Houdini 20.0 Reference Windows

Display Options (3D viewer)

On this page

When the Apply operations to all split views checkbox at the bottom of the window is on (the default), all changes in this window apply to all split viewports.

When using a split view, the easiest way to make viewport-specific changes is to use the viewport options menu in the top-right corner of the viewport. The menu contains most viewport-related settings you might want to change, such as shading mode.

However, if you want to change other display options in only one viewport, move the pointer over the one of the split views and press N to make it “active” (the view’s menus are drawn in yellow). Then open the display options window and turn off Apply operations to all split views before making changes.

Markers tab

Set display options for

This menu controls what category of geometry the options below apply to.

Scene Geometry

Display options for unselected objects at the object level look like.

Selected Scene Geometry

Display options for selected objects at the object level. By default this is the same as “Scene geometry”.

Ghost Scene Geometry

When you are at the geometry level (inside a Geometry object), you can optionally show other objects “ghosted” (semi-transparent). This shows display options for these ghosted objects.

Display Model Geometry

At the geometry level (inside a Geometry object), display options for the output of the node with the display flag.

Current Model Geometry

Display options for the selected geometry node’s output. At the geometry level, when you select a node that doesn’t have the display flag, the view shows the output of the selected node (by default, as wireframe) along with the output of the network. This lets you edit the node and see the effects in the view.

Template Model Geometry

Display options for “templated” geometry. You can set the template flag on a geometry node to keep the node’s output visible at the geometry level alongside the display geometry.


Allows you to use the marker settings from another category for the current category. To manually set options for the current category, choose “Unique settings”.

Marker type visibility menu

Most marker types have a small visibility icon menu to the right of the label that allows you specify different levels of visibility for the marker type. This can be useful for reducing clutter in the viewport. Click the icon to show a menu of visibility levels:

Always visible

Show the markers on all visible components.

Only selected components

Only show the markers on selected components (points, primitives, vertices).

Area around pointer

Only show the markers near the mouse pointer.

Under pointer only

Only show the marker on the component (point, primitive, vertex) under the mouse pointer.



Draw small dots showing the point locations.


Draw the point number next to the point location. Point numbers start at 0.


Draw the point normal direction as a small line from the point location. You can change the line length using the Normal scale option on the Guides tab.

If a point does not have a normal attribute, Houdini draws a generated normal (as used by the viewport and Mantra), in a dimmer color.

UV coords

Draw each point’s UV texture coordinates next to the point location.

XYZ coords

Draw each point’s 3D coordinates next to the point location. Drawing three decimal numbers for each point can clutter the screen quickly, so you might want to use the visibility menu for this option.


Draw a trail line from the current point position back to where it was in the previous frame. This information is taken from the point’s v (velocity) vector attribute. You can scale the trail length using the Vector scale option on the Guides tab.


Draw the number of overlapping points when multiple points overlap on the screen (for example, if three points overlap, draws a 3).

This is in screen space, so it’s most useful when the geometry you're looking at occupies a lot of the viewer. If the geometry is small/far away, you’ll get a lot of useless overlap readings.

Don’t turn this on for all geometry in the scene. It works best on the display geometry, or only the selected geometry.


“Primitives” are polygon faces, curves, and NURBS surfaces (as well as more exotic types such as metaballs, primitive shapes, and crowd agents).


Draw lines representing the limits of NURBS, Bézier surfaces, curves, and metaballs.

It can be useful to display hulls in place of the actual geometry when extremely large numbers of these types of geometry fill the screen, to reduce clutter and speed up the display.


Draw the primitive number next to the primitive. Primitive numbers start at 0.


Draw the primitive normal direction as a small line from the primitive’s location (for primitives that have normals). You can change the line length using the Normal scale option on the Guides tab.


Draw edit points on NURBS and Bézier curves/surfaces.

Profile curves

Draw profile (trim) curves on NURBS surfaces.

Profile numbers

Draw the profile curve number next to the start of profile (trim) curves. Profile curve numbers start at 0.

Tint Backfaces

Tint backfacing polygons so that they are easily distinguishable from frontfacing polygons.



Draw small boxes showing the vertex locations. Draws the vertices slightly offset inward on the polygon face so you can tell the vertices apart from each other and the point. Unselected vertices are hollow, and selected vertices are filled.


Draw the vertex number next to the vertex location. Draws numbers for vertices of different polygons in slightly different shades, to make them easier to tell apart.


Draw the vertex normal direction (if the vertex has a normal attribute) as a small line from the vertex marker’s location. Draws normals for vertices of different polygons in slightly different shades, to make them easier to tell apart. You can change the line length using the Normal scale option on the Guides tab.

UV coords

Draw each vertex’s UV texture coordinates next to the vertex marker location.

UV Backfaces

In UV viewports, change the fill color of back-facing polygons to make it easy to tell them apart from front-facing polygons.

UV Overlap

In UV viewports, change the fill color of regions where multiple primitives overlap.


The options in this column control how Houdini draws geometry in the current category.


The shading mode for surfaces in the current category. For example, you can have scene geometry drawn as smooth shaded, and templated geometry drawn as wireframe.

Lock Shading Mode

Ignore the viewport shading mode control for surfaces in the current category.

For example, this is on by default for templated geometry, so (for example) changing the viewport shading mode between shaded and wireframe does not change the look of templated geometry.


Draw curves/surfaces with a grayed-out, less prominent look.


Ignore any color (Cd attribute) or any texture on curves/surfaces. Uses the default wireframe colors and the default material. This generally used for “templated” geometry. You can change the defaults on the Material tab.

Draw as X-Ray

Show the wireframe of any geometry hidden behind other surfaces.

Allow Lighting

Allow lighting to affect the look of geometry in the current category. This overrides the viewport’s lighting setting.


3D Boundaries

Show boundaries of meshes (non-shared edges).


Don’t highlight open edges.

All Views

Highlight open edges in all viewport types.

3D View Only

Draw open edges in 3D viewports.

UV View Only

Draw open edges in UV viewports.

UV Boundaries

Show boundaries of UV texture coordinates on meshes (UV island edges).


Don’t highlight edges of UV islands.

All Views

Highlight edges of UV islands in all viewport types.

3D View Only

Draw edges of UV islands in 3D viewports.

UV View Only

Draw edges of UV islands in UV viewports.


Enabling either 3D or UV boundaries requires Houdini to do extra processing of the geometry in the scene. It should be fast to tumble, but can slow animation playback.

UV Display

UV Attribute

The name of the geometry attribute used for UV boundaries and UV map visualization.

Show UV Map

When enabled, use this texture map for visualizing UVs. This will only affect geometry with a UV attribute and no material assignment (though turning materials off will remove the assignment, and allow this texture visualization to be seen).

UV Map Scale

Scale the UV coordinates by this value when visualizing UVs with a UV Map. This does not affect textured materials.

