---
title: "Subdivision"
canonical: "https://documentation.chaos.com/space/VBLD/117638239/Subdivision"
format: markdown
---
This page provides information about the Subdivision node in the V-Ray Node Editor in Blender. Overview The Subdivision node smooths the surface of the mesh by dividing the object's triangles into sub-triangles at render time. This results in a fast smoothing effect on the object. This node needs to be connected to the  Object Output  node to be applied to the selected geometry. UI Path:  ||V-Ray Node Editor|| >  Object  >  Add  >  Geometry  >  V-Ray Subdivision Properties Geometry Generation  – Determines when the geometry is compiled during rendering. On The Fly   – The subdivision geometry goes into the dynamic memory pool during rendering, from which it can later be removed, if needed. This process is at the expense of some speed, but potentially saving on RAM.  Pre-tessellated   – The subdivision geometry is pre-compiled into an acceleration structure at the beginning of the rendering and remains there until the end of the frame. This may speed up the rendering but increase the memory usage. Preserve Map Borders – Specifies different modes for subdividing the UV boundaries from  None ,  Internal , and  All . None – UVs are always subdivided regardless of whether they are on a UV seam or not; Internal – Only preserve UVs if they are on an internal UV seam; All – Does not subdivide UVs on UV seams. Subdivide Uvs  – When enabled, the UVs of the object are subdivided at the borders. Preserve Geometry Borders  – Keeps the geometry boundaries in place while subdividing the mesh. Classic Catmull-Clark  – When this option is enabled, V-Ray uses the Classic Catmul-Clark method for subdividing the mesh instead of the hybrid one used by default. This option should be enabled only if the mesh is composed entirely of rectangular faces or it does not work. Quality Overrides Use Global Settings  – When enabled, the Quality Overrides are read from the  Globals Rollout . This option is enabled by default.  View Dependent  –  When enabled,  Edge Length  determines the maximum length of a subtriangle edge, in pixels. A value of 1.0 means that the longest edge of each subtriangle is about one pixel long when projected on the screen. When View Dependent is off, Edge Length is the maximum subtriangle edge length in world units. This options is disabled by default. Edge Length  – Determines the quality of the displacement or subdivision. Each triangle of the original mesh is subdivided into a number of subtriangles. More subtriangles means more detail in the displacement, slower rendering times and more RAM usage. Less subtriangles mean less detail, faster rendering and less RAM. The meaning of Edge Length depends on the  View Dependent  parameter. Max Subdivisions  – Controls the maximum subtriangles generated from any triangle of the original mesh. The value is in fact the square root of the maximum number of subtriangles. For example, a value of 256 means that at most 256 x 256 = 65536 subtriangles are generated for any given original triangle. It is not a good idea to keep this value very high. If you need to use higher values, it is better to tessellate the original mesh itself into smaller triangles instead.