Documentation

​
​

Development

User Acquisition

Monetization

Industry

Asset Transformer SDK

2026.1

Supported
​

User Manual

Python API

C# API

Changelog

Discussions

Asset Transformer SDK

2026.1

Supported
​
​
​
Get started
  • System requirements
  • Licensing
  • Setup
  • Migrate from Scenario Processor
Framework
  • Framework
  • Preferences
Input/output files
  • Supported formats
  • Import files
  • Export files
  • Export textures
  • CAD vs Mesh
Scene
  • Product structure
  • Occurrence
  • Component
  • Material
  • Image
Algorithms
  • CAD
  • Healing
  • Optimization
  • Combine and bake
  • Reconstruction
  • UVs
Viewer
  • Viewer
Pixyz UI
  • Setup
  • Overview
  • Create a custom UI
Guidelines
  • Data preparation
  • Process point clouds
Debug and diagnose
  • Log files
  • Debug best practices
Other
  • Samples
  • Glossary
  1. Asset Transformer SDK (ex Pixyz)

System requirements

This page outlines the hardware and software requirements needed to run the Pixyz SDK effectively.
Read time 3 minutes
Last updated 9 months ago
This section lists the hardware and software requirements to run the Pixyz SDK.

Operating system

Operating system version

CPU

Additional requirements

WindowsWindows 10 or newerX64 architectureVulkan 1.3 Runtime or later (installed by default with NVIDIA drivers)
Windows Server2019/2022X64 architectureVulkan 1.3 Runtime or later (installed by default with NVIDIA drivers)
Linux
Ubuntu Debian
X64 architecture
Vulkan 1.3 Runtime or later
Libc 2.33 (e.g Debian bookworm or Ubuntu 22.04)
macOSSonoma 14 or newer
X64 architecture
Apple M1 or above (Apple silicon-based processors)

Install Vulkan

  • Windows: install from here
  • Linux:
sudo apt-get install -y libopengl0 libegl1 libgl1-mesa-glx libvulkan1 libgomp1

Hardware

Since loading and optimizing complex and heavy CAD files requires a lot of computing power, you should use Pixyz with the best hardware configuration possible (powerful CPU-GPU, confortable quantity of RAM). Please note that system requirements heavily depend on the type and complexity of CAD and 3D assets your company or studio is dealing with (eg: a game character vs. a complete oil and gas platform)
  • One instance of Pixyz is multi-threaded (up to 32-threads per Pixyz process)
  • 1 CPU with 4 to 32 CPU cores is recommended. Ideal configuration is 8-12 core per Pixyz task (can be configured within Pixyz settings (see Preferences))
  • Memory usage depends on the volume of data that Pixyz will deal with during the task
    It is not easily predictable but for simplification matters, the number of polygons in your scene will be the major indicator of memory consumption. We recommend sizing a comfortable amount of RAM in order to prevent Pixyz from exiting during the process due to lack of memory.
    As a reference, 1M polygons takes about 0.5Gb of RAM and a point cloud of 1M points takes about 50Mb of RAM
Find below the recommended and minimum system configurations to run Pixyz efficiently:

Recommended

  • Processor: Intel Core i7 3.0 GHz or higher
  • RAM: 16 GB or more
  • Graphics Hardware: NVIDIA GeForce RTX 3080
  • Disk Space: 1 GB or more (with dynamic swap)
  • Operating System: Windows 10 and 11 64-bit, Linux Ubuntu/Debian, Docker, Windows Server 2019/2022.

Minimum

  • Processor: x64 dual-core 2GHz
  • RAM: 4 GB
  • Graphics Hardware: OpenGL 4 compatible
  • Disk Space: 200 MB
  • Operating System: Windows 10 and 11 64-bit, Linux Ubuntu/Debian, Docker, Windows Server 2019/2022.

GPU (optional)

Important
AMD graphic cards are not fully supported, using one might result in a poor experience of Pixyz GPU-based features. We recommend using NVIDIA graphic cards.
A few functionalities are accelerated on GPU. If there is no GPU on the machine, your integration will fail on using GPU-accelerated capabilities. This will activate the CPU fallback for any GPU algorithm which can lead to significant performance reduction (up to 100x slower than with a GPU).
GPU-acceleration can be turned off using the following Pixyz API command:
core.setModuleProperty("Algo", "DisableGPUAlgorithms", "True")
List of GPU-accelerated functions
  • All functions in the 
    view
    module
  • algo.createvisibilityinformation
  • algo.createVisibilityInformationFromViewPoints
  • scene.getHiddenPartOccurrences
  • algo.removeOccludedGeometries
  • algo.removeOccludedGeometriesAdvanced
  • algo.removeOccludedGeometriesFromPoints
  • algo.removeOccludedGeometriesFromViewPoints
  • algo.findOccludedPartOccurrences
  • algo.findOccludedPartOccurrencesAdvanced
  • algo.createVisibilityInformationAdvanced
  • algo.orientPolygonFacesAdvanced
Activating GPU acceleration in Docker image
Instructions to setup NVIDIA drivers and NVIDIA-Docker modules for Docker images.
Copyright © 2026 Unity Technologies
LegalPrivacy PolicyCookiesDocumentation Terms of UseDo Not Sell or Share My Personal InformationYour Privacy Choices (Cookie Settings)

"Unity", Unity logos, and other Unity trademarks are trademarks or registered trademarks of Unity Technologies or its affiliates in the U.S and elsewhere (more info here). Other names or brands are trademarks of their respective owners.

Some pages are machine-translated for convenience, and may contain inaccuracies. In the event of conflicting information, the English version is authoritative.

  • On this page

    • Install Vulkan

    • Hardware

      • Recommended

      • Minimum

      • GPU (optional)

        • List of GPU-accelerated functions

        • Activating GPU acceleration in Docker image

Report a problem with this page