Octane Render - System Requirements (Windows and MacOS)

Last modified: 10 March 2025

Hardware requirements

Octane can now leverage RTX hardware acceleration to evaluate previously CUDA-only shaders, such as the new dirt shader. After removing this limitation we have experienced significant 1.6x+ performance increases in many production scenes that previously had almost no speed improvements with RTX activated.

OctaneRender® requires an NVIDIA® CUDA®-enabled video card. It runs on Ampere, Maxwell, Pascal, Titan, Volta, and Turing GPUs. Texture limits and differing power efficiency ratings also apply, depending on the GPU microarchitecture.

GPUs from the GeForce® line are clocked higher and render faster than the more expensive Quadro® and Tesla GPUs.

GeForce cards are fast and cost-effective, but have less VRAM than Quadro and Tesla cards. OctaneRender scales well in a multi-GPU configuration, and can use different types of NVIDIA cards at once, such as a GeForce RTX 4080 combined with a RTX A6000. The official list of NVIDIA CUDA-enabled products is located at https://developer.nvidia.com/cuda-gpus.

OctaneRender does not require RTX, but it does render some scenes much faster when RT Core hardware is present.

To use the engine's out-of-core features, we recommend using at least the following hardware:

8-core CPU
16 GB RAM
A CUDA-enabled card with at least 2 GB VRAM

Drivers

OctaneRender™ v4 requires an Nvidia driver supporting at least CUDA 7.5. On Windows and Linux, a recent enough Nvidia driver is necessary.

For MacOS versions, use the most suitable driver released from the Nvidia website.

Failure to install these driver versions may result in stability problems. The OctaneRender™ Team cannot provide support to users of different driver versions.

Internet Access

Except for the demo versions, all OctaneRender editions require authentication with its designated license key, and it requires internet access during the initial launch. Once you launch the program, OctaneRender requests your OTOY credentials, and it attempts to retrieve an available license from the OctaneRender LiveTM server.

Software Requirements

OctaneRender® for LightWave 2024® requires:

  • LightWave 2024® Plugin - If you do not have an Octane subscription, you can purchase it on our purchasing page. An Octane Render subscription is free for a Single GPU and included with LightWave 2024.

  • Matching Version Numbers - If you are using an Octane Render Standalone version, License version numbers must be identical between Standalone and the plugin.

  • LightWave® Host Application - LightWave® host application versions 2024 and above for Windows®, and versions 2024 and above for MacOS®.

To install OctaneRender® on Windows®, make sure you install the appropriate drivers for your GPU. After this, you only have to run the installer.

OctaneRender is a CUDA® 9.1 application that requires an NVIDIA® graphics driver version 388.x or higher.

You can download the latest video card drivers for Windows here.

NVIDIA® cuDNN Library File

If the cuDNN (CUDA Deep Neural Network library) is not installed, you will be prompted to install it by the following prompt:

cuDNN_Prompt.png

  1. OctaneRender® requires NVIDIA® cuDNN to run. If necessary, download the cuDNN library file from either https://render.otoy.com/downloads/ef/08/f5/63/cudnn_8_0_4_win.zip (or copy the URL into the prompt)

  2. The library file should be placed in either of these folders: C:\Users\[user]\AppData\Local\OctaneRender\thirdparty\cudnn_8_0_4\ (where X.x.x is the version) C:\Program Files\OTOY\[OctaneRender 4.x]\

Downgrading Drivers

Some drivers are released in a beta state and may cause errors when running OctaneRender®. You can revert to the previous driver by downgrading or installing an earlier but more suitable driver version. Windows®

After downloading a suitable driver from https://www.nvidia.com/Download/index.aspx, use DDU to remove the installed driver, then perform a clean installation of the preferred driver.r on-board graphics or a second graphics card for the Windows display adapter and dedicate a more powerful CUDA-enabled card for rendering.