Viewport 2.0 Options

From the panel menus, select Options > Viewport 2.0 Options.

Performance

Consolidate World

This option attempts to combine geometry caches for shapes using a common material. In many scenes, this can produce a significant improvement in performance at the cost of additional memory.

When combining geometry, Consolidate World moves vertices of multiple objects into a new shared object-space. Therefore, if your plug-in shader make assumptions about the object-space coordinates of an object, Consolidate World will break these assumptions and your shader may not render correctly.

To use this feature, vertex normals must be re-normalized to unit-length. Therefore, materials that do not use unit-length normals are not compatible with Consolidate World.

When an object changes, it is unconsolidated. It is then re-consolidated if it remains unchanged for a few frames. Therefore, if you notice that an object draws differently in the viewport a very short time after being edited or unselected, it may be due to Consolidate World. If this is undesirable, then disable this option.

Tip: If the workspace becomes unstable, try disabling this option.

Important: Some operations incur slight delays when using this option.

Vertex Animation Cache

Disabled by default. When on, Maya caches the resulting mesh data for each frame of animation, allowing for better performance during various tasks, including scrubbing animation, tumbling the camera, moving transform nodes, and adjusting lighting with real-time preview in Viewport 2.0.

Important:

Vertex Animation Cache does not support sub-frame animation.
Ensure that you have sufficient memory available for large or complex scenes with many objects or heavy geometry before enabling Vertex Animation Cache. Performance may be compromised if sufficient memory is not available.
In order for this option to work, any history nodes that change the topology of mesh data must come before any node that changes based on time. This includes the Deform User Normals option on the skinCluster node and the polyTransfer node, which must be turned off in order to work with Vertex Animation Cache.
If your Playback speed is clamped, for example with a Max Playback Speed of 30 fps, Vertex Animation Cache may require several cycles to completely cache all of the animation in the timeline.
An alternate method to increase efficiency is to set the Time Slider preferences with Playback speed set to Play every frame, and Max Playback Speed set to the desired frame rate (Window > Settings/Preferences > Preferences).

Disable: Turns off caching. You can turn Vertex Animation Cache off if you are working only on interactive workflows like modeling or lighting and you don't need to play back the scene.
Hardware: Caches the vertex buffers on the GPU. Hardware is faster than System if there is enough memory on the graphics card to store all of the scene, texture and animation frames. Otherwise; if there is not enough memory, you may experience system slowdowns. Switch to System if this occurs.
System: Caches the vertex buffers in system memory. System is slower than Hardware, but typically there is much more memory available for caching buffers.

GPU Instancing

Enable this option so that if you have multiple instances (or especially a large number of instances) of a Maya shape, such as a poly object, and all of which use the same material, then they render using hardware instancing. This eliminates graphics driver state change overhead and render pipeline overhead for much faster rendering results.

In some occasions, you may want to disable GPU instancing, for example:

Occasionally, when your instances have negative scale and you have other features such as backface culling enabled, your lighting may appear incorrect. In general, if you have negative scales and your instances appear incorrect, disable GPU Instancing.
If you have a large number of very small instances, enabling this option may slow down your scene. For example, if you have a large scene with 1000 non-instanced objects, and you select them all and then duplicate-instance once, you will have 1000 instance-pairs. In situations like these, the overhead for GPU instancing may cause your scene to be slightly slower if the instance batches are very small and there are a large number of batches. There is no instancing overhead for objects which have no instances.
Plug-ins that use MPxGeometryOverride may or may not support GPU instancing.

Thread Dependency Graph Evaluation

When enabled, Maya attempts to evaluate separate characters (sub-graphs of DG) in separate threads.

Light Limit

Use this option to set the maximum number of lights that are used in rendering. Hidden lights are not included. The default is 8 lights and the maximum is 16.

Transparency Algorithm

Select from the following transparency sorting algorithms.

Simple

No transparency sorting.

Object Sorting

Objects are sorted by depth. Objects further away are drawn before objects closer to the camera.

Weighted Average

This is an order-independent transparency algorithm, and no sorting of objects or polygons is involved. The final color is computed as a weighted average of all the transparent objects in a pixel. The advantage of this algorithm is that it is order independent and is therefore fast even for a large number of transparent objects, hair, particle systems, and so forth. This algorithm is ideal for distant objects where you require a quick algorithm for rendering transparency but do not require high quality. It is accurate for single level transparency and is stable, with no jumping artifacts from sorting.

dx11Shaders and ShaderFX support all transparency algorithms.

Depth Peeling

Depth peeling transparency renders per-pixel transparency via multiple passes. At each pass, the next transparent layer in each pixel is peeled and composited behind previous layers. After the set number of peeled layers, all remaining layers are handled in a single pass using the weighted average algorithm. This helps to balance transparency quality versus performance, as depth peeling provides very high quality transparency but with a performance cost.

Note: Multisample Anti-aliasing is not supported with Depth Peeling. It only affects opaque objects but not transparent objects.

Alpha Cut

This transparency algorithm provides accurate object sorting for fully transparent and opaque objects. It generates less optimal results when rendering semi-transparent regions. Alpha Cut provides performance gains that are comparable to the Simple mode, while matching or exceeding the image quality of the Object Sorting algorithm.