Auto Detect UV Attribute Type

Automatically detects the attribute type in a UV texture editor viewport.

UV Viewport Attribute Type

The attribute type for UV viewports, when Auto Detect Attribute Type is off.

Render Tab

The render tab only appears when viewing LOPs. You can find the current renderer’s settings here. Each Hydra renderer has its own list of settings.

Crop Render images to Camera Aspect

When looking through a camera in LOPs, render delegates can render to the full viewport area or only the area within the camera frustum. If enabled, a fixed resolution applied to the image so that panning and zooming in 2D does not change it or restart the render.

Revert to Renderer Defaults

Reset all the renderer settings above to their default values.

Image Resolution

Specify what determines the image resolution when required by Crop Render Images to Camera Aspect or the Image viewport. Viewport Size

Resolution of the camera frustum area within the viewport.

Render Settings Prim

Take the resolution from the current Render Settings Primitive. If there isn’t a current primitive, use the Viewport Size instead.

Fixed Size

Specify a resolution in the next parameter. This can only be used in the Image viewport.

The resolution from any of these sources can then be scaled down by fractions, from full resolution to 1/16.

The Image viewport displays the renderer output in much the same way as MPlay or the Render View - as a 2D image with specific resolution. You can zoom in to inspect individual pixels and select an area of interest to speed up rendering (LMB drag in the view state). 3D navigation is not available in this viewport mode, so viewing through cameras is recommended. All 3D handles, guides, grids, and selections are also not displayed for a clear view of the image.

Guides Tab

Floating gnomon

Draw a small gnomon in the bottom left corner of the view, showing the current orientation of the world axes.

Origin gnomon

Draw a gnomon at the world-space origin (0,0,0). (See also how to display an object’s origin/pivot.)

Particle gnomon

Draw a small gnomon at the origin of each particle system. If the particle system’s Display Particle Axes option is on, and the particles have a center of mass attribute (com), this will show the center of mass.

View pivot

Draw an indicator at the world-space location about which the view rotates when tumbling.

Group and attribute list

Show the group list in the upper right corner of the viewer.

Video safe areas

Draw rectangles showing the video “safe area” for picture and titles.

Field guides

Draw a traditional cell-animation grid overlay on top of the view. This can make it easier to align elements by eye and find points of reference between frames.

Camera view mask

Draw the view mask and overlay of the current camera.

The parameters on the camera to control the display mask are set to a 1.85 aspect ratio and hidden by default. To show them, use Edit rendering parameters in the parameter editor’s gear menu, and add the For rendering ▸ Viewport ▸ View Mask properties.

XZ, XY, and YZ Reference Planes

Draw a 20×20 reference grid centered at the origin in the XZ, XY, and/or YZ plane.

Node guides

Draw “guide geometry” created by some nodes to indicate invisible information (for example, wind direction).

Node handles

Show interaction handles created some nodes (for example, camera handles).

Follow selection mask

Automatically show components for selection. For example, when this is on and you go to select points, the viewport will automatically show point markers.

IK Critical zone

Draw the IK critical zone guides for bones.

Object names

Draw the node name of each object (for example, lamp) at its local origin.

Object paths

Draw the full path of each object (for example, /obj/subnet1/lamp) at its local origin.

Displayed nodes

Draw the display geometry of objects.

Current geometry

Draw the geometry contained in the selected geometry node (when it’s different from the display node).

Template Geometry

Draw templated geometry.

Selectable templates

Draw nodes with the selectable template flag on.

Additional information

Draw Time

Draw timing information (frames per second (FPS), draw time, and elapsed time), in the corner of the viewer. This can be useful for testing scene performance. You should use the FPS as a reference for animation speed, and the elapsed time when comparing playback times.

The FPS display is limited to a maximum of 120, to avoid showing extremely high and misleading values.

Render Time

Display render time and progress as reported by the current renderer. Only supported in LOPs.

Render Stats

Display render statistics reported by the current renderer, as specified in usd_renderers.py. Only supported in LOPs.

Geometry information

Draw information about the current scene/model (for example, total number of polygons and number of selected polygons) in the bottom-right corner of the viewer.


Object Selection

Draw selected objects with a highlighted wireframe or outline. If you turn this off, selected objects will look exactly the same as unselected objects. This only applies to object-level selection.

Wire over Shaded

Draw a colored wireframe over the selected object.


Draw an outline around the selected object.

Fill Selections

Fill selected components with the highlight color. For example, selected faces will be filled with the selection color. When this is off, Houdini only highlights the wireframe, allowing you to see the shading/texture of selected faces. This setting affects all wire-over modes: Wireframe Ghost, Hidden Line Invisible, Hidden Line Ghost, Flat Wire Shaded, and Smooth Wire Shaded. Pure Wireframe is not affected by this setting as it does not display faces, nor is Flat Shaded or Smooth Shaded, both of which always fill the face. This is only available when viewing a SOP.

Closure Selections

Draw the primitive to which the selected components “belong” in a secondary highlight color. This may be useful for NURBS and Bézier surfaces, where it might not be clear which surface a control point out in space belongs to. This is only available when viewing a SOP.


Do not highlighting the primitives to which the selected components belong.

Hull Geometry Only

Only highlight NURBS and Bézier surfaces with selected control points/hulls. This is the default.


Highlight primitives to which the selected components belong. For example, selecting a polygonal point will draw the faces sharing that point using a secondary highlight.

Display style

Indicate selected LOP primitives by shading them with a solid fill or drawing an outline around them. This is only available when viewing LOPs.

Guide sizes

Auto DPI Scale

Automatically scale all guide sizes according to the monitor’s reported DPI. You can turn this off to manually scale all guide sizes (this does not affect the Guide font size).

Scale Normal

Scales the length of normal markers.

Scale Vectors

Scales the length of vector direction markers, such as velocity trails.

Point marker size

Scales the size of location “dot” markers, such as points.

Origin gnomon size

Scales the size of the large gnomon drawn at the world origin. You can turn it off completely using the Origin gnomon checkbox on this tab.

Guide font

The relative size of marker text drawn in the view, such as point numbers.

Handle highlight

Controls the amount of highlighting when the mouse pointer is over a handle. “Off” does no highlighting. “Small” brightens a handle when the mouse is over it. “Normal” draws a prominent glow around a handle when the mouse is over it.

Offset vertex markers

Vertex markers and normals are inset into the polygon so they can be selected independently when part of a seamless mesh. When disabled, the markers and normals are placed at the actual vertex, potentially overlapping.

Visualize tab

This tab lets you arrange and save your common visualizers. The top section of this dialog shows your common visualizers and in the bottom you can see your scene visualizers. You can add or delete visualizers in each section. You can also drag-and-drop visualizers between the two sections or rearrange them within each section. Clicking Save As Default saves the arrangement of the common visualizers as your default common visualizers which will be available in all scene file.

