This page provides details on the settings found on the IPR tab of the Render Settings.


Page Contents



To use IPR you must have V-Ray assigned as a rendering engine and use the IPR function of Maya. The IPR tab in the Render Settings window allows you to control the settings of IPR.

IPR GPU supports these GPU modes per operating system:

  • Windows: CUDA or OpenCL

  • Linux: CUDA or OpenCL

  • Mac OS X: CUDA


To use the GPU engine, a video card that supports following is required:

  • OpenCL 1.1 or higher.

  • CUDA 2.0 or higher.

  • The CUDA engine for IPR GPU works on NVIDIA cards with the Fermi, Kepler and Maxwell architectures.

  • For users of NVIDIAs, we recommend the use of the CUDA engine when rendering with IPR GPU.




UI Path: ||Render Settings window|| > IPR tab




CUDA IPR engine




IPR engine – Specifies the back-end for IPR mode. The Same as production renderer option will use the same engine that is used for production rendering specified in the VRay tab.The other possible options are:

CPU – The CPU engine is used. This engine does not require a graphics card and supports many of the regular V-Ray renderer features, including procedural textures and complex materials.
– A GPU engine based on OpenCL is used. This engine uses graphics cards that support OpenCL and can be very fast depending on the hardware, but has somewhat limited abilities with regards to shaders. The OpenCL and CUDA engines have the same set of capabiltities, but for NVIDIA GPUs it is recommended to use the CUDA engine.
– A GPU engine based on the NVIDIA CUDA platform. This engine uses NVIDIA graphics cards that support CUDA and is the recommended engine for NVIDIA GPUs. Like the OpenCL engine, it can be very fast depending on hardware, but has limited abilities with regards to shaders. The OpenCL and CUDA engines have the same set of capabilities. The recommended choice for NVIDIA GPUs.

Coherent Tracing – Improves the rendering speed at the cost of a visually unpleasant intermediate result. Can be used in animation or baking mode.

GPU Resize Textures – Determines how textures' resolution/size will be handled to help optimize memory usage. This parameter is only effective when Type is set to Open CL or CUDA. The possible values are:

Resize all textures – Adjusts the size of high-resolution textures to a smaller resolution in order to optimize render performance. The GPU engine will load as much texture tiles on the GPU as it can, then swap the ones that are needed between GPU RAM and CPU RAM.
Full size textures – Textures are loaded at their original size.

GPU Texture Size – When GPU Resize textures is enabled this value specifies the resolution to which the textures will be resized. This also controls the resolution for baking unsupported procedural textures to bitmaps so they can be loaded into memory.

GPU Texture Format – Controls the file format of the textures used in GPU rendering. You can choose between 8 bit, 16 bit and 32 bit.


Trace depth – Represents the maximum number of bounces that will be computed for reflections and refractions. The individual material reflection/refraction depth settings are still considered, as long as they don't exceed the value specified here.

GI depth – The number of bounces for indirect illumination. Other GI settings (e.g. whether GI is enabled or disabled) are taken from the production V-Ray renderer.



Max. render time (min) – Since IPR uses progressive path tracing it would never stop rendering unless specified otherwise. This option allows the user to specify a maximum render time in minutes.

Max. paths per pixel – Allows the user to limit the maximum quality achieved by IPR, by specifying a maximum number of rays traced for each pixel. Once the specified number has been reached V-Ray will stop rendering.

Max. noise – Specifies the noise threshold for IPR. Once the threshold is met for a certain part of the image, V-Ray will stop sampling it and will reassign rendering power to part of the image that require more sampling.



The parameters in this section affect the performance of IPR. Note that the optimal values for a given machine and network configurations may be different from the defaults. The user is encouraged to experiment with these values to find the optimal ones.

CPU Ray bundle size / GPU Ray Bundle size – These control the number of rays that are sent to the IPR render servers for processing when using IPR, running on CPUs or GPUs respectively. When using distributed rendering, the smaller sizes cause more frequent client/server communication with smaller network packets thus decreasing the speed of the renderer but increasing the interactivity and vice versa. Note that this number is not the exact amount of rays, but is proportional to it. It is not recommended to increase this value beyond 512.

CPU Rays per pixel / GPU Rays per pixel – The number of rays that are traced for each pixel during one image pass. The greater the value, the smoother the picture from the very beginning of the rendering with GI, but interactivity may be significantly diminished. Increasing this value also reduces amount of data transferred from the render servers back to client machine.

Progressive RPP – When enabled, IPR starts rendering the image with a lower Rays Per Pixel value and then progressively increases it. This allows us to speed up the initial preview of the image.

Undersampling – When enabled, IPR starts rendering the image at a lower resolution in order to speed up the initial preview. Later the image is rendered at its final resolution.

Show statistics – When enabled, some useful values are shown in the upper right corner of the V-Ray Frame Buffer window:

  • IPR version
  • Number of calculated camera paths (i.e. image samples) per second, in thousands (Kpaths/s)
  • Time elapsed since the current image started rendering. This value is reset whenever a change is made to the scene that requires recalculation of the image


Advanced Parameters

Restart when the Render Layer changes – When enabled, V-Ray will restart the IPR process whenever the active render layer in Maya is changed.

Use out of process IPR – When enabled, the IPR session will run in an external process. When disabled (default), IPR will run in the same process as the host application, which allows for faster starting and updating.

Low GPU thread priority – When enabled, V-Ray will try to lessen their use of GPUs working on displaying graphics to the monitor(s) in order to give them more resources to complete other processes and tasks for the OS. This is done by using a lower value for Ray bundle size for those GPUs with attached displays (Rays per pixel is still the same for all GPUs). This can affect overall performance, and the render speed might be reduced. It is recommended to utilize a separate GPU for the display, if possible.


Viewport Parameters

 These settings control the way IPR works in the Maya viewport.




Resolution (%) – Specifies the resolution at which IPR will render, as a percentage of the actual viewport resolution. Can be lowered to increase the performance.

Selection Mode – Specifies how selected objects are displayed in the viewport.

None – Do not show any selection feedback. Note that you can still select objects by clicking them.
Wireframe – Show a selection wireframe over the rendering similar to what Maya does.
Box – Show a simple selection bounding box that allows a better view to the rendered image.

sRGB – When enabled, converts the image to sRGB color space in the viewport. Note that this setting is ignored if Maya color management is enabled in Maya 2016 and later.

The list of Show check boxes allows fine control over which Maya components will be visible in the viewport while IPR is running.