quickly understand the build construction logic and detect potential issues even if you didn’t create it in the first place.

Last but not least, builds are extremely quick to load even if the final render scene complexity is super high.

Indeed, the number of scene assembly nodes involved to the build assembly is completely uncorrelated to asset complexity. When you bring a very complex asset to your build, only the high-level node description (describing mainly the path to the asset) is stored. As a result, the build description file is always very small even if the assets you import in your build are very complex.

Furthermore, thanks to the build assembly engine lazy loading, when you don’t display the content of a node (in the 3D View or in the Explorer for example), nothing gets loaded. Clarisse gives you the flexibility to choose to work completely blindly on your builds using assets that are even offline if you choose to.


The good practice for creating great builds is to make them with a relatively small number of nodes. The idea is to keep the graph as clear and manageable as possible so that you don’t end up with thousands of nodes even if it is possible. Just because you can, doesn’t mean you should! :)

The high level build assembly in Clarisse is achieved through a dedicated editor called the Build View. Node parameters and attributes are modified through the Attribute Editor and scene states are visualized using the Explorer, Browser or the Hierarchy View.

New to Clarisse?

In Clarisse, every single item you create is an object in the conceptual sense of it. They can all be represented as a node. Every item/object defines attributes (called parameters in other packages) which are the entry point to manipulate the item. To translate a camera for example, you would select it and change its Translate attribute in the Attribute Editor. For more information about items in Clarisse, please refer to Managing Items