AMD Plans to Use RDNA 3.5 Integrated GPUs in Mainstream APUs Until 2029, Premium SoCs Transitioning to RDNA 5 Architecture

AMD Plans to Use RDNA 3.5 Integrated GPUs in Mainstream APUs Until 2029, Premium SoCs Transitioning to RDNA 5 Architecture

Future developments in AMD’s graphics architecture have been outlined, indicating that the RDNA 3.5 technology will persist across their APU product line until 2029. Premium System-on-Chip (SoC) variants, however, are set to introduce the next-generation RDNA 5 integrated graphics processing units (iGPUs).

AMD’s RDNA 3.5 and RDNA 5: A Comprehensive Overview

Since the debut of RDNA 3, AMD’s Accelerated Processing Units (APUs) have remained largely unaltered in terms of architectural innovation. Although RDNA 3.5 provided notable enhancements, it primarily served as an optimized iteration of the RDNA 3 framework. Meanwhile, AMD has unveiled its RDNA 4 architecture focused on discrete graphics, introducing advanced AI and Ray Tracing (RT) capabilities, as well as supporting the new FSR Redstone technology. These significant advancements could have been beneficial had they been integrated into AMD’s APU family.

The latest APU offering, the Ryzen AI 400 “Gorgon Point, ”continues to utilize RDNA 3.5, albeit with enhanced clock speeds. Additionally, the Strix Halo and Gorgon Halo “Ryzen AI MAX”series also adopt the RDNA 3.5 architecture, but they are available in larger configurations, featuring up to 40 compute units, compared to a maximum of 16 compute units in the mainstream lineup.

Insights from insider Kepler_L2 suggest that RDNA 3.5 will define AMD’s APU ecosystem well into 2029. This assertion aligns with recent disclosures from other industry insiders, such as Golden Pig Upgrade. Notably, categorization will occur where RDNA 3.5 integrated graphics will be targeted towards entry-level and mainstream APU designs. The Medusa Point “Ryzen AI 500″series is anticipated to be the final major family utilizing the RDNA 3.5 architecture before a transition to RDNA 5.

According to Kepler, the premium segment of AMD’s APUs will feature iGPUs built on the RDNA 5 architecture. This suggests that upcoming Medusa series devices, or high-end Halo offerings, will integrate these advanced graphics systems, promising enhanced performance and enriched feature sets.

AMD Ryzen APU Integrated Graphics Configurations

APU Family Codename GPU Codename GPU Compute Units (Max) TFLOPs
Ryzen 2000 Raven Ridge Vega 11 Compute Units 1.76 TFLOPs
Ryzen 3000 Picasso Vega 11 Compute Units 1.97 TFLOPs
Ryzen 4000 Renoir Vega+ 8 Compute Units 2.15 TFLOPs
Ryzen 5000 Cezanne Vega+ 8 Compute Units 2.04 TFLOPs
Ryzen 6000 Rembrandt RDNA 2 12 Compute Units 3.40 TFLOPs
Ryzen 7000 Phoenix Point RDNA 3 12 Compute Units 8.30 TFLOPs
Ryzen AI 300 Strix Point RDNA 3.5 16 Compute Units 5.93 TFLOPs
Ryzen AI MAX 300 Strix Halo RDNA 3.5 40 Compute Units 14.85 TFLOPs
Ryzen AI 400 Gorgon Point RDNA 3.5 16 Compute Units ~8.50 TFLOPs
Ryzen AI MAX 400 Gorgon Halo RDNA 3.5 40 Compute Units ~15.00 TFLOPs
Ryzen AI 500 Medusa Point RDNA 3.5+ TBD TBD
Ryzen AI 500 Medusa Premium RDNA 5 TBD TBD
Ryzen AI MAX 500 Medusa Halo RDNA 5 TBD TBD
Ryzen AI 600 TBD Point RDNA 5 TBD TBD
Ryzen AI MAX 600 TBD Halo RDNA 5 TBD TBD

With the arrival of the Medusa series, AMD is poised to enhance core counts and introduce specific optimizations that will enable their iGPUs to compete more effectively against Intel’s offerings. Intel’s recent upgrade to the Xe3 architecture is anticipated to significantly enhance performance, surpassing that of AMD’s current mainstream APU portfolio. Moreover, Intel is expected to unveil the Xe3P, a highly anticipated architecture built on the Celestial design, as part of a new family of Arc products.

A Wccftech forum post by user Kepler_L2 discusses the structuring of Zen 6 / RDNA 5 products, mentioning Olympic Ridge: IOD
Image Source: Anandtech Forums

Looking ahead, Intel has plans to advance beyond Xe4, exploring new horizons in integrated graphics with a strategic partnership with NVIDIA, which is expected to introduce x86 SoCs equipped with RTX-level graphics capabilities. Intel has made clear that this collaboration will not impede its existing graphics development paths, suggesting a dynamic shift in the integrated GPU landscape is on the horizon.

Source & Images

Leave a Reply

Your email address will not be published. Required fields are marked *