AMD’s RDNA 3 lineup is still very new, but the company is already looking to the future. In a recent interview, AMD executives spoke about the potential of RDNA 4. They also spoke about AMD’s approach to artificial intelligence (AI) and how it differs from Nvidia’s.
It seems AMD may be looking to increase its AI presence in the future. It’s a good decision – so far AMD has lost the AI battle to Nvidia. Here’s what we know about his plans for RDNA 4.
Japanese website 4Gamer spoke to AMD execs David Wang and Rick Bergman, mostly discussing the use of AI in AMD’s best graphics cards. While Nvidia has a longer history of AI, the interview suggests AMD may want its ROCm software suite to compete with Nvidia’s CUDA libraries. In fact, AMD claims that the two platforms are on par.
As for the future of AI for AMD, the company just seems to have a different approach to using AI in gaming GPUs. Where Nvidia has gone all-in, AMD has been slower to introduce AI to consumer cards – but it claims it was a matter of choice.
“The reason why Nvidia is actively trying to apply AI technology to applications that can run without using AI technology is that Nvidia has installed a large-scale inference accelerator in the GPU. In order to use it effectively, they seem to be working on a topic that needs to mobilize many inference accelerators. That’s their GPU strategy, which is great, but I don’t think we should have the same strategy,” Wang said in the 4Gamer interview, which was machine-translated from Japanese.
AMD realizes that gamers don’t need or want to pay for features they never use, which is why AMD believes inference accelerators should be used to improve games rather than anything else. To that end, AMD notes that it was able to achieve competitive performance with FidelityFX SuperResolution (FSR) without AI, while Nvidia’s DLSS uses AI.
Wang’s idea to implement something fun using AI is to improve the behavior and movement of enemies and non-player characters (NPCs) in modern games. When it comes to image processing, AMD seems to have bigger plans.
Jacob Roach / Digital Trends
“Even if AI is used for image processing, AI should be responsible for more advanced processing. In particular, a topic like “neural graphics”, which is currently gaining momentum in the 3D graphics industry, might be appropriate,” Wang added.
Wang spoke at length about the technologies used in AMD’s latest flagships, the RX 7900 XT and RX 7900 XTX. The Multi-Draw Indirect Accelerator (MDIA) used in RDNA 3 contribute to a remarkable performance increase of up to 2.3x over the previous generation. To that end, AMD wants to continue to build on this technology and add even more advanced shaders to RDNA 4. Wang said the company aims to make this the new standard specification for the GPU programming model.
We also got a brief, if not very specific, teaser of RDNA 4 from Rick Bergman, AMD’s executive vice president of the Computing and Graphics Business Group. Bergman said, “We promise to evolve to RDNA 4 with even higher performance in the near future.”
The use of “near future” is interesting here. AMD’s RDNA 3 lineup is still in its infancy. There are plenty of mobile GPUs coming to laptops, but the desktop range is still very small. It’s hard to imagine RDNA 4 coming out any sooner than late 2024 – there are so many current-gen maps yet to be released. However, when the time comes, it will be interesting to see what AMD is doing to improve its AI game.