Although this dialog allows you to add, delete, activate, and deactivate common and scene visualizers, it is mostly used to review and save your list of common visualizers. In particular, it does not display your Per-Node visualizers. The visualizer menu, accessible via RMB over the icon on the right side of the viewport, is the recommended way to access all visualizers relevant to a scene.

Geometry tab


Level of Detail

Increases or decreases the display resolution of metaballs, NURBS, and Bézier surfaces.

Polygon Convexing

Fixes concave polygons by tessellating them to triangles so they appear the correct shape. There are two options for determining when to redo the convexing:


Only redo convexing on full topology changes. Ignore changes to point position (P).


Redo convexing on full topology and point attribute changes.

Subdivision Polygon Limit

The maximum polygon count that subdivision can produce when subdivision is enabled for that object. If exceeded, try the next subdivision level down until the resulting polygon count is acceptable, which may be no subdivision at all. This is expressed in millions of polygons. Increasing this limit may cause very long updates for subdivision objects and long draw times.


Volume Quality

Controls the display quality of volumes in the viewer.

Very Low

Draw volumes as parallel slices along one axis. This is the fastest option but produces a visual pop as the volume rotates in the view. Overlapping volumes will produce visual artifacts.


Draw volumes as slices parallel to the viewport. This is the fastest of the view-aligned, useful for working interactively with dozens of volumes. Overlapping volumes will render correctly.


Draw volumes as slices parallel to the viewport, with more tightly-spaced slices than the “Low” option. Balances quality and performance.


Draw volumes as slices parallel to the viewport, with even more tightly-spaced slices than the “Normal” option. Slowest but best quality option. Adds some random variation to the volume sampling to break up the slices and 3D texture sampling.

Enabling HDR Rendering will remove any banding artifacts from volumes.

Volume Filtering

Controls how the viewport interpolates volume values when drawing fog volumes.


The viewport exclusively uses trilinear interpolation. This is the fastest to draw, but low resolution volumes may appear blocky in the viewport.


Higher-order inteprolation will be used when not interacting with the viewport. This mode falls back to trilinear when tumbling, moving handles, or playing the timeline.


The viewport exclusively uses higher-order b-spline interpolation. This is the slowest, highest quality option.

Ambient Occlusion

Intensity of self-shadowing for fog volumes from ambient light sources. This default is overridden by the volvis_ambientshadows detail attribute. Note that the default headlight is treated as an ambient source by fog volumes.


Wire Width

The thickness (in pixels) of wireframe lines. The range of available values depends on the OpenGL driver.

Wire Blend

Opacity of the wireframe in “Shaded wireframe” mode, from 0 (transparent) to 1 (opaque). You may want to turn this down to keep surface color/texture visible underneath a dense wireframe.

Interior Wire Alpha

The visibility of “interior” wireframes (such as inside tetrahedrons), from 0 (not visible) to 1 (drawn the same as exterior wireframe).

Shade Open Curves

Apply lighting to curves. This is useful when using curves to represent hair.

Select wireframe surfaces as if solid

Lets you select surfaces in wireframe mode by clicking where the surface would be if it was drawn.

Wire over packed geometry

Show polygon outlines on meshes in packed geometry when drawing a wire-over-shaded mode. Disabling this hides the outlines on packed geometry, making meshes appear as a single shape, which packed geometry technically is.


Display particles

How Houdini draws particles and disconnected points.


Draw points as uniform dots, controlled by Point Size (in pixels). In this mode, close and far points are drawn at the same size.


Draw points as single pixels. This may be useful for very dense particle simulations.


Draw particles as streaks. This only affects particles (disconnected points are drawn as dots).


Draw particles as filled circles, with the radius controlled by Disc size (in world space units). In this mode, particles are drawn as actual geometry, so closer particles appear bigger than far particles. This only affects particles (disconnected points are drawn as dots).

Lit Spheres

Draw particles as spheres which are lit by the lights in the scene and respect material properties.

Display Sprites

If particles have sprite attributes (see the Sprite node), draw the sprite image at the particle location.

Point Size

The size (in pixels) of particle and unconnected points, when Display particles is “Point”.

Disc Size

The size (in Houdini units), of the filled circles, when Display particles is “Disc”.

Orient Discs/Sprites to N

When Display particles is “Disc” or “Sprite”, this orients the discs with the particle’s normal. If the particle doesn’t have a normal, it uses the particle’s velocity.

Sprite Texture Limit

The minimum and maximum display size of a sprite texture.


Point Instancing

Use OpenGL point instancing to replace the points of Instance objects (and any other objects with the proper instancing properties) with the instanced geometry in the viewport. See instancing for more information.

To increase viewport performance, you can lower the percentage of the total number of points that are replaced with referenced geometry. For example, setting it to 50 will only replace half the points with the instanced geometry. The other half will instance the Stand-in geometry.

Stand-In Geometry

What to draw for points that can’t be replaced with the referenced geometry (when the Point Instancing percentage is less than 100, or the number of points to instance exceeds the Instancing Limit). “Display off” draws nothing, “Location marker” draws a dot, and “Bounding box” draws the bounding box of the referenced geometry.

Normals & Tangents

Auto Generate Normals

The type of normal to generate when no normals exist on the geometry.

Point Normals

Normals are generated on points, which is fast but cannot visualize cusps on sharp edges (such as boxes).

Vertex Normals

Normals are generated on vertices which can visualize sharp edges. The surface will be smoothly shaded as long as the angle between two polygons is less than the cusp angle (below). Vertex normals are slower to generate and display than point normals.

Cusp Angle

The angle, in degrees, at which the surface stops smoothly shading and instead creates a cusp. The angle is measured between neighboring polygons. This has no effect if Point Normals are auto-generated.

Vertex Normal Limit

The upper limit, in millions of polygons, where normal auto-generation will switch from Vertex Normals to Point Normals. Large models often take too long to generate vertex normals and display speed is affected as well. In this case, it is better to use a Normal SOP to generate vertex normals on the geometry if vertex normals are desired, rather than have the viewport continually recalculate them.

Auto Generate Tangents

If tangent-space normal maps are used in a material, the viewer requires that tangents are present. It can auto-generate them if they don’t exist in the geometry. The material may have artifacts if the tangents used to generate the tangent space normal map don’t match the tangents generated by the viewer method.


Uses the MikkT algorithm to generate a uniform tangent space. This is a good tangent space but is very slow to generate, especially if the geometry is deforming.


Uses a fast implementation taken from the Polyframe SOP (Two Edges).


Don’t generate tangents. Disable normal mapping for tangent space normal maps.

Scene tab


Antialiasing Samples

Smooths edges of lines and polygons in the viewport. Increasing this increases the amount of framebuffer memory Houdini uses. Only use modes higher than 4× if your graphics card has 2GB of VRAM or more. Modes above 16× are significantly slower and give diminishing returns on quality, so it is best to find a “good enough” setting rather than maxing this value.

