GPU Cache Preferences

GPU Cache preferences let you set the system graphics card parameters that control the behavior and performance of the gpuCache plug-in. You can set the following preferences in the GPU Cache category of the Preferences window.

For information about GPU caching, see Pipeline caching overview.

Hardware Information

Displays read-only information about the graphics card installed in the system.

Manufacturer: Graphics card manufacturer.
Model: Graphics card model.
Driver Number: Displays the version of the graphics card driver installed on the system.
GPU Memory in MB: Amount of graphics card memory being used by the system.

Set All to Automatic

Turning on Set All to Automatic sets the Automatic option for all GPU Cache preferences to on. Turning Set All to Automatic to off does not disable the Automatic options.

The default Automatic options are determined by the system's hardware and operating system profile. When the Automatic options are off for a preference, you can input values or choose between the available options.

VRAM Management

Maximum VRAM for All Cache Buffers: Specifies the maximum total size of the vertex buffer objects allocated for the gpuCache plug-in and GPU caching operations.
Maximum Number of GPU Cache Buffers: Specifies the maximum number of vertex buffer objects allocated for GPU caching.
Minimum Vertices Per Shape for GPU Caching: Specifies the minimum number of vertices a shape must contain before it uses a vertex buffer object. Shapes with fewer vertices than the specified value use system memory.
Low VRAM Operation: Specifies which memory source to use to draw geometry when insufficient video card memory is available to maintain the vertex buffer objects each frame.

OpenGL-Based Selection Options

OpenGL Selection Mode: Specifies whether vertex arrays or graphics card memory is used to draw geometry when performing OpenGL picking for selection.
Wireframe Threshold Until Raster-Based Selection: Specifies the threshold value that determines if OpenGL picking or raster-based picking is used for wireframe selection. If the number of vertices per cache node in the cache exceed the set value, raster-based picking is used.
Surface Threshold Until Raster-Based Selection: Specifies the threshold value that determines if OpenGL picking or raster-based picking is used for surface selection. If the number of vertices per cache node in the cache exceed the set value, raster-based picking is used.

Advanced Display Settings

Disable Vertex Arrays: Specifies if vertex arrays or OpenGL primitives are used.
Two Sided Lighting: Specifies if OpenGL two-side lighting or an emulation of Maya two-side lighting is used by the graphics card.
UV Coordinates: Specifies whether or not UV coordinates are written to or read from GPU cache files.
Background Reading: Turns on background loading of GPU cache files into the graphics card memory. Objects in the GPU cache display in the scene vew as the cache loads.

You can delete, duplicate, and rename a gpuCache node while it loads. Background Reading applies to all GPU cache files including Scene Assembly Cache representations.
Background Reading Refresh Interval: Specifies the refresh frequency of the scene view draw when GPU cache files load in the background.

Except where otherwise noted, this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License