This algorithm is ideal for quickly rendering foliage as well as light-weight objects for the purposes of pre-visualization or low LOD stand-ins.

Transparency Quality: For the Depth Peeling transparency algorithm, use this slider to select the number of layers to peel before Maya starts using the weighted average algorithm. Choose from a range of 2 to 10 passes.

Maximum Texture Resolution Clamping

Clamp Texture Resolution

When on, any textures with a resolution above the Max Texture Resolution are downscaled so that you can still work in textured mode while saving memory.

If your scene contains more high resolution textures than your video card has enough memory to handle, your textures are not loaded and your scene appears in non-textured mode. An error message appears indicating that the texture RAM limit has been exceeded and suggesting that you reduce the Max Texture Resolution clamp.

This error message appears whenever the video card has insufficient memory to handle the textures, which can occur if Clamp Texture Resolution is not enabled; or, if the Max Texture Resolution is se updated limit too high and the texture resolution is not downscaled enough for the available GPU memory.

Max Texture Resolution

Set a value between 32 and 8192.

Reload All Textures

After changing the Max Texture Resolution, click to apply the updated limit to the textures already loaded in your scene. Otherwise, the updated limit is only applied to newly loaded textures.

Bake Resolution for Unsupported Texture Types

Color Textures, Bump Textures: Maya shading nodes that are supported as baked textures are listed in Internal texture baking in Viewport 2.0. You can set the resolution for baked textures using these attributes.
Rebake All Textures: Click to see the effect of the change on currently loaded textures.

Screen-space Ambient Occlusion

Enable: Select this option to enable screen space ambient occlusion.

Note: When Multisample Anti-aliasing or Screen-space Ambient Occlusion are enabled, UI items such as locators or wireframe mesh may draw on top of semi-transparent objects.
Amount: Specifies the intensity of the ambient occlusion that occurs. Default is 1.0 but you can choose from a range of 0.0 to 3.0.
Radius: Specifies the radius of the sampling area (in screen space).
Samples: Specifies the number of occlusion samples that occur to improve the appearance of the ambient occlusion effect.

Note:

When post-effects such as screen-space ambient occlusion are on, the drawing of wireframe and components are not affected. Filled display for surfaces are affected. However, wireframe and components for surfaces are not blended with the filled drawing if the surface is semi-transparent. For example, if you create an object, then set its shader transparency to semi-transparent and select the object or its components, you will see a difference between enabling and disabling Screen-space Ambient Occlusion.

In addition, unlike the other transparency options, when you use Weighted Average for the Transparency Algorithm, wireframe drawing is not blended with the transparent surface drawing.

Hardware Fog

Falloff

Select among the following options:

Linear: Enables the options Start and End and creates a fog that slowly diminishes from the center of light.
Exponential/Exponential squared: Enables the Density option that determines the thickness of the fog. This option creates a fog that diminishes quickly, following an exponential curve. Exponential squared creates a more realistic fog.

Density

Controls the optical density of vapor in the fog layer. Increase this value to thicken the fog.

Start

Determines the distance from the camera at which the fog begins.

End

Determines the distance from the camera at which the fog ends.

Tip

You can animate the End value so that your fog appears to spread or dissipate during an animation.

Color

Sets the color of the fog.

Alpha

Sets an alpha value to determine the opacity of the fog.

Note: Hardware fog is not supported with image plane. To create a washed out, fog effect, adjust the Color Gain and Color Offset in the imagePlaneShape node.

To exclude specific objects from fog rendering (for example, if one of your objects already has fog baked in), or adjust the amount of fog effect applied to an object, adjust the Hardware Fog Multiplier attribute in the object's Attribute Editor, Object Display section. When this value is set to 0, the object is not affected by fog. When this value is set to 1, fog is applied to the object. Adjust the slider to a value between 0 and 1 to adjust the amount of fog effect.

Anti-aliasing

Smooth Wireframe: Select this option to enable anti-aliasing of all lines in the scene view, whether UI or geometry wireframe. When a line is drawn, Maya blends it around the edges. By contrast, multisample anti-aliasing is per scene sampling and affects all objects. This option only takes effect when Multisample Anti-aliasing is disabled.
Multisample Anti-aliasing: Select this option to enable multisample anti-aliasing.

Note: When Multisample Anti-aliasing or Screen-space Ambient Occlusion are enabled, UI items such as locators or wireframe mesh may draw on top of semi-transparent objects.
Sample Count: Increase the number of samples for better anti-aliasing quality. A high Sample Count gives slower performance but with better anti-aliasing results.
Note: Sample count options are video card dependent. Only valid options are displayed in the drop-down menu.

Floating Point Render Target

These options are also available in the Color Management preferences.

Enable: When on, you can render to a floating point buffer instead of an integer one.
Format: Select from the following:

Note: Supported formats are card dependent and only valid options are displayed in the drop-down menu.

R32G32B32A32_FLOAT: Select this option for 32-bits/channel, RGBA channels.
R32G32B32_FLOAT: Select this option for 32-bits/channel, RGB channels.
R16G16B16A16_FLOAT: Select this option for 16-bits/channel, RGBA channels.