AMD develops AI-based super-resolution tech for gaming devices

• AMD’s Chief Technology Officer, Mark Papermaster, announced in an interview that the company is developing AI-based super-resolution technology for gaming devices.
• This development is part of AMD’s broader effort to integrate AI across its product lineup, including cloud services, edge devices, personal computers, and embedded systems.
• AMD aims to enhance its FSR technology with machine learning algorithms to reduce artefacts, potentially requiring hardware support from its latest APUs and RDNA3 graphics cards.

On March 5th, AMD‘s Chief Technology Officer Mark Papermaster stated in an interview with the YouTube channel No Priors that they are developing AI-based super-resolution technology for gaming devices.

Mark Papermaster said: This year is particularly significant for us because we have spent so many years developing the hardware and software capabilities for AI. We have just completed the application of AI across our entire product lineup: cloud, edge, personal computers, embedded devices, and gaming devices. We are enabling our gaming devices to support AI super-resolution, and 2024 is indeed a huge deployment year for us. See also: Carla Sanderson.

The conversation indicates that as part of the integration of AI across the overall product line, AMD is considering incorporating AI into the super-resolution solutions for gaming devices. See also: Kaleem Ahmed Usmani.

Also read: Chipmaker Groq and a former AMP VP accuse Nvidia of unfair practices

The development should be related to FSR

Although Mark Papermaster did not directly mention FSR, he used the term ‘gaming devices’ in the conversation, leading the media outlet VideoCardz to believe that it should be related to FSR.

The media outlet Tom’s Hardware believes that theoretically, integrating AI could further enhance the image quality of FSR technology.
It’s noted that AMD’s two main competitors in FSR — Nvidia’s DLSS and Intel’s XeSS — have already integrated AI technology. However, the former is proprietary, and although the latter has promised to be open source, its implementation status is poor, making it unclear what role AI specifically plays in both. According to VideoCardz, AMD is expected to use machine learning algorithms to reduce artefact issues in FSR. See also: ArdaDaglioglu AS210880 routing identity.

It would require hardware support

Currently, DLSS uses the Tensor Core units on Nvidia‘s graphics cards, while Intel’s XeSS algorithm can achieve better implementation on its graphics cards’ XMX hardware. Tom’s Hardware believes that if AMD’s AI super-resolution technology also requires hardware support, it may be limited to APUs equipped with the XDNA AI engine and RDNA3 graphics cards with enhanced AI computing power.