HDR Rendering

Produces higher quality render of volumes and transparency. This doubles the amount of framebuffer memory Houdini uses. When this is on, flipbooks will contain HDR images.

You can use this with a LUT (in the Color Correction section) to view super-white values.

Enable X-Ray Drawing

Draw objects with the X-ray flag (bones and nulls) as wireframes when they're behind solid surfaces.

X-Ray Strength

Controls the strength of the X-Ray wireframe. Values less than 1 will dim the lines, and values greater than 1 will widen the lines.

Enable Object Origins

Draw axis and pivot points at the object origin of objects with the Display Origin flag.


Show Name

Show the viewport options menu (titled with the viewport name) in the top right corner of the viewport.


Name for the viewport, for example persp1 for a perspective viewport.

Show Camera Name

Show the camera options menu (titled with the name of the camera the view is looking through) in the top right corner of the viewport. If the view is not locked to a camera, the menu is titled “no cam”.

Show State Status

Show additional text information about the current state, if any.

Show Status Badges

Show icon badges beside the viewport menus to indicate various viewport settings or state that is relevant to the user.

Onion Skinning

Allow Onion Skinning

Geometry objects are drawn with onion skins if their Onion Skinning parameter is set. This is a global toggle to quickly disable all onion skinning in the scene, without the need to turn off each individual objects' Onion Skinning parameter.

Frames Before

Shows a number of skins for frames in the future, before the action at the current frame has taken place. These skins can be tinted with the color swatch to differentiate them from the past frames.

Frames After

Shows a number of skins for frames in the past, after the action at the current frame has taken place. These skins can be tinted with the color swatch to differentiate them from the future frames.

Frame Increment

The number of frames between skins. A frame increment of 1 will show tightly packed frames, whereas higher increments will skip frames. Fast motions are easier to see with a small frame increment, while slow motions are easier to see with a larger frame increment.

Skin Opacity

Each skin can be drawn with transparency so that the main geometry and other skins can be seen more easily. A value of 1 makes the skins opaque, and values closer to zero make the skins more faint.

Color Correction


Applies gamma correction to user geometry in the viewport. This does not affect guides, handles, and other interface elements.


Applies a Lookup Table (LUT) to the use geometry, after gamma is applied.

Apply to Background Image

Apply the gamma correction and LUT to the background image as well as the user geometry.

Viewport Split

Multi-viewport layouts can have their viewports resized. This group contains per-layout options for the split values, in fractions of the total viewer space. These can also be changed by dragging on the dividers in the viewport directly.


Restore the current viewport layout to its default layout.

Camera tab

The controls on this tab apply to the “active” split viewport. To make a split viewport active, move the mouse pointer over it and press N. The menus of the active viewport are drawn in yellow.


View Type

What kind of view the viewport shows – “Perspective”, “UV”, or an orthographic view direction such as “Front”.

Field of View

Set the horizontal field of view, in degrees, of the default viewport camera. This has no effect when the viewport is looking though a camera.

Aspect Ratio

Sets the ratio of width divided by height (eg. 16:9 would be 1.777). This only has an effect when the viewport isn’t looking through a camera.

Crop Offset

Sets the center of the crop region, in 0..1 screen space. The default is the center of the viewport (0.5, 0.5).

Crop Size

Sets the size of the crop region, in 0..1 screen space. The default is (1,1) - no scale.

View Mask Opacity

The opacity of the mask showing which areas of the view will not be rendered (when looking through a camera), from 0 (fully transparent) to 1 (full opacity black).

For example, a very low value would display a transparent mask; a medium value would display a gray mask; and a high value would display a black mask.

Stereo Display Mode

The display mode of stereoscopic images in MPlay. You can use passive polarization glasses to view the images in horizontal interlace mode.


Homing automatically adjusts

Which clipping planes are automatically adjusted when you home the viewport. If you set explicit near and/or far clipping planes, this lets you prevent them from being changed by homing.

Near/Far Clipping Planes

The nearest distance to and the farthest distance from the view camera to draw geometry. That is, geometry nearer to the view camera than the near clipping plane, or farther away than the far clipping plane, is not visible in the view.

Due to GPU Z-buffering issues, the far/near ratio should never be more than one million. If you enter a larger range is selected, Houdini will simply move the near clip plane out until the far/near ratio is one million.

Minimum Home Size

When homing, treat geometry smaller than this size as if it was this size (in Houdini units). This prevents huge camera moves when you home a scene that is very small (such as a single point).

Depth of Field


These controls only affect the OpenGL view, not the image generated by a renderer. These effects can be useful for visualization, or if you are generating imagery straight from OpenGL (for example, using the OpenGL render node).

Camera Depth of Field

Turn this on and set the view to look through a camera to simulate depth of field in the OpenGL scene view. You can control the range that is in-focus using the camera’s F-stop parameter (the F-stop must be non-zero to get depth of field).

This effect works by blurring the OpenGL rendered pixels, so it is quite fast but can give strange results in areas with lots of blurring, since it is a post-process that can only work with the available pixels. So for example it can’t simulate light blurring out from behind objects.


Simulate Bokeh flares for bright points in out-of-focus areas. This is only available if the GPU and operating system support OpenGL 4.4.


The bokeh effect is not currently supported on macOS.


Do not simulate bokeh.


Circular or oval bokeh effect (use the Bokeh Aspect below to create ovals).

From File

Use an image file to define the bokeh shape. Bright and opaque pixels define the shape, dark or transparent pixels define the background. If the image is not monochrome, the view uses the image colors to tint the bokeh flare.

From COP

This is the same as From File above, but gets the image from the output of a compositing node instead of a file. The bokeh image must be in the color plane (C).

Bokeh Aspect

Optionally stretches the bokeh shape vertically or horizontally. The default value 1 does not stretch the shape (so if Bokeh is “Circular”, the shape will be a circle). Values greater than 1 stretch the shape horizontally. Values less than 1 stretch the shape vertically.

Bokeh Boost

Artificially increases the brightness of bokeh hot spots. This can also produce more overall bokeh in the scene.

Foreground Image

Display Camera Foreground Image

Show the current camera’s foreground image, if any. The foreground image is specified as a parameter on the Camera object, and it can be an image file or a COP reference. Foreground images are displayed over the regular geometry.

Depth Plane

When enabled, use the specified plane to adjust the depth of the image so it z-composites into the scene. The image should be rendered from the same camera. This can be used as a standin for a huge number of objects that would otherwise take a lot of memory and time to draw. The plane must be found in the image file or COP as an AOV.

Depth Style

Allows for multiple representations of depth.

Linear Depth

Single channel depth as positive distance from camera.

NDC Depth

Normalized single channel depth in the -1 to 1 range (near to far).

