Foveated Rendering

Foveated rendering is a rendering technique that uses an eye tracker integrated into a headset to reduce the image quality of the content being rendered in the user’s peripheral vision. The area that is not in the user’s focus gets blurred. The goal of the technique is to give users the best visual quality possible by using the least amount of computational resources.

All Varjo VR and XR headsets have foveated rendering capabilities. However, the feature only works if it is also supported by the application being used.

Why use foveated rendering?

Using foveated rendering is beneficial because lowering the image quality in the user’s peripheral vision improves system performance. This is because approximately 30-50% fewer pixels need to be rendered on Varjo headsets when using this technique (the exact amount depends on image quality settings and other factors).

Foveated rendering makes sense because a person can see with the highest resolution only at the center of the eye’s field of view and can detect much fewer details and colors in the periphery. This means that rendering things in full resolution in a person’s peripheral vision is essentially wasted because the eye cannot see the full detail.

With integrated eye tracking, a VR or XR headset can take advantage of this phenomenon and reduce the resolution in the user’s peripheral vision, so the rendering requires fewer resources. The user won’t be able to tell the difference in visual quality because the resolution change in the person’s peripheral vision is not observable, but the user will still benefit from the improved system performance.

Read more:

• Dynamic Foveated Variable Shader Rate with the OpenXR Toolkit and Varjo Aero (a.k.a. Foveated Rendering)
• Varjo API documentation