Nvidia decided to release the NVIDIA PhysX source code to the public


Last week NVIDIA announced via its blog the availability of the NVIDIA PhysX source code. so everyone can use its high-quality, reality physics simulation engine.

NVIDIA decided to release the PhysX physics simulation engine source code to the public and place it in the free projects category.

About Nvidia PhysX

NVIDIA PhysX is part of many popular game engines, including Unreal Engine, Unity3D, AnvilNext, Stingray, Dunia 2, and REDengine.

More than 500 games have been built on the basis of PhysX, including "Batman: Arkham Asylum", "Batman: Arkham City", "Bioshock Infinite", "Borderlands 2", "Lords of the Fallen", "Monster Hunter Online", "Daylight" "and" Witcher 3 ".

In such games, PhysX is used to implement effects such as destruction, explosions, realistic movements of characters and cars, smoke from waves, trees leaning in the wind, water flowing and flowing around obstacles, flapping and tearing of clothing, collisions and interactions with hard and soft bodies.

Some of the areas in which PhysX can be used as an open product:

  • Synthesis of data that reflect the state of objective reality, for research in the field of artificial intelligence and for training neural networks.
  • Creating realistic environments to train robots.
  • Simulation of real conditions in the process of running autonomous vehicles and autopilots.
  • Take realistic gaming environments to a new level.
  • The use of high-performance clustering systems to achieve a high level of detail and precision in the simulation of physical processes.

NVIDIA makes available to the public PhysX

The reason for the release of the PhysX source code is the urgent need to use physical process simulation in areas beyond computer games, the demand for projects related to artificial intelligence, robotics, artificial vision, unmanned vehicles and high performance computing.

In its current form, PhysX SDK is a multiplatform solution for the multiprocess simulation of physical processes in games, scalable for various computers, from smartphones to powerful workstations with multi-core CPUs and GPUs.

Active use of GPUs to speed up operations allows you to use PhysX to process effects in very large virtual worlds.

The announcement was made by Senior Director of Content Engineering and Technology, Rev Lebaredian, which is also responsible for Nvidia GameWorks technologies. In the post it says the following:

“We are doing this because the physical simulation ended up being more important than we imagined.

It is central to so many different things that we decided to make it available to the world in open source form.

The first version of the open source version is PhysX SDK 3.4 , but a new version of PhysX 20 is scheduled to be released on December 4.0, which will offer the first functional update for a free project.

The new version will implement the TGS (Temporal Gauss-Seidel Solver) algorithm, which will improve the quality of character and object simulation, which consists of many articulated parts.

In PhysX 4.0, assembly support using Cmake will also appear and the scalability of filter rules for kinematic and static objects will be increased.

This is very good news for devs who want to port their games for Linux, as now they will have at their disposal one more powerful tool.

And thus to be able to deliver more and more high quality products in Linux and gradually ending the differences that still exist in some titles with respect to Windows.

And, of course, opening the possibility for the Rev, where PhysX can be used in the autonomous vehicle, AI, and high-performance computing.

Where to find the source code?

The code for the engine and its associated SDK is open under the BSD license as well as GPU acceleration as well is under this license, so the compilation of it is currently supported for Windows, Linux, macOS, iOS and Android platforms.

PhysX also to have integration with Unreal 3 and 4 engines with Unity3D. The full announcement and more details can be found directly on the NVIDIA blog.

The source code is also available on GitHub.

The content of the article adheres to our principles of editorial ethics. To report an error click here!.

Be the first to comment

Leave a Comment

Your email address will not be published.



  1. Responsible for the data: Miguel Ángel Gatón
  2. Purpose of the data: Control SPAM, comment management.
  3. Legitimation: Your consent
  4. Communication of the data: The data will not be communicated to third parties except by legal obligation.
  5. Data storage: Database hosted by Occentus Networks (EU)
  6. Rights: At any time you can limit, recover and delete your information.