OpenGL Depth

Normalized single channel depth in the 0 to 1 range (near to far).

Camera Position

XYZ position in camera space.

World Postiion

XYZ position in world space.


Reduce this to make the foreground image more transparent so that the geometry behind it can be seen.

Lights tab

Houdini may disable some controls on this tab if your graphics card does not support them.



How to show lighting in the viewer.

No lighting

Ignore scene lights, show objects at constant brightness.

Headlight Only

Ignore scene lights, show objects as if lit by a single virtual sunlight centered on the camera. You can set the Headlight direction below.

Normal Lighting

Use the scene lights to light objects.

High Quality Lighting

Use a slower, higher-quality lighting model that better matches how the rendered scene would look.

High Quality Lighting with Shadows

Use high quality lighting and approximate the effect of shadows for lights that have shadowing turned on. This is slower and uses more memory, but greatly increases the quality of the view.

Houdini recalculates a light’s shadow map(s) when its position, orientation, or projection changes. You may want to disable shadows while editing a light to improve interactivity.

In the High Quality Lighting modes, the Flat shading modes are not available. Smooth shading is always used. The High Quality Lighting modes only work with a perspective view. Orthographic view mode will use normal quality lighting instead.


Show the effect of direct lighting on surfaces.


Show the effect of ambient lighting on surfaces.


Show the effect of specular reflection on surfaces.


Show the effect of light emission (glow) from surfaces.

Max Light Count

Improve shading performance of “High Quality” lighting by limiting the number of lights used. The lights with the highest intensity are used.

Light Sampling

The number of samples to use when rendering area and environment lights when Quality is “High Quality”. Higher numbers produce more accurate results, with slightly lower performance.

Max Light Samples

Improve shading performance of “High Quality” lighting by limiting the total number of light samples across all lights.

Fast Interactive Sampling

Drops the number of samples used in HQ Lighting to improve performance when interacting with the viewer, which includes tumbling, moving handles, or playing the timeline. This may result in slightly different lighting while interacting with the viewer.



The brightness of the headlight when Lighting is “Headlight”.


The headlight is treated as a distant light with parallel rays. This sets the direction of the headlight rays, relative to the camera. The default is over, behind, and to the right of the camera.

Specular Highlights

Show specular highlights caused by the headlight, when Lighting is “Headlight”. When this is off, the headlight casts diffuse light only.

Headlight Occlusion

Do an ambient occlusion pass after lighting to increase contrast on local features and give a better sense of depth to models.


Ambient Occlusion

Enables screen-space ambient occlusion, when Lighting is “High quality”. This shadows objects based on the amount of ambient light that could reach a surface. Areas in corners and sunken areas will be shadowed. Turning this on decreases display performance.

The numeric value increases the quality and range of effect of the occlusion, softening the overall effect.


Changing the scene file’s Unit Length option will change the extent of the ambient occlusion effect.

Shadow Quality

Controls the quality of shadow visualization in the view. Higher levels give better shadow approximations, especially for area and environment lights, but decrease display performance.


All lights are shadowed as if they were point lights, producing hard shadow edges. This is the lowest quality setting.

Antialiased Point

Improve the shadow edges by softening jagged edges caused by light map aliasing.


Area lights use many shadow maps to produce a soft shadow effect. Environment lights perform more sampling. This has no effect on other light types (point will be used in these cases). Moving an area light with this option on will result in slower interactivity.

Antialiased Area

Soften the jagged edges of shadows, which improves the soft shadow look.


Correct for shadow map artifacts, which look like moiré patterns or “shadow acne” on surfaces. In general, increase the first value to fix self-shadowing artifacts on surfaces that are edge-on to the light, and the second value for surfaces perpendicular to the direction of the light.

Light Map Size

Controls the resolution of the shadow maps, based on the individual light object’s settings, or using a global setting for all lights. Increasing the shadow map size will reduce the jaggedness of shadow edges and improve fine shadow detail. Larger maps may affect performance and will use more graphics memory.

Override Light Map Size

Ignore the light object’s shadow map size and use the specified resolution. All Light maps are square.

Use Light Map Size up to

Use the light object’s shadow map size, but limit it to the maximum resolution specified. If an object’s shadow map size is not square, then the width is used for both the width and height of the viewport shadow map.

Shadowmap Memory

Memory limit for shadowmap generation.

Shadowmap Time

Time limit for shadowmap generation. Multiple redraws are queued to show the partially updated scene when this limit is exceeded.


Enable Reflections

Enable reflections using reflection cubemaps. This simulates reflections by rendering the scene to a cubemap with the reflection object removed, at the reflective object’s centroid. Reflective objects are those with a material with a GL Reflect parameter that is greater than zero.

HDR Reflections

Use a FP16 cubemap to store high-dynamic range reflections. When disabled, an 8b cubemap is used (standard 0-1 color range). HDR reflections look brighter, but use twice the texture memory.

Reflection Map Size

Resolution of the cubemap’s square images, in pixels. Larger maps produce sharper reflections at the expense of increased reflection map generation time and texture memory use.

Minimum Reflect

Require that a material have a GL Reflect parameter set to at least this value, otherwise do not consider the material reflective. No reflection cubemaps are generated for objects with non-reflective materials. This can reduce the number of reflection maps generated for very dull materials.

Material tab

Material Effects

Use Materials

Color objects in the view using their assigned materials. When this is off, Houdini draws objects using the Default Diffuse, Default Specular, Default Ambient, and Default Emission below.

Show Geometry Color

The Cd attribute on geometry is multiplied into the material color when this is on.


Draw objects with per-pixel alpha, texture maps with alpha, or material transparency using alpha blending. When this is off, Houdini draws any pixels with non-zero alpha as opaque.


Transparent objects are rendered as opaque if the alpha is greater than zero. This doesn’t require an additional pass or blending, and is well suited to rendering billboards.


Transparent objects are only sorted by object order. Overlapping surfaces within an object may be rendered incorrectly, unless objects are sorted manually in the scene hierarchy list or a Sort SOP is used at the end of the object’s geometry network.


Transparent objects are sorted per-pixel, producing a more realistic display of complex transparent objects.


Transparent objects are sorted per-pixel and are shadowed, if shadows are enabled. More render passes are used to resolve transparent layering issues, if they are needed.


Draw geometry with materials that have a displacement map as displaced surfaces using adaptive tessellation. This may cause a reduction in performance, which can be adjusted with the Level slider.

OpenGL 4.0 is required for this feature. This option will be disabled if OpenGL 4.0 is not available.


Lower values produce fewer tessellated triangles, resulting better performance but lower quality results. Values greater than one produce more triangles than recommended by the adaptive tessellation, up to the OpenGL tessellation limit which is hardware-dependent.


Displacement should be turned off when attempting to select edges or polygons on a surface with a displacement material. Selection does not take the displacement into account so it will be difficult to accurately select components with displacement on.

