AMD RDNA 3 Performance and Architecture Details Suggested for FAD 2022

At Financial Analyst Day Summer 2022, AMD confirmed new details regarding its upcoming RDNA 3 architecture that will feature in Radeon 7000 series GPUs later this year. Some of the information was reconfirmed from estimates presented in previous years, but it also confirmed some earlier leaks to be true. Thanks to this, we can better understand what to expect for gaming performance from the next generation of graphics cards.

One promising stat is that RDNA 3 will apparently deliver over a 50% increase in performance per watt over the existing RDNA 2 architecture featured in the current Radeon RX 6000 series. This was also the case during the transition from RDNA 1 to RDNA 2.

Additionally, AMD has announced that RDNA 3 GPUs will be manufactured using the more advanced TSMC 5nm manufacturing process, which will allow for more performance or efficiency, depending on AMD’s design preferences. This will mark a transition from the 7nm TSMC node that was widely used for RDNA 2.

One thing to keep in mind though is that AMD didn’t elaborate on how it measured efficiency increases in performance per watt, and didn’t provide actual performance estimates. He also didn’t elaborate on how much the Node advantage contributed to the stat compared to architectural changes alone. Either way, the minimum 50% improvement in performance per watt is something to be excited about.

(Image credit: AMD).

The chiplet approach

The confirmation of the chip packaging technique is the other major element of this announcement. We’ve been expecting this for a while due to several leaks that pointed to this. The first mention of using chiplets for RDNA 3 came in the spring of 2021 from a serial industry leaker. The second came in February earlier this year from an AMD engineer, who shared a bit too much information about the projects they were involved in.

In short, this announcement confirms that AMD has developed the manufacturing techniques and technologies necessary to implement multiple GPUs on a single graphics card, much like it did with Ryzen Zen processors. This design could allow AMD to effectively increase performance to new heights over what would be possible with a traditional single-GPU design.

You can think of this multi-GPU (MCM) design like using SLI or Crossfire, but without the compatibility issues and poor performance that come with these older technologies, provided RDNA 3 works as intended. The difference should be that the OS won’t know that there are multiple GPUs, as the data computation is handled at the architectural level. AMD will also be the first company to implement such a design for a game architecture.

Amd Rdna4 Radeon Performance Chart Release Date

(Image credit: AMD).

Massive gaming performance probably boosts sound

The idea that AMD has once again managed to improve efficiency by 50% from generation to generation is a big deal. The implementation of chiplets is also particularly interesting to improve performance. Given the power of today’s best graphics cards, mainstream gamers should be able to achieve new levels of performance such as 4K gaming at over 144 fps at ultra settings in today’s most demanding games. today. 1440p 240Hz should also be a fairly easy affair. We may even be able to start enjoying advanced visual effects like ray tracing without killing our frame rates.

These levels of performance are already achievable with some high-end graphics cards. Examples include the $2000+ USD 450W+ Nvidia RTX 3090 Ti and the $1100+ USD 350W+ AMD RX 6950 XT. However, these expensive halo products mean little to the masses. Next-generation graphics cards are expected to bring these levels of performance at more affordable prices and with lower power consumption.

Comments are closed.