Default Material


Selects the type of material used when the default material is applied to geometry (no material is assigned). Simple

An untextured material with variable roughness and color settings for diffuse, specular, emission, and ambient values.


A material that uses a MatCap texture (Material Capture) to define the lighting and shading of the surface rather than the lights in the scene.


Diffuse color to use for surfaces that have no material.


Specular color to use for surfaces that have no material.


Ambient color to use for surfaces that have no material.


Emission color for surfaces that have no material. This produces a constant shading effect.

Diffuse Rough

Diffuse roughness to use for surfaces that have no material.

Specular Rough

Specular roughness to use for surfaces that have no material.

MatCap Texture

The file path to a Material Capture (MatCap) texture for the MatCap default material. A MatCap texture encodes lighting and shading information, much like a lat-long environment map.


The intensity multiplier for a MatCap texture.

Material Assignment

Apply Material Stylesheets

Evaluate and apply material assignments and overrides based on material stylesheets affecting the object and geometry. When disabled, the effects of any stylesheets will not be scene but the scene may update and draw faster.

Apply Per-primitive Stylesheets

Allow material stylesheets to be evaluated on a per-primitive basis when applied to geometry. When off, only the object assignment is used.

Apply Stylesheets to Packed Geometry

Allow material stylesheets to be applied to portions of geometry that is packed inside a packed primitive. When off, only the packed primitive assignment or object assignment will be used.

Use Material Overrides

Allow material overrides from material stylesheets or the material_override attribute. When off, material overrides are ignored.

Single Object Material Limit

The maximum number of materials and material variants allowed to be assigned to a single object by material stylesheets, overrides, or shop_materialpath attribute. When this is exceeded, the remaining primitives will be assigned either the default material, object material, or the packed primitive material. A note is made of this in the viewport information button on the right toolbar.

Material Assignments

Limit when material assignments are updated, as this can be expensive.

Always Update

Update whenever material assignments need to be updated.

Disable during Playback

Update when material assignments need to be updated unless Houdini is playing or scrubbing.


Do not update unless Update Materials is manually clicked.

Interactively Update Materials

Update materials when they change. When disabled, changes to the materials themselves will not be reflected in the viewport unless Update Materials is pressed.

Update Materials

Forcibly update all material assignments, stylesheets, and materials. This is useful for stylesheets with external script files when those files change (they are not monitored for changes), or when automatic updates have been disabled.

Fog tab


These controls only affect the OpenGL view, not the image generated by a renderer. These effects can be useful for visualization, or if you are generating imagery straight from OpenGL (for example, using the OpenGL render node).

Uniform Fog

Uniform Fog

Draw a simple atmospheric fog effect in the viewport, essentially blending a tint color with each pixel based on how distant the pixel is from the camera. The effect does not correspond to any volume object (it’s uniform across the entire scene space), and is not affected by lighting.

This effect post-processes the OpenGL rendered pixels, so it is very fast, and can give increased depth cues. It is best used with low Density levels, because higher levels can reveal the “fake-ness” of the effect.

You can also simulate a sun burning through the fog, which can create a nice hazy artistic effect. Of course the “sun” does not actually contribute light to the scene, since this is a post-process effect.

Fog Color

Tints the generated fog. You can use this to look it look more like smog, haze, or dust.


Controls how visible the uniform fog is (how much the fog color blends into the rendered pixels). You should keep this small (below 50) to avoid washing out the entire scene. This value is scale-dependent, so if you are working with, for example, a centimeter-scale scene you will need to adjust this value.


An additional scale factor on the final fog result to make it heavier (greater than 1.0) or lighter (less than 1.0).

Depth Range

The distances at which the fog begins and ends, in Houdini distance units from the camera. All depths beyond the end are clamped to the end depth. Any distance before the fog start does not have any fog.


Fog can be limited to areas above or below an elevation in the scene. Fog will begin at the height value, and slowly increase to its density at height plus falloff (or height minus falloff in the Below case). The falloff gives a softer transition from areas with no fog to areas with fog.

Depth Clip

Anything beyond the depth clip value not have fog applied. This is useful for preventing fog on skyboxes or other image-based background elements.


Enable simple lighting of uniform fog based on a distant light in the scene. The distant light’s color will tint the fog in that direction. Adjust the slider to control the size of the light bloom around the sun (scattering). Very small values produce a crisp sun disc (0.01), whereas larger values will tint most of fog in that direction (above 1). This requires a distant light in the scene. In case more than one active distant light exists, the brightest one is chosen.

Sun Intensity

For uniform fog with the Sun enabled, this provides an additional intensity adjustment on the distant light’s color.

Volumetric Fog

Volumetric Fog

Simulates volumetric lighting/shadows through a single volume, aligned to the camera frustum. Where uniform fog is a post-processing pixel effect, volumetric fog simulates an actual volume, with the lights and geometry in the scene creating light rays and shadow shafts.

You must have high quality lighting turned on in the viewer to see this effect.


You can control volumetric fog display using the RMB menu on the High Quality Lighting button on the [display toolbar||/basics/view#display_toolbar] on the right side of the viewer.

This is only available if the GPU and operating system support OpenGL 4.4.


The volumetric fog effect is not currently supported on macOS.


Currently the volume created by this effect does not blend well with, and is not shadowed by, actual volume objects in the scene. This should be improved in a future version of Houdini.

Fog Color

Tints the volume. You can use this to look it look more like smog, haze, or dust.


Controls how dense the fog volume is. This is not the same as the density setting on a volume object (where 1 would be opaque). Instead it has a similar range to Density for uniform fog, generally from 1 to 100 for meter-scale scenes. This value is scale-dependent, so if you are working with, for example, a centimeter-scale scene you will need to adjust this value.


This controls the resolution of the volume. Low quality is fast but coarse, Medium is a balance between performance and quality, and High and Very High quality are more accurate but much slower.


An additional scale factor on the final fog result to make it heavier (greater than 1.0) or lighter (less than 1.0).

Depth Range

The portion of the view frustum that will have lit fog, in Houdini distance units from the camera. Set this so the depth range is as tight as possible to the lights you want to light the fog. (This is to avoid wasting precision on areas without lights.)

Light Intensity

This effect uses the (rendered) light intensity of the scene lights by default. However, physically correct lighting may look too subtle or too blown out in OpenGL. This is a global scale on the intensity of the scene lights. The view will also look for a gl_fogintensity property on each light. If that property exists, it is used instead of this value.

The default for this scale is 1. It may be useful to set this to 0, so only the lights you specifically add the gl_fogintensity property to will affect the fog.

Light Scattering

The first value controls the intensity of the lit fog when the light ray is parallel to the viewing direction. The second controls the intensity when the light ray is perpendicular to the viewing direction.

To reduce bright light blooms while still keeping god rays visible, lower the first value.

You can set this value per-light by adding the gl_fogscattering property to the light node.



Bloom is a simple effect that mimics subtle atmospheric or lens effects around bright spots in the scene. Adjust the slider to modify the radius of the bloom, which is initially based on the brightness of the pixel being bloomed. This will also affect the intensity of the bloom (larger blooms reduce the overall intensity of the bloom).

Bloom Intensity

Brighten or darken the blooms.

Bloom Threshold

Any pixel brighter than this threshold will begin to bloom (white being 1.0). Increasing this value will produce less blooming in the scene.

Node Override


The path to a node which has Fog Properties that override the settings above (Uniform Fog, Volumetric Fog, Bloom). This lets you switch between override nodes to switch between different looks. Node parameters can also be animated, unlike display options. If a property doesn’t exist as a parameter on the node then it is taken from the display options instead.

Grid tab

3D Ortho

These options control the display of the grid in orthographic viewports. To show grids in a perspective viewport, use the XZ, XY, and YZ Reference Plane options on the Guides tab.

Display Ortho Grid

Show a grid in orthographic viewports.

Grid Offset

The distance in X, Y, and Z between the grid origin and the world-space origin.

Grid Spacing

The horizontal and vertical distance between grid lines, in Houdini units.

Grid Ruler

Draw every nth grid line thicker. Set this to 0 to not draw thicker lines.

Texture Viewport

These options control the display of the grid in UV editor viewports.

Display Reference Grid

Show a grid using the given Grid spacing.

Grid Spacing

The distance between grid lines, in UV coordinate values.

Display Grid Over Image

Show a pixel grid over the texture image, using the Grid pixel spacing and Grid pixel offset.

Grid Pixel Spacing/Grid Pixel Offset

The placement and size of the pixel grid, in pixels. You can use decimal values, for example an offset of 0.5 to snap to pixel centers.

Clamp Grid to Image

Don’t show the grid outside the boundaries of the image.

Display Tile Boundaries

Draw guide lines representing the boundary around the 0-1 tiling of the image.

Background tab

Color Scheme

The background color scheme for viewports.

Display Environment Lights as Backgrounds

Display environment lights as the background in all viewports (when the environment light’s Enable light in viewport parameter is on).

Display Background Image/Video

Show an image or video in the background of the viewports. This can be useful for tracing over an image/video, or looking at a 3D model in front of a context image/video.

When a viewport is looking through a camera, the camera’s Background Image parameter specifies the image. For viewports that are not locked to a camera, Houdini uses the images specified in the tabs below.

Use Texture Filtering

Oversample the background image or video so zooming on the image or video looks smooth and blurry instead of blocky.

Apply Zoom to Background

Zooming the viewport changes the size of the background image or video. When this is off, the image or video remains the same size when you zoom in or out. This option is ignored when Auto-place image is on.

Disk File/COP Image

Get the image or video from a file or a compositing node. Click the chooser button to choose a file/node. Houdini uses OCIO to color correct images on disk. It may be able to guess the image’s color space based on the image filename (for example, myimage-srgb.exr) or format.

Minimum/Maximum UV

The smallest and largest UV values display horizontally (U) and vertically (V) in the viewport.


Environment Map

Draw the image as a 360 degree environment map rather than a 2D image. The image should be a lat-long map or a 6-face cube map.

Auto-Place Image

Automatically scale the image to fit the image inside the viewport (not available for UV viewports). When this option is off, you can use the Image Offset and Image Scale controls to place and scale the image manually.

Fit Mode

Controls how the image should be fit to the viewport. This option is ignored when Auto-Place Image is off.

Image Offset/Image Scale

Manually control the placement and scale of the background image in the viewport, when Auto Place Image is off. A scale of 1 uses the original image size. A scale of 0.5 would make the image half size. Not available for UV viewports.


The strength of transparency of the background image or video. Lower values increase the transparency and higher values decrease the transparency.

The default value is 1, which makes the background image or video fully opaque.


The brightness of the background image or video. Lower values darken the image or video and higher values lighten them.

The default value is 1, which uses the background image or video’s original brightness.

Image Quality

Controls the fidelity of the background image by downscaling it. Lower values give lower quality background images but use less memory.


Clip Start/End

The first and last frames of the background video, which determine the video’s range. Click Trim to open a video preview to select the range.

The default values are 1, 1. Click Reset to reset this parameter back to its default values.

Playback Start Frame

The frame on the Playbar when the background video appears and starts playing.

The default value is 10.

Playback Speed

The speed that the background video plays during playback.

The default value is 1.

Frame Rate

The playback speed of the background video. The parameter maps this imported value, or an updated value that you set, to the scene frame rate (the frames per second set in the Global Animation Options window) so the video plays back at that specific frame rate.

For example, if this parameter is 1 and the Global Animation Options' FPS parameter is 24, the video advances by one frame every 24 scene frames during playback. Similarly, if this parameter is 60 and the Global Animation Options' FPS parameter is 30, the video advances by two frames for every scene frame during playback.

The default value is 24. Click Reset to reset this parameter back to its default value.

Color Space

The color space to apply to the background video.

The default value is sRGB - Texture.

Texture tab


Display Textures

Show surface textures in the view. Turning this off can increase display performance.

Multi Texturing

Displays all layers of multi-layered textures. Turning this off will only display the first texture, but can increase display performance.

Projected Textures

Show spotlight projection maps in the view (see the Projector map parameter on the light object). This adds one drawing pass per spotlight in the scene. Even when this option is on, the projected maps will not affect transparent objects or volumes. This option only works when using High Quality lighting.

Mipmap Textures

Increase the quality of displayed textures by producing a mipmap for enhanced filtering. When this option is off, textured objects will sparkle when viewed at a smaller scale. This option causes textures to use 50% more memory on the graphics card.

Anisotropic Filtering

Higher values increase the quality of textured objects when a surface is nearly parallel to the view direction. Increasing this will slightly decrease display performance.

Texture Cache

Texture Cache Size

The maximum amount of memory to use for textures, both 2D and 3D (in megabytes).

OpenGL always stores textures in main memory and may swap them to and from the graphics card, so this setting can affect the amount of memory available to Houdini. The maximum texture cache size is limited on 32 bit computers.


Remove all currently unused textures from the cache and force an update of all textures that are in use.


Checks all textures with a file source to see if they are updated and reload them if they are.

Single Texture Limit

The maximum amount of memory that a single texture can consume (in MiB). Even with a large amount of available VRAM, very large textures can have performance issues. This option is automatically set based on the detected VRAM (where possible), from 64MiB to 256MiB. It is not saved to .hip files.

Viewport Texture Use

Displays the amount of texture cache currently in use. To flush the cache, use Render ▸ Update Textures from the main menu.

Dynamically reduce texture scale

When drawing the viewport, if the total size of all textures used is greater than the texture cache size, texture thrashing will result. This causes severe performance degradation as textures are continually reloaded and deleted on every redraw. If this option is on, Houdini detects this situation and on the next redraw scales textures to a size that will fit in the cache.

Reduce 2D textures

Apply dynamic texture scaling to 2D textures. The box next to the option shows the current dynamic scale.

Reduce 3D textures

Apply dynamic texture scaling to 3D textures. The box next to the option shows the current dynamic scale.

2D Textures

Limit Resolution

Limit the size of 2D textures to the specified size (if it is less than the maximum size allowed by the graphics card). When this option is off, texture sizes are only limited to the maximum size supported by the graphics card. This option can reduce graphics card memory usage and improve performance, at the expense of decreasing display quality of large textures.

This only affects new textures and will not cause updates of loaded textures (use Render ▸ Update Textures from the main menu to apply the limit to these textures).

Cube maps are also affected by this size, though it applies to their faces. The maximum size of shadow maps is controlled on the Lights tab.

HDR Textures

Show High Dynamic Range texture images. Using HDR images will give better shading on textured surfaces when lighting is on in the view.

Full HDR

All HDR formats are allowed (32 bit and 16 bit). This uses the most graphics memory, as 32 bit HDR images use 4× the amount of RAM that an 8 bit image does. This may also negatively affect performance on older graphics hardware.

16b HDR only

Use only 16 bit floating point textures images. Any 32 bit HDR images will be down-sampled. This provides a good balance between rendering quality, graphics memory use, and performance.

8b SDR

Down-sample all textures to 8 bit (Standard Dynamic Range). Any super-whites are clamped to (0,1) (black,white). HDR textures may look washed out in this mode.

8b Compressed

Compress all textures to 1/4 or 1/8 their size by reducing color resolution. This option will reduce the memory used by textures, but also reduce the color accuracy of textures.

Scale Textures

Scale all 2D textures to a percentage of their original size. You can use this to quickly improve performance of a highly-textured scene if it is stuttering on redraw.

This scale is applied before the Limit Resolution option.

3D Textures

Limit Resolution

Limit the resolution of 3D textures, which are often used to represent volumes.

If you need better quality, increase the limit values. Do not turn this off unless you really know what you're doing. Larger 3D textures can use a huge amount of graphics card RAM. High resolution (simulation) volumes can use far more memory than reasonable to display on a graphics card. (When this option is off, volumes are limited to the maximum OpenGL 3D texture size.)

This only affects new 3D textures.

Use 2D Texture Settings

Copy the HDR textures and Scale textures settings from the 2D textures options. When this is off, you can specify separate values for 3D textures below.

(8 bit Compressed is not supported for 3D textures, so if the 2D setting is “8b Compressed”, 3D textures will use 8 bit SDR.)

HDR Textures

Show High Dynamic Range color and opacity for volumes and other 3D textures. This can improve the quality of volumes at the expense of performance and graphics memory use.

Scale Textures

Scale all 3D textures to a percentage of their original size. This can be useful for displaying extremely large simulated volumes.

This scale is applied before the Limit Resolution option.

Optimize tab


Visible Objects

Only object names that match this pattern will be visible in the view.

Remove Backfaces

In shaded view modes, do not draw polygons that face away from the camera. This can improve display speed on very heavy polygonal geometry.

For closed (airtight) surfaces, back faces are not visible anyway, but open surfaces often have visible back faces that will disappear when this option is on. For example, if you create a cylinder (without end caps) and look down at it with this option on, only the front half of the cylinder will appear in the view.

Draw Hulls Only

Draw the hulls of certain primitives instead of the normal geometry. This applies to NURBS surfaces and curves, Bezier surfaces and curves, metaballs, volumes, and VDBs.

Distance-based Packed Geometry Culling

When enabled, replace packed geometry with bounding boxes if the total polygon count of all the packed geometry within the view frustum exceeds the Scene Polygon Limit. Distant packed geometry is replaced with boxes first. This preserves viewport performance for large scene files at the expense of visual quality, making it easier to navigate the scene.

Scene Polygon Limit

The maximum number of polygons to draw in the scene. When this is exceeded, some packed primitives will be replaced by their bounding boxes until the total number of polygons is under the limit.

Bounding Boxes

Determines whether to use wireframe or shaded bounding boxes when replacing geometry in the scene, or omit them entirely.


Remove the geometry but don’t show anything in its place.


Always show wireframe bounding boxes for culled geometry, regardless of the main viewport shading mode.


Always show shaded bounding boxes for culled geometry, regardless of the main viewport shading mode.

Current Shading Mode

Use wireframe bounding boxes when the geometry is being wireframe, or shaded if the geometry is being drawn shaded.

Optimize Packed Geometry

Perform a series of optimizations on the packed geometry which make it faster to display. The downside of this is that the wireframe representation will no longer match the original geometry’s topology and it will take slightly longer to initially process the geometry.

Crowd Agents

LOD Agent Reduction

Automatically generate a series of reduced-polygon levels of detail for agent shapes, and swap them in based on an agent’s distance to camera, to make display of far-away agents faster. Each level has half as many polygons as the previous one. The closer the slider is to the “performance” side, the more low-poly levels Houdini generates.

Shape Point Cutoff

The minimum number of polygons in a reduced-polygon model generated by LOD agent reduction. Houdini will not generate any levels of detail with fewer than this many polygons.

Max LOD Levels

The maximum number of reduced levels of detail generated by LOD agent reduction. For example, if an agent model has 100000 polygons and Max LOD levels is 4, Houdini will generate levels with polygon counts of 50000, 25000, 12500, 6250, and then stop. Note that Houdini may not actually generate this many levels if it reaches the Shape point cutoff.

Base LOD Level

Which reduced level of detail to use for the display of the closest agents. The default is 0, meaning use the original model. For example, setting this to 1 would use the first reduced level of detail (with half as many polygons as the original model) even close to the camera. This lets you quickly and drastically reduce the overall polygon count of a crowd scene for better display performance.

Single Bone Deform

Choose whether agent models are deformed by all bone weights (up to 4), or optimize performance by only deforming based on the highest-weight bone.


Always use full deformation (best quality).

Reduced LOD Agents Only

Use full deformation for the primary level of detail, and single bone deformation for reduced levels of detail.


Always use single bone deformation (fastest).

Wireframe Display

Agent rigs can be displayed in several ways when in wireframe mode.


Rig joints are connected with a line segment. This can be faster than bone display depending on the graphics hardware.


Rig joints are connected with shaded bones (default).


Interactive Mode

Tells viewport plugins (“render hooks”) to continuously update during user interaction (tumbling, tracking, dollying). This only affects viewport plugins, it has no effect on normal Houdini viewport drawing. The plugin must look at this option and implement its own 'performant' drawing mode, it is